Office of Inspector General
          Report of Review
PRCfl*           *
     Special Review of EPA's Information
               Systems Program
                   Volume II


              El SKG3-15-0098-4400038


                  March 24, 1994
                                        Racycted/Recyctabl*
                                        Printed on paper that contains
                                        at least 50% recycled fiber

-------
Inspector General Division
  Conducting the Audit:
Region Covered:

Program Offices Involved:
Technical Assistance Division
Washington,  D.C.

Agencywide

Agencywide

-------
          Special Review of EPA/s Information systems Program
                      TABLE OF CONTENTS
                          VOLUME II
APPENDIXES
     APPENDIX  I:   SENATE SUBCOMMITTEE ON SUPERFUND,
                    RECYCLING, AND SOLID WASTE MANAGEMENT
                    REQUEST TO INSPECTOR GENERAL
                    FOR A SPECIAL IRM REVIEW .......... , . . 1-1
                                                    r .
     APPENDIX II:   SENATE SUBCOMMITTEE ON SUPERFUND,
                    RECYCLING, AND SOLID WASTE MANAGEMENT
                    REQUEST TO EPA ADMINISTRATOR
                    FOR A SPECIAL IRM REVIEW ............. II-l
                                    »

     APPENDIX III:  JOINT AGENCY AND OFFICE OF
                    INSPECTOR GENERAL TEAM MEMBERS ...... III-l

     APPENDIX IV:   FOCUS GROUP SUMMARIES ............... IV-1

     APPENDIX V:    INTERVIEW SUMMARIES .................  V-l

     APPENDIX VI:   AGENCY PARTICIPANTS IN FOCUS
                    GROUPS AND INTERVIEWS ............... VJ-1

     APPENDIX VII:  OIG REPORTS,  GAO REPORTS,  EPA
                    MANAGEMENT REPORTS, AND
                    CONGRESSIONAL TESTIMONY. . . ..* ........ VII-1

     APPENDIX VIII: APPROACH AND METHODOLOGY ........... VIII-1

     APPENDIX IX:   IMPLEMENTATION STATUS OF PRIOR
                    HIGH-PRIORITY RECOMMENDATIONS ...... '. IX-1

     APPENDIX X:    CHART:  EPA OFFICES WITH MAJOR
                    IRM RESPONSIBILITIES ................  X-l

     APPENDIX XI:   OIRM ISSUE PAPER:  STRENGTHENING
                    IRM AT EPA. ...................... ;.'.  XI-1

     APPENDIX XII:  GLOSSARY OF ACRONYMS AND
                    ABBREVIATIONS ................ . . ..... XII-1

     APPENDIX XIII: REPORT DISTRIBUTION ....... . ........ XIII-1
                           Report NO.   E1SKG3-15-0098-4400038

-------
Special Review of EPA's Infornatioa Systems Program
   (This page intentionally left blank)
                 Report No.   E1SKQ3-15-0098-4400038

-------
                                                     ,«««.,^^. -r
                                                     APPENDIX I
                        September 27,  1913

Honorable John C. Martin
Inspector General
United States Environmental Protection Agency
            DC 20460
Dear Hr. Martini                                             ,-

     Thank  you for presenting  euch compelling t«»timony at our
bearing on  Jtane 10, 1993. unfortunately, your efforts have railed
eeriouB new queitioni about the Agency1 • management of the fiscal
and  information  lyitemi  programs,  compounding  the  previously
identified  problem* with EPA's contract management.

     Based  on  information you presented at our recent hearing and
th« body of accumulated evidence on this issue, we **• requesting
that your office perform a comprehensive management review of the
EPA's  fiscal  and  information systems programs  with as  much as
specific attention on the Superfund area as is possible.

  •   We recognise  that EPA has taXen some important steps in the
past year to address its contract management problems. •  However/
there has been no  comparable no holds -barred, systemic management
review  of  the fiical  and information  systems  programs  of th«
Agency.

     The  apparent lack of  credible basic management information
about  where Agency  funds,  including Superfund money,  are being
sprat and how, and the questions raised about what the resulting
accomplishments  may -be, go  to the heart of  all  Agency  programs
including the  Superfund program.

     Xt is  critical that these issues, to the extent they pertain
to the  Superfund program,  are addressed as part of the Superfund
reauthorisation process over the next year.  Therefore, we) rvqueet
that you  carry out the global management  reviews and report the
results to  us  as soon as possible but no later than January 1994.
We would ask that  any legislative recommendations are provided to
the Subcommittee no later than November 1,  1993.  The subcommittee
understands that to accoapliah your  work  in  this timeframa, you
have  asked  and  Agency has  agreed to be  a participant  in your
reviews.
                                1-1

-------
                                                   APPENDIX I
     We  truly  appreciate  all  your efforts  in identifying and
remedying the serious management problems at the Agency, and look
forward to receiving copies  of  the two reports requested by this
letter.
                                          Sincerely,
Frank Lau
Chairman/.Subcommittee on Super fund,     Sinking Minority Member
Recycling, and Solid Waste Management     j
                               1-2

-------
         MAX UUMl MNUNk OUIftlMI

                       "—•—-"                              APPENDIX II
        .
NMMWM.MWIMMKOIM     MR
Bititd States Senate
          mm. tMIH tfrgttgetm               CeMUnTttONniVWONMINTANpnAUCWDtKt

                                          WMrtlMflTOft. PC «
                             September 27, 1993

     Ma.  Carol Browner
     Administrator
     U.S. Environmental Protection Agency
     401  M Street
     Washington,  DC 20540

     Dear Xa.  Browner:

           Aa you ara aware, the Subcommittee haa had  serious  eoneerna
     about the Agency'• management of fiaeal and information systems, we
     have requested that the  Office of the  Inspector General  (OZ6)
      (letter attached)  perform a  comprehensive  management review of
     SPA'a  fiaeal  and  information systems  with  aa  much  apecific
     attention on the  Superfund area aa  ia possible,

           Given the pace of Superfund reauthorisation,  it is  critical
     that these issues are addressed over the next few months,   we have
     consequently asked  that  the  Inspector General  (10) provide  any
     suggestions  for   legislative  reform by  November 1,  1999,  and
     complete  his overall review of  fiscal and information management
     systems no later then January 1994.

           In order for this to  occur/ it ia important that the staff of
     the  ozo receive full cooperation and participation from fiaeal and
     information system aa well aa program officials  during the course
     of  the studies.   This cooperation should  include assistance in
     identifying problems and their cauaea aa  well solutions either of
     a management or legislative nature.

           The  OZG  will periodically brief  staff of  the  Subcommittee
     during the conduct of this work.  We appreciate  your staff* full
      iavolvwewfc in assisting the IG in  reviewing these critical Agency
      support systems.
                                                Sincerely
      Prank Ladtenberg       ^»«X             Aavra DujSKoerger
      Chairman, Subcommittee on Superfund,       funking Minority Member
      Recycling, and Solid Waste Management
      United States Senate
                                       II-l •

-------
                                                  APPENDIX  III
          JOINT AGENCY AND OFFICE OF INSPECTOR GENERAL
                       REVIEW TEAM MEMBERS
Project Director

Gordon Milbourn III, Special Assistant to the
     Assistant Inspector General for Audit
OIG

Proi ect Manager

Craig Silverthorne, Chief
ADP Audits and Support Branch
OIG

Team Members

Tom Bennett, Auditor
ADP Audits and Support Branch
OIG

Eileen Falcone, Auditor
ADP Audits and Support Branch
OIG

Asa (Jack) Frost, Director
Information Management Staff
C3WER

William Gill, Computer Specialist
Office of Information Resources Management
OARM

Steve Hufford, Chief
Information Management Branch
Office of Information Resources Management
OARM
                              III-l

-------
                                                 • APPENDIX IV

            SPECIAL IRM REVIEW -  CONGRESSIONAL REQUEST
                      FOCUS GROUP SUMMARIES

     The following transcripts of the four focus groups are
provided to give the reader the complete results of the focus
group discussions.
                               IV-l

-------
                                                  APPENDIX IV

                    FIRST FOCUS GROUP SUMMARY

     The session started with the group voting, using a simple
ballot, to prioritize the previously identified problems and root
causes addressed from previous reports, and to add descriptions
of any other major IRM problems not on the list.  The ballot was
structured so that respondents would answer based on their own
systems experiences.  Respondents were also asked to indicate on
the ballot how they used the particular information system on
which they were basing their responses.  The group identified
four high-priority problems and identified root causes and
solutions for each of them.

     The Focus Group voted the following previously identified
problems as the three manor problems:

1. Systems don't provide credible information on the resulting
accomplishments from money spent.  (19 points)

2. Systems don't adequately address cross-media pollution
problems. (19 points)

3. Duplicate systems have been developed.  (19 points)

     One new problem surfaced that was not on the list of
previously identified problems.

4. Systems have low levels of utility and friendliness to
regional users.
                               IV-2

-------
Other  Problems Identified  (from initial  ballot)

        RCRIS

        System is  overly complicated &  difficult to use.

        System continually needs  upgrading  6  rewrites.

        CERCLIS

        CERCLIS t  IFMS/FMS are  not  electronically linked.

        We track too many things  and definitions are too
        convoluted.

        Purpose of system not clearly identified 6 kept
        "narrow" for national systems.

        Systems developed with  no clear customers.

        Usefulness of  systems for EPA regional managers.

        FRPS

        Doesn't allow  easy user access.

        Doesn't contain  parametric  data, only violations.
                                                              APPENDIX IV
                                                           (med)

                                                           (med)



                                                            (high)

                                                            (med)


                                                            (high)


                                                            (med)

                                                            (high)



                                                            (high)

                                                           . (med)
         AIRS/AFS

         The different data bases don't communicate/ share    (high)
         info well - this will have a great impact on multi-
         media issues if not addressed.

         Over all direction & policy is lost in the day-     (high)
         to-day operations — need to take time to evaluate
         and refocus if it's required.

         Need to have standardized & consistent minimum      (high)
         data elements that are agreed to by alj. programs
         using the system.

         Need one identifying number to track a facility     (high)
         on (especially multi-media sources)  — too many
         different numbers used now!

         Stop collecting data for data sake - identify       (med)
         the need for & use of the data - use data
         for program management in decision making.
11/17/93
         IFMS/EPAYS/GICS/PPAS

         Not user friendly.
         Don't provide  info  useful to  regional /
         first-line manager.
                                    •».
         (EPAYS/IFMS)i  Systems  are too difficult for a
         manager to use.  Need easy to  use data systems
         that provide managers  info quickly.
(high)


(high)
                                   IV-3

-------
                                                  APPENDIX  IV

                             SUMMARY

CROSS-CUTTING ROOT CAUSES:  The following root causes were
identified as having impact on more than one priority problem.
The causes are identified to the problems by number in
parentheses.

1.   Program and data  (IRM) staff don't work together to design
     useful systems, therefore no motivation exists to assure
     data quality.  (1)  (3)

2.   National systems are designed to meet EPA Headquarters needs
     (and those of Congress), and do not meet regional needs.
     (1) (2)

3.   National systems are media specific.  (1) (3)

4.   Program and data management and staff often lose sight of
     the philosophy/reason behind the original development of a
     system,  i.e. the intended purpose of the data.   (2)  (4)

5.   Automation is not always the answer.  Systems are not the
     best place for answering all questions.   If activities
     resulting in accomplishments are qualitative as opposed to
     quantitative, these would be unmeasurable in quantitative
     systems. (1) (2)   (3)  (4)

6.   There is no comprehensive master plan [for IRM] and there
     are unrealistic expectations. (1) (2) (3) (4)

                             SUMMARY

GROUP DISCUSSION OF POTENTIAL SOLUTIONS;   The group considered
the following solutions before in depth discussions to relate
solutions to root causes for the priority problems:

1.   Developing a master plan for EPA information systems.

2.   Setting clear expectations for systems rather than adding
     functionality to existing systems ad infinitum.

3.   Better planning for, and culling, the set of Agency
     information systems.

4.   Prioritizing customers and their competing needs.

5.   Viewing data as timeless.
                               IV-4

-------
                                                  APPENDIX IV

6.   Addressing the mismatch between data system assumptions  and
     congressional expectations  (e.g., a "best practicable
     treatment" approach in legislation will never create
     information systems that can answer specific questions about
     ambient conditions at individual .sites).

7.   Seizing the opportunity, after NAFTA, Mexico/SEDESOL  has to
     build good systems from the start.

8.   Addressing problems with 25 FINDS numbers being assigned to
     the same facility under the air program.

9.   Recognizing automation is not the best answer to all
     problems.

10.  Avoiding an increase in internal regulations.

11.  Creating core data elements across systems so States
     wouldn't have to enter duplicate data.

     This discussion led to developing one particular solution:
(stakeholders forming a core committee to formulate an effective
IPM 5 year plan) and brainstorming to flesh out' that solution.
The group then identified and categorized solutions to the root
causes.

                      ROOT CAUSES/SOLUTIONS
       Based on Previously Identified Problems/Root Causes
            for  IRM Problems Supplied to Participants)

PRIORITY PROBLEM #1:  Systems don't provide credible information
on the resulting accomplishments from money spent.  (19 points)

ROOT CAUSES FOR PRIORITY PROBLEM #1:

1.   EPA establishes surrogate measures (activity measures and
     "bean-counting"  measures rather than trends/outcomes
     measures)  to quantify program success (e.g.,  compliance
     rates, permits issued,  inspections,  penalties collected,
     "best technology applied",  etc.).   This, in turn,  is
     because:

2.   Data quality is less accurate, for "environmental"  data and
     is subject to many qualifiers (season,  age,  sex,  species),
     which must be interpreted

3.   Environmental data is expensive to collect and limited in
     amount, geography,  year,  etc.
                               IV-5

-------
                                                  APPENDIX IV

4.   Federal/state/local stakeholders who would incur the
     transaction costs to collect more environmental data do not
     think it's worth it.

5.   Reports like Reilly's "Risk Based Priorities" are a better
     tool for relating program expenditures to environmental .
     results (systems aren't the best place to accomplish this
     task).

PRIORITY PROBLEM #2: Systems don't adequately address cross-media
pollution problems.  (19 points)

1.   "Media" based organizational structure of the Agency does
     not encourage system designs which address risk-based
     environmental problems.  Furthermore, systems tend to lack
     common data elements (e.g., identifier numbers) which would
     facilitate utility across organizational lines.

2.   Separate media programs don't encourage cross media
     communication.

3.   Systems developed are local in purpose and focus on single
     media concerns.

5.   Needs from media to media are vastly different.

6.   Systems are statutorily based, not risk or problem based.

7.   Too many different identifier numbers exist, not one
     number that is common.

8.   Systems do not communicate to share common data elements.

PRIORITY PROBLEM #3.  Duplicate systems have been developed.   (19
Points)

1.   Information must serve multiple clients with multiple needs.

     --National systems are developed by Headquarters offices to
     serve, their needs without considering the needs of other
     Headquarters offices, Regions, and States
     --Headquarters and Regional managers have different data
       needs
     --State programs require more and different information and
       detail
     --Too little interaction exists by Regions at National level

2.   National systems are too complex.
                                 «.
     --Systems are overly complicated and too difficult for

                               IV-6

-------
                                                  APPENDIX  IV

       managers or staff to use
     --Systems cannot produce needed regional reports, therefore
       Regions develop report writers
     --System development or revision is difficult and takes  too
       long to obtain national consensus/ therefore people
       develop their own
     --Developers with no environmental program knowledge develop
       systems that are not useful to the Programs they should
       serve, therefore people develop their own

3.   There is no comprehensive Agency or national program office
     planning for system development or identifying future needs.

     --There is no master plan or life cycle management of
       systems
     --There is no agreement on standard Agency-level data fields
     --Headquarters uses end of year money for unnecessary new
       system enhancement
     --Program is unwilling to resolve mainframe versus PC based
           debate, so they decide to do both

4.   Existing systems are inflexible and cannot be easily adapted
     to meet changing needs.

     --canned reports only provide information valuable to
       Headquarters or Congress
     --new system developed contains almost the same information .
     --data fields are not always normalized between systems
       requiring some overlap or duplication, eg. the FRDS system
       doesn't address all needs of users (states) because the
       system carries only violation indicators not parametric
       values
     --developers fail to step back and refocus to correct the
       direction; instead a new system is developed

PRIORITY PROBLEM #4.  Systems have low levels of utility and
friendliness to regional users.

ROOT CAUSES FOR PRIORITY PROBLEM #4.

1.   EPA's national systems are written in obsolete database mgmt
     software.  This obsolete software (Focus,  ADABAS, S2K,
     Clipper, etc.)  makes it impossible or too expensive to
     create user-friendly systems.  No amount of user
     identification and user participation in the development
     process can change this.

2.   There is a need to better identify "who" the system must be
     friendly/useful for.
                               IV-7

-------
                                                  APPENDIX IV

3.    The "user" group has changed over time to include everyone
     with a PC on their desk.

4.    Costs are high to make systems user-friendly.

5.    Program staff and data staff have not worked together on
     what's needed and what's to be done with what's been
     collected.

6.    We lose sight of the philosophy and reason behind the
     original development of the system, and along with it, the
     intention of how the data are to be used.

7.    The systems' become cluttered with add-on activities.

8.    National requirements (for Congress) are not the same as
     regional requirements

9.    We rely too much on contractors and work with artificial
     deadlines.

SOLUTIONS:  The group believed the first two of the following
solutions would have a large impact in addressing EPA's IRM
problems discussed during the two days.  For those two solutions,
the group did a force-field analysis (looking at pros and cons)
of implementing those solutions:

1.    Formulate an IRM 5-year plan by having stakeholders
     (Administrator, Assistant Administrators, regional media
     program managers and staff, state and local representatives)
     form a core committee to:

     a)   conduct viability study of existing systems (to keep,
          revise, or pull them).

     b)   identify the future direction of data use and data
          management.

     c)   make hard decisions, prioritizing competing customer
          needs  (not just attempting to please all customers).

     d)   examine whether it would be cost-effective to pursue a
          core data concept for the Agency, i.e., relational data
          concept.

     e)   examine and answer the following questions  [for
          national systems]:

          i)   what is the decisign that can/will be made from
               this data/system?

                               IV-8

-------
                                                            APPENDIX  IV


            ii)   what level of  data  is necessary to  provide basis
                  for the  decision?
            iii)  is  it necessary for this  data  to reside  in a
                  national database system?
            iv)   are the  data "timeless"?
            v)    are the  customers transient and can we say "no" to
                  their needs?


                      FORCE-FIELD ANALYSIS FOR SOLUTION 1
                                    PRO
            there  is a legislative mandate to do so
            it aligns  with a multimedia  approach
            it is  an opposing force  against individuals  who try to make
              the  data systems work  rather than question the usefulness
              of the system
      -     it would simplify identification of facilities  and sources
            it would reduce the data reporting burden on businesses, states,
              and  locals
            it provides for a broad-based  approach to determine expectations
              and  goals of what the  system is to do
            Administrator would need to  designate a "team"  with a very
              succinct mission and tight schedule
            if the cost of maintaining current systems is factored in with the
            cost of duplicate data entry,  the up front costs would be worth
            the long term benefit

                                    CON
            bureaucratic intransigence to  change
            senior managers don't have attention span to provide meaningful
              involvement and direction
            attempting to reach consensus  on core elements  may make it
              difficult to define what is  really "core"
            AAs may be territorial or resistant
            little historical interest by  EPA top managers  to  information
              management issues

2.    Enhance  electronic  transferability  of core  national  data,  so
      that  regions  can develop  their own  modular  front end data
      systems  (interfaces to the  data)  to meet  regional data
      needs.


                      FORCE-FIELD ANALYSIS FOR  SOLUTION 2

                                     PRO
            if the cost of maintaining current systems is factored in with
              cost of  duplicate entry at various levels,  the up-front costs
              Of this  new "core system" may be worth the long-term benefit.
            this would eliminate many disputes between HQ and  regions and
              among regions selecting "important"  data fields
            front-end  programming expertise is developed at regions,  not by
              transient HQ contractors
            the necessary technology is now available and affordable
            this would reduce time and cost in developing new national
              systems
            regional users would get  what  they want every time
            NPR culture would empower lower levels (regions) to take the lead
              in meeting their own needs (there may be 10 different front-end
              systems)                  ~~


                                    IV-9

-------
                                                      APPENDIX IV

           this would make the use of data easier
           there would be regional mgmt  consensus behind this idea
           supports multi-media enforcement
           speeds up development of enforcement cases and comparison of
            statistics and trends

                                  CON
           entrenched NCC bureaucracy doesn't want to relinquish turf to
            regions
           some regions have no front-end developing capability, and
            training costs are high to  acquire expertise
           need for HQ "uniformity" make it hard to let go of front end
            systems
           current national system managers will be resistant to change
           HQ offices would have to "give up" some control.  The territorial
            instinct in HQ is strong
           some individuals (programmers, data system managers) will continue
           to try to make the existing systems work rather than question
           the systems

                           OTHER SOLUTIONS

3.   Implementation of HR 3425, which will establish a Chief Info
     Officer  and a Steering Committee at  a high  level

4.   Look to  the National Performance Review for recommendations
     for  how  EPA can be better organized  to address environmental
     risks,  i.e. address  barriers we  have within media

5.   EPA  should pay for systems management costs incurred by
     States,  to improve data quality  (but decision to do this
     should be done on a  system-by-system basis  to ensure cost-
     effectiveness)

6.   Design specifications committees/workgroups must include
     regional representation at the workgroup leadership level.
     They must also include state/local reps if  the systems are
     to be used by them,  or if they will  provide data to .the
     systems.

7.   HQ traditionally has had the lead on developing systems.
     Shift the lead and contractor control to the region(s)
     chosen by national consensus to  be the lead.

8.   Use  a bottom-up approach rather  than a top  down development
     approach for selected systems.

9.   In the absence of a  large committee/global  approach,
     establish clear expectations for each system by asking:
     a)    what is the decision that can/will be  made from this
           data/system?
     b)    what level of data is necessary to provide basis for
           the decision?
                                    «.

                                IV-10

-------
                                             APPENDIX IV

c)   is it necessary for this data to reside in a national
     database system?
d)   are the data "timeless"?
e)   are the customers transient and can we say "no" to
     their needs?
                         IV-ll

-------
                                                  APPENDIX  IV

                    SECOND FOCUS GROUP SUMMARY

                        PRIORITY PROBLEMS

1)   Exclusion of information management from regulation and
     guidance development. Lack of ownership of information in
     EPA systems by program people (What information do we need
     to run our programs).  (Success is defined by getting a
     regulation "out the door").
                               and
     Lack of understanding or participation by upper management
     (Office Director and above) in information resources
     management.  Information management programs are not treated
     as core function or critical to EPA environmental
     protection.

          (23 Pts.)

2)   Lack of consistent Agency architecture (hardware, -software,
     data),  strategy and lack of power to enforce it.  No
     consolidated approach to EDI.  [Indirectly trying to enforce
     it through budget].    (13 Pts.)

3)   Lack of centralized data administration function in Agency
     (also none within program offices).   (11 Pts.)

4)   Un-implementable policies, standards,  and guidance.  (They
     are basically "OK",  but there is not power to enforce and
     often can not afford it).   (10 Pts.)

5)   Too much dependency on contractors and inadequate in-house
     IRM technical expertise.    (9 Pts.)

6)   Lack of defined IRM infrastructure,  communication within
     infrastructure, and common understanding of roles and
     responsibilities  (how infrastructure works).   (Read about
     Lotus Notes as an Agency standard in "Government Computer
     News")

     Unwarranted complexity of getting work assignments through
     existing contracts  (takes too long to get something through
     contracts).  (6 Pts. each)

7)   EPA does not view information as a valuable tool to empower
     the "public" (local environmental groups, researchers,
     labor,  industry) to deal with environmental problems beyond
     what the government can do.   (3 Pts.)

8)   Budget (realities) process precludes long-range, strategic
     planning.

                              IV-12

-------
                                                  APPENDIX IV

     Lack of effective use of "life-cycle planning process" and
     understanding of this by management (needs to be in more
     understandable language).

     State-of-the-art computer equipment out of reach
     (availability on Agency contracts lags and budgetary
     constraints).

     Data integrity problems with EPA's mission critical systems.

     GROUP COMMENTSt     Data usually developed for principle
     users, not secondary users.  Source of data is from
     States/often voluntary.  However, EPA is accountable.  EPA
     dependent on others (trust) for data (because of statutes).
     Often there is a Federal/State difference in interpretation
     because delegated programs are not exact or precise.
     Because of priorities and limited resources we are working
     in a continuum.  Upper management tends to "hang on" to
     numbers (which are not real-time and constantly changing).
     States want regulatory flexibility making it difficult to
     aggregate data (esp. on a delegated responsibility).  This
     allows inconsistencies.  There is also an interrelationships
   .  of problems.   Public access may be mandated but we do not
     understand costs (how much and what is the
     effectiveness/benefits) and have not determined if it is
     even appropriate in many cases.  Secondary user often does
     not want to take the time to understand the data.  Example
     Chesapeake Bay program's "Chessie System" -"there are
     negligible number of users but a large amount of money was
     spent which may have been better spent on cleaning the
     environment.

           (2 Pts.  each)

9)    No defined career paths for information management
     specialists similar to that developed for scientists in EPA.

     Development of duplicate systems.

     GROUP COMMENTS;     This is a problem.   OW example is a
     Congressional add-on to report on contaminant sediments
     caused creation of an emergency data base because of a
     mandated date.  A system was in development which could have
     provided the data but would have taken longer to 'complete.
     How many systems store sample data?  Need to put price tag
     ($ threshold)  on what is major system.   We never "kill" old
     systems.  Often contractors get to know the systems, have
     access to them and market them to other offices,
     perpetuating duplication.   RACF has helped RCRIS by
     preventing contractor access.  Probably a problem, esp. with

                              IV-13

-------
                                                  APPENDIX  IV

     tracking systems and with other Federal agencies.  OW
     example - USGS, NOAA (NCPDI gets data from PCS).  Even
     though they may cooperate, nobody gives up systems because
     they may lose funding.  Seven states have their own NPDES
     system because they say their systems are integrated.  In-
     house systems - Management wants up-to-date information
     (real-time) and begins to create own systems.  However, they
     often use information that is not ready for release but use
     it anyway.  Management often will not let us (give ua  time)
     do it right but there is always time to do it again.   Often
     there are spin-off systems (LANs) because people do not like
     the platform (mainframes) .,  There is no standard
     architecture and programs allowed to do it.  No one has
     power to say no.

          (1 Pt. each)

No Votes

10)  Lack of credible basic management information about where
     Agency funds, including Superfund, are being spent and how,
     and the resulting accomplishments (previously identified
     problem).

     GROUP CQMMBMTSi     Can not account for adequacy of dollars
     spent (i.e., how do you know you are spending dollars  on the
     right thing.  Need to measure results on environment and how.
     systems support that.   Not an issue in some .programs
     (OSW/OPPT).  Systems managers and SIRMOs do not play heavily
     in budget process.  Congress often asks questions from
     different people and do not get the same answers because of
     different perspectives of respondents (Congressional staffs
     go to various sources and often protocol for QA responses is
     not followed in agency).   A problem of communication,  lack
     of screening responses and is a vulnerability leading  to
     problems in credibility.   Everyone views "accomplishments"
     differently  (different customers/users).  Congress may ask
     "Are waters cleaner?",  but systems were not designed to
     answer that.

11)  Difficulties in identifying cross media pollution problems
     (previously identified problem).

     GROUP COMMENTS:     This is a program definition/data
     definition problem,  Oftep. asking questions of systems that
     they were not designed to answer.  No one (upper management)
     can say no!  Maybe need to look at Congress as a customer
     and ask what it is they need.  No strong leadership at the
     top (esp. in IRM).  Never received questions regarding
     integration.  Maybe this is an access problem.  Often

                              IV-14

-------
                                                  APPENDIX IV  .

     Congress does not understand complexity of the environment.
     Statutes and regulations often conflict and systems are
     expected to perform this function.  Agency is
     compartmentalized.  Because subject areas (media) are so
     different would most users understand the data?  Maybe we
     are selling users short on abilities to understand data.

12)  Significant cost overruns and delays in developing
     information systems (previously identified problem).

     GROUP COMMENTS;     The group consensus was that this was a
     result of other problems not a problem in and of itself.
     Budgets and statutes continually change.  However, it also
     happens in- systems development regardless (has there been a
     system that hasn't?).   Inadequate staff and dependence on
     contractors are problems.  Sometimes Congress never provided
     enough money to begin with.  Regulations and guidance are
     written with no thought of the impact on information
     systems.  You have the same problems (i.e. result - cost
     overruns/delaysJ in implementing programs.  There is also
     game playing.  Sometimes systems are not a line 'item in a
     budget.  It would be better if they were.

13)  Exposure of systems to unnecessary risks  (access and other)
     (rewording of previously identified problem - "Exposure of
     Agency's financial payment systems to unnecessary access
     risks").

     GROUP COMMENTS;     This is a problem for systems fas
     reworded;.  Exposure of systems to unnecessary risks.

                       SECOND FOCUS GROUP
                    ROOT CAUSES AND SOLUTIONS

PROBLEM;  Lack of understanding or participation by upper
          management (Office Director and above)  in information
          resources management.  Information management programs
          are not treated as core function or critical to EPA
          environmental protection.  Exclusion of information
          management from regulation and guidance development.
          Lack of ownership of information in EPA systems by
          program people (What information do we need to run our
          programs).   (Success is defined by getting a regulation
          "out the door").

ROOT CAUSES;

     *    Senior management does not understand linkage between
          information systems and accomplishing the mission.


                              IV-15

-------
                                        APPENDIX IV

Senior management does not understand  role  of  data in
framing options  in the decision-making process (Data
loses its  identity).

Upper management (Office Directors and above and Branch
Chiefs in  Regions) are accountable for things  other
than IRM,  i.e.,  program mission, and do not pay
attention  to IRM unless they can see a return.

Designations of  IRM functions not reflected in
performance standards and functions are not empowered
 (responsibility  without authority).   (Function
examples:  System managers, information management
coordinator, PC  site coordinator  (PCSC), LAN
Administrator/LAN Manager, SIRMO, RACF Administrator,
Account Manager, Records Managers, Telecommunications
Contacts,  ADP Coordinators, ADP Training Coordinator,
EMail Coordinator, Contracts Management, etc.)

Too much turnover at the highest management levels
which does not support long-term investment in IRM -
working for short term results to improve resumes for
next j ob.

Lack of insulation from "political winds".

Management does  not understand IRM (in terms of  costs,
logistics, and resources).

Upper management "grew up" in an era that did  not use
information systems for decision-making and is not
comfortable linking benefits of information systems  to
decision-making.

EPA does not recognize long range investment/benefit  of
information - too many new initiatives (initiative of
the month  syndrome), play budget games, cut base
budgets.

Inadequate communication of IRM issues among upper
'management.  (Example: RACF and its effects of
implementation).  Issues go to lowest  level.   Selection
process of contactees on IRM communications is faulty.

Organizational location of IRM function(s)  misplaced in
the Agency and program offices.   (Sometimes functions
within Division,  and can only service that division).

NDPD physically  remote/lack of communication and
understanding of programmatic. IRM needs.

                    IV-16

-------
                                                  APPENDIX IV

          COMMENTS; One participant indicated remote location had
          nothing to do with poor communication or understanding,
          however, another participant indicated lack of travel
          money impacted the ability of NDPD to be present at
          meetings in which they should be represented.

     *    No long-term strategy for tying IRM to Agency mission.
          Management by Committee - Too many committees, lack of
          coordination between committees, no consideration of
          impacts on information management and the other
          committees' work and the effects they are having.
          Committees have no authority to make decisions (i.e.,
          State Capacity).   Too much consensus building.

     *    Lack of budget, planning and resource (staffing)
          stability in IRM.

     *    Agency is not a stable steward of information.

     *    Capabilities of computers (technology) are often
          oversold - management does not understand support that
          is necessary to develop and maintain systems and what
          can or can not be done with computers.  Some managers
          may be proponents of computers but never use them..

     *    Undefined process for making IRM decisions in the
          Agency.

     *    Program Managers make decisions without assessing
          impacts or consulting OIRM/NDPD.

     *    NDPD and OIRM make unilateral decisions without
          consulting about or assessing impacts on customers.
          Lack of effective communication plan and obtaining
          information from user community for strategic planning

     *    No follow-through on IRM decisions.

     OTHER COMMENTS;  Some participants indicated that several
     years ago the SIRMOs met, formed a group and provided
     solutions to OIRM/NDPD but there was no follow-through by
     OARM.  SIRMO group eventually disbanded.
SOLUTIONS;
          Create Assistant Administrator for IRM (Corporate
          Information Officer - CIO) that reports to
          Administrator with no ancillary responsibilities other
          than IRM (including IRM_ .strategic planning) .   Must have
          responsibility, authority and expertise.   (There was

                              IV-17

-------
                                        APPENDIX  IV

discussion on whether this should be a career position
and the need for someone to Jbe a political appointee  to
ensure adequate power was voiced;.

Administrator should hold AA'-s accountable and
responsible for IRM effectiveness  (able to devote
resources).

Revise or restructure IRM in Agency  (including giving
SIRMOs more authority).  Need a fresh view.

Develop a consistent IRM structure (functions and
responsibilities) in program offices and have
consistent treatment of regional and headquarters IRM
functions.

Define IRM functional responsibilities and
accountability.

IRM functional positions need to be filled with
qualified people.

OARM/IRM and programs needs to hire staff with IRM and
programmatic knowledge or obtain expertise through
rotations, details, etc.

Provide training in IRM functions.

Educate senior management on roles and responsibilities
in IRM.

IRM Community must improve communication with senior
management (use less jargon).

Develop IRM strategic plan tied to mission.

Any major IRM initiatives,  standards, etc. that are
developed or changed need to go through green border
review process with cost/benefit analysis.

OARM (OIRM/NDPD) need to be customer oriented.

Establish working capital fund for IRM (tie budget
cycle to system life-cycle).

AA's need to be part of focus group with no delegation
of attendance.

Establish process for incorporating IRM in regulation
development and guidance.

                    IV-18

-------
                                                  APPENDIX IV

PROBLEM;  Lack of consistent Agency architecture  (hardware,
          software, data), strategy and lack of power to enforce
          it.  No consolidated approach to EDI.   [Indirectly
          trying to enforce it through budget].

ROOT CAUSES:

     *    No IRM leadership or direction  (via plans) and no
          enforcement.  Program offices go out on their own
          because there is no leadership.

     *    EPA can not respond effectively to rapidly developing
          technology.

     *    The EPA procurement process stinks  (ineffective and
          cumbersome).  Procurement drives hardware, software,
          etc. (not architecture baaed on  user needs).  Lack of
          sufficient ADP procurement expertise in 0AM.

     *    Too much contractor input (at NDPD) on architecture.

     *    The people making architectural decisions (NDPD) do not
          understand how computers (technology) are used in the
          Agency or by the States/local governments (not paying
          attention to customer's needs).

     *    No consideration of needs in exchange of
          data/information in architectural decision-making
          process  (technology for technology's sake).

          GROUP COMMENT: NDPD held a meeting last summer to
          address architectural issues (only one participant knew
          of the meeting).  Only two systems managers were
          invited from HQ.  The rest were IRM Branch Chiefs and
          contractors.  The group remarked that no results have
          been seen from this effort.

     *    NDPD does not talk/listen to customers.

     *    No mechanism for determining Agency standards or making
          changes to existing architectures.

     *    Lack of data standards and the ability to implement or
          enforce them.

     *    No common data definitions,  especially through the
          legislation/regulations.
                              IV-19

-------
                                                  APPENDIX IV

     *    Current "stovepipe" standards, regulations, and
          management inhibit standardization.  Some special
          purpose (vs. corporate) data is OK.

     *    SIRMO's have no power to ensure that standards are
          enforced (offices avoid getting their signatures).

     *    Resources (money) not available to implement standards
          (i.e., locational data policy).

     *    Items in the budget dealing with data standards are the
          first ones cut.

     *    No consideration of data as a corporate resource;
          responding to stovepipe statutes.

     *    NDPD/OIRM do not view themselves as a corporate
          resource for data and training.  NDPD offers no
          training for mainframe packages  (Statistical Analysis
          System (SAS),  etc.).  They do not view themselves as
          responsible for this corporate resource.

     GROUP COMMENT;  Several individuals did point out that NDPD
     has WIC training but no one shows up.  There is no
     management (programmatic managementJ commitment for IRM
     training.
SOLUTIONS;
          Develop an implementation plan for EDI in the Agency
          (not written by OPPE).

          Develop a legal policy for EDI signatures for external
          Agency entities.

          Develop a process for determining what is corporate
          data vs. special purpose data resulting in an
          enforceable "corporate data" policy.

          Establish a "data czar" who reports directly to the
          Administrator.

          Maintain relational reference tables  (such as zip
          codes, county FIP codes, etc.) as a corporate data
          resource.

          Ensure that there are sufficient resources in the
          budget to implement standards.
                              IV-20

-------
                                        APPENDIX IV

Require red or green border review  (as appropriate) on
Agency standards  (both new and revised), including a
cost/benefit analysis and an analysis of impacts on the
users and IRM community.

Exercise Agency discretionary authority in implementing
16 and GAO audit recommendations.

Undertake a comprehensive review of the Agency's
architectures (both data and technology).

Undertake a comprehensive review of how the Agency
spends its approx. $260 million annual IRM budget
(agency-wide IRM program budget).

Review and evaluate ADP procurement process (both large
and small purchases) to facilitate purchasing, increase
electronic commerce, and look for better models in
other agencies to emulate.  Need to do this at two
levels - work to improve on issues outside..the Agency
via National Performance Review and also issues
internal to the Agency.
                    IV-21

-------
                                                  APPENDIX  IV

                    THIRD FOCUS GROUP SUMMARY

                        SUMMARY SOLUTIONS

At the completion of the focus group exercise, the group was
asfced to develop five broad solutions based on the detailed
solutions presented by the group in response to specific
problems.  The following Jbroad solutions were presented by  the
group.

     1) The components of an information system  (e.g.,
requirements, needs, etc.) need to be identified and agreed to up
front.  There was general agreement that failure to identify and
agree to information system requirements up front results in
failure of the system to meet perceived needs later in the  life
cycle.

     2) Need to better anticipate needs by better communications
with customer  (i.e., Congress and oversight agencies).  Planning
needs to take place in a cooperative non-adversarial environment.

     3) There is a need for more resources.  There is a need to
consider all resource requirements before designing information
systems.  Resource requirements must then be prioritized and
compared to available resources.  Priorities may then be
addressed with available resources.  There was concern that
systems are implemented at the Regional level without funding the
resource requirements (unfunded liabilities) needed to operate
the systems (i.e., data entry, system maintenance, etc.).

     4)   Longer term stability of information systems is needed.

     5)   User input to information systems development is
critical.  There was general consensus that better communication
with users is necessary to anticipate needs.  In addition,  the
group generally agreed that better communication would lend
itself to building a cooperative atmosphere and better commitment
for data input and data quality.

REACTIONS TO IDENTIFIED PROBLEMS
       (Based on Previously Identified Problems/Root  Causes
            for IRM Problems Supplied  to  Participants)

The session started with the group discussing the previously
identified problems and root causes addressed from previous
audits and reports.  The following were not viewed as Regional
problems, however some of. them were discussed in the comments
below:
                              IV-22

-------
                                                  APPENDIX IV

PROBLEM:  Difficulties in addressing cross-media pollution
          problems.   (Primarily a Superfund focus group and not
          felt to be a Superfund issue).

PROBLEM:  Development of duplicate systems.

PROBLEM:  Exposure of financial payment systems to unnecessary
          access risks.

COMMENTS ON IDENTIFIED PROBLEMS
Many of these comments have bearing on the problem of cost
overruns and delays in development.  Many were carried over to
specifically identified problem discussions.

PROBLEM:  Lack of credible basic management information about
          where Agency funds, including Superfund money, are
          being spent and how, and the resulting accomplishments.

     1.   There is a lack of linkage between how funds are spent
          and accomplishments.

     2.   We do have "credible" management systems, but if
          questions coming from the Hill change. we may not have
          the answers in the data - Resource Limited.

     3.   Many of this summer's questions appeared to be directed
          at Superfund reauthorization - Is Superfund fair?  (And
          may not be recurring questions).

     4.   Constantly changing focus/changing questions - i.e.,
          Environmental justice -- there has been no
          data/information collected oh that in past.

     5.   Sometimes by the time questions are answered there is
          no longer need for answers.  Congress (and senior
          management) needs to understand.the impact of asking
          questions (resource and $ costs).

     6.   IFMS/CERCLIS still not linked - Not a Regional problem
         -but National systems problem.  However, Region would
          benefit from integrated system.

PROBLEM:  Difficulties in addressing cross-media pollution
          problems.

     1.   Superfund does not play in multi-media enforcement -
          not amenable because addressing past actions.  Region
          has used GIS successfully for multi-media issues
          however.  Region targeted major facilities for RCRA,
                              IV-23

-------
                                                  APPENDIX IV

          air, water - most generally OK.  Small facilities
          generally don't have multi-media problems.

     2.   Cross-media integration not an issue in Superfund.  We
          collect multi-media data for site.  Some data never
          need to be stored in database, question of where to
          store it (i.e,., different media specific databases).
          Mostly deal with program accomplishment reporting.
          Does Congress want an organic emphasis on reporting?
          Reporting structure is to different Committees/Sub-
          committees for different laws (i.e., specific program
          elements).

     3.   Planning horizon short (because of budget
          processes/politics),  a longer planning horizon would
          allow for more stability.  Also differences in program
          focus over time have occurred (technology vs. risk).

     4.   Congressional Sub-committee setup not encouraging
          multi-media activities.  Have to report on specific
          program elements.  Strategic planning can't work (tried
          in one Region but failed) because resources tied to
          legislation and there is an inability to move resources
          to where they are needed (Superfund resources can't go
          to water or vice versa).

     5.   Funds are program specific and not used for multi-media
          tracking system, therefore can not report if there is
          no activity.  Does this include environmental data?
          Typically EPA's tracking & info, systems do not.

     6.   Superfund collects multi-media data, but its relative
          impact is small  (dealing with localized problems).   So
          much planning goes into remedial actions that
          collateral impacts are generally not issues.

     7.   Sub-committee chairmen need to coordinate with each
          other.

PROBLEM:  Data integrity problems with EPA's mission-critical
          information systems.

     1.   Need user friendly, self-feeding systems (worry about
          data quality).   Have problems - reasons not always
          apparent.  EPA sometimes pushing IRM technology limits
          - causes delays, etc.  Major reason for cost overruns
          last year because of loss of TOSS (loss of contract
          personnel - prematurely - knew contract was going
          away).


                              IV-24

-------
                                                  APPENDIX  IV

     2.   Data accuracy/quality concern in future years.  Gaps  in
          budget formulation cause problems.  Data element
          definitions change, systems do not handle anomalies.
          Happens as technology changes (mainframe replaced by
          PCs) .

     3.   No traceability between Regional budget submissions and
          final budgets  (after negotiations, etc.).  SCAP data
          elements constantly changing - no data comparability
          from year-to-year - source of data quality problems
          (related to Congress?).

PROBLEM:  Development of duplicate systems

     1.   Systems developed by Headquarters not useful to Region
          - therefore Region develops own applications.

     2.   Changes to mainframe system happen slowly, need to
          supplement with PC systems.

     3.   Need to define what are common data elements (relate  to
          reporting needs) and which ones aren't (therefore allow
          for variability in individual's needs).   [Relates also
          to data integrity problem].

                       PRIORITIZED PROBLEMS

Problems are presented in priority order based upon the mulci-
voting scores given to each problem by the focus group
participants.  Scores, presented as  total points received, are
provided in parentheses following each problem.

     1)   Most information systems are designed for centralized
          Headquarters management and are not useful to the
          Regions.  Systems need to be designed as tools for
          staff, not merely to answer questions  (although they
          may do this).   (25 points)

     2)   EPA does not do a good enough job of anticipating
          questions from Congress.   (23 points)

     3)   EPA receives data of questionable quality from the
          States.  This was of particular concern because EPA
          receives most of their data from the States.
          (14 points)

     4)   Congressional Committee structure.   (13 points)

     5)   Data definitions.   (10 points)


                              IV-25

-------
                                                  APPENDIX IV

     6)   Linkage between financial system  and program systems
           (e.g., IFMS and CERCLIS).   (9 points)

     7)
                                          are asking  different
          questions.   (8 points)
     8)   Congress needs to revisit the Agency mission to  give
          more flexibility to programs.  There ia a continual
          fight for resources.   (5 points)

     9)   Overemphasis on contract, support versus in-house
          support.  This is applicable to all Agency  functions,
          not just IRM.   (4 points)

     10)  Need to develop information system with more
          flexibility so they can respond to changing needs.
          Inflexible systems lead to data integrity problems.
          There is no clear expectation or common understanding
          of what is desired from information systems.   (3
          points)

     11)  Cost overruns.   (0 points)

     12)  Need to take out negativism.   (0 points)

                      ROOT CAUSES/SOLUTIONS

PRIORITY PROBLEM #1:  Most information systems are designed for
centralized Headquarters management and are not useful to
Regions.  Systems need to be designed as tools for the staff,  not
to answer questions  (although they may do this).

ROOT CAUSES FOR PRIORITY PROBLEM #1:  Root causes for priority
problem #1 are presented in priority order as determined by group
multi-voting.  Scores, presented as total points received, are
provided in parentheses following each root cause.

     1.   Centralized management systems are designed by
          Headquarters for their needs.  (41 points-)

     2.   Misinformation from OIG, GAO, OMB.  What appears in
          audit reports and studies is often not an accurate
          representation of the condition.  (23 points)

     3.   Data system developers do not understand program or
          staff needs.  Part of this problem may be related to
          the fact that contractors (versus Agency personnel)
          develop most of our information systems.  (21 points)

                              IV-26

-------
                                                  APPENDIX  IV

     4.   Information systems are used for different uses than
          originally intended.  For example, the HWDMS system was
          originally designed as a tracking system.  However, the
          Agency tried to use as management system.   (16 points)

     5.   Complex vs. simple information systems - need vs.
          effort.  There is a need for information at different
          levels (e.g./ public, field offices. Congress,
          management, etc.) and different requirements (i.e.,
          effort, cost, motivation) for obtaining and maintaining
          data/information at these levels.  A data quality
          problem exists when Regional resources are required to
          collect information to meet Headquarters needs when
          perceived benefit and/or resources are not provided.
          (14 points)

The following additional root causes were identified for priority
problem #1.   However, because of our desire to address more of
the prioritized problems, we did not spend focus group efforts on.
developing solutions for these root causes.

     6.   Information systems development has been reactive
          (versus proactive) to Congressional needs.  (9.points)

     7.   There are too many managers and too few staff in
          Headquarters.  (6 points)

     8.   Information systems are used by data managers to build •
          empires.   (0 points)

SOLUTIONS:  Solutions are identified to the problems by number in
parentheses for clarity.  Subsequent sessions were not specific.

     1.   The people who are designing information systems need
          to understand what is needed. (1)

     2.   Consultation must take place between Regional and
          Headquarters personnel.  (1)

     3.   Need to determine if other Federal agencies have
          information systems we can use.   (1)

     4.   Information system designers should be required to
          function as a Regional project manager.  (1)

     5.   Need pilots and ground truthing  (like that used with
          the RP2M program).  (1)

     6.   Need to tie into existing information systems - close
          the loops.   (1)

                              IV-27

-------
                                             APPENDIX IV

7.   Information systems need to allow flexibility for the
     Regions.   (1)

8.   Headquarters needs to listen to the Regions and vice
     versa.  (1)

9.   Headquarters needs to be more consistent in providing
     either FTEs or contractor dollars to maintain and
     perform data input into information systems.  (1)

10.  We need to identify what we truly need up front.   (1)

11.  We need to anticipate future IRM needs fdata and
     information^.   (1)

12.  Information systems needs to stop being all things to
     all people.   (1)

13.  Need more experienced auditors (particularly experience
     in program areas).  (2)

14.  More regional involvement in audits.  (2)

15.  Auditors need to discuss resources as they influence
     outcomes.   Auditors need to analyze if program
     operations are doing the best they can with given
     resources, instead of auditing against a standard that
     assumes adequate resources.  (2)

16.  Audits need to identify positives.  (2)

17.  Regions need a second opportunity to review report*
     before final reports are issued.   (2)

18.  OIG, GAO,  OMB should care if information reported in
     the audit (results of audit) are. correct.  (2)

19.  Auditors need to follow the 1988 GAO audit guidance.
     (2)

20.  Programs should be less defensive and admit when there
     are problems.   (2)

21.  Both sides need to be more objective in the spirit of
     Total Quality Management.   (2)

22.  Programs need to do a better job responding to audit
     reports (both draft/final).  (2)
                         IV-28

-------
                                             APPENDIX  IV

23.   Need more in-house staff for certain development
     functions.   (3)

24.   Information system developers need to listen to end
     users.  (4)

25.   Information system developers need to understand  needs.
     (4)

26.   Adequate resources need to be provided.   (4)

27.   Programs need to admit when a new information system is
     needed, instead of trying to operate with a system that
     no longer works (ex. CERCLIS is broken beyond repair,
     nested loops in CERCLIS keep growing).  (4)

28.   Need to recognize that we need to have different
     information systems to do different things.  It is not
     always necessary to combine all the satisfaction  of all
     needs into one system.  Decisions of this nature  need
     to be discussed and resolved at a high level of
     management.  (4)

29.   Don't assume you need an integrated information system.
     (4)

30.   Feed information as needed.  (4)

31.   People making decision regarding IRM issues should have
     a background in computers.  (5)

32.   Determine if one information system is needed (versus
     multiple systems).  (5)

33.   Scrap old information systems when necessary (CERCLIS).
     (5)

34.   Design information systems properly.   (5)

35.   Assess cost of fulfilling data needs.  (5)

36.   Decision makers need to have background in ADP (should
     not have to rely on contractor's technical expertise).
     The Agency needs in-house expertise.   (5)

37.   Need long-term planning (high turnover change).    (5)

38.   Decision should be made at Program level  (by those
     persons committing resources)  in concert with the
     Administrator,  GAO, and OMB. .  (5)

                         IV-29

-------
                                                  APPENDIX  IV

PRIORITY PROBLEM #2: Problems anticipating Congressional
questions.

ROOT CAUSES FOR PRIORITY PROBLEM #2:  Root causes for priority
problem #2 are presented in priority order as determined by group
multi-voting.  Scores, presented as total points received,  are
provided in parentheses following each root cause.

     1.   The Agency is in a reactive mode.   (43 points)

     2.   Poor communication with Congress.   (39 points)

     3.   Too many hidden and/or different agendas.   (28 points)

     4.   Too busy.   (8 points)

     5.   No marketing plan for customer.   (6 points)

     6.   Changing Congressmen/Congresswomen and changing
          interest.   (4 points)

     7.   Congress is not interested until it is time to get
          votes.   (2 points)

     8.   Have to think like Congress  (anticipate congressional
          needs).  The group was not sure if this was a realistic
          expectation.  (0 points)

     9.   Don't debate diaries.   (0 points)

SOLUTIONS:

     1.   More ongoing communication with Congress  (borderline
          lobbying).

     2.   Decide up front what data EPA needs in order to manage,
          then be consistent  (need long term stability).

     3.   Programs need to be proactive in asking Congress  what
          they want to know.

     4.   Congress needs to be to be more specific  regarding  what
          they want (i.e., "specifics" vs. "general inquiries").
          This level of specificity needs to take place
          throughout the information systems development life
          cycle.

     5.   Meetings between EPA and Congress should be attended by
          higher level people.  Subcommittee attendance at
          Regional EPA briefings has been poor.

                              IV-30

-------
                                                  APPENDIX  IV

   .  6.   Need to focus more on IG, GAO, etc.  (i.e., the ones who
          are raising the issues - misinformation).  Need to  work
          on relationship with these groups and need to work  with
          an open mind.

PRIORITY PROBLEM #3:  Concern for quality of data from States
(most data comes from States).

ROOT CAUSES FOR PRIORITY PROBLEM #3:  Root causes for priority
problem #3 are presented in the order in which they were
presented by the group.  Because of our desire to Address more
priority problems, we did not attempt to prioritize root causes
for this or subsequent problems.

     1.   State personnel are not clear on what data to input or
          how to input data.

     2.   There is a general lack of resources (e.g., people,
          equipment, computer systems linkups, procurement
          system, etc.).

     3.   EPA always seems to want more data or changes the data
          that it needs.

     4.   Information systems developers do not involve States in
          system design, even though States provide most of the
          data.  According the group, reductions in Headquarters
          eliminated State involvement during data element
          definition development.

     5.   There is no National or Regional consistency.  This
          results in States entering data under different
          pretenses.

     6.   Do we have good data quality audit reports?

     7.   Need to emphasize to States the benefits of using the
          information systems we provide.  The group believes
          that the States will be better motivated if they
          recognize a benefit.

     8.   No way to enforce data quality at the State level.

SOLUTIONS:

     1.   Need more resources..

     2.   Better define what EPA needs are.  This will help get
          "buy-in* from States, distinguish between EPA and State
          needs, and provide specifics that States could

                              IV-31

-------
                                                  APPENDIX IV

          implement in their own systems to pass information to
          EPA.

     3.   Need to have State involvement in information system
          design.

     4.   Need to allow States more flexibility in tailoring
          information systems.  We may be able to provide
          resources for State information systems while
          identifying core data requirements for EPA systems.

     5.   Assure we have good data audit reports.  Need to
          conduct audit work at the time of information system.
          design.

     6.   States need EPA assistance in the procurement of
          equipment.

     7.   States need to receive more and continual training on
          information systems.  States have high turnover.

     8.   Headquarters needs to establish a hotline for data
          definitions and information.  This will support
          National consistency.

     9.   Need to determine if we can use the grants process to
          apply leverage to the States or determine if there is
          another mechanism to provide money to States while
          ensuring that data meets EPA needs.

PRIORITY PROBLEM #4:  Congressional Committee structure.

The participants in the focus group agreed that this problem/
although valid/ was not a problem for which we could brainstorm a
solution.  Therefore/ the group choose not to address this
problem.

PRIORITY PROBLEM #5:  Data definitions.

ROOT CAUSES FOR PRIORITY PROBLEM #5:  Root causes for priority
problem #5 are presented in the order in which they were
presented by the group.  Because of our desire to address more
priority problems/ we did not attempt to prioritize root causes
for this problem.

     1.   There is a continual change in the definitions of
          programmatic outputs.

     2.   There is a redundancy in terms (i.e., a fine line
          between definitions).  The group believes that there is

                              IV-32

-------
                                                  APPENDIX  IV

          a lacJc of understanding of definitions between some
          constituents (e.g., EPA Programs and others such  as the
          Inspector General and CongressJ.

     3.   There is no common understanding of definitions.

     4.   There is a false assumption that Superfund sites  have
          common traits.  The group expressed that each Superfund
          site has unique characteristics which make "national"
          definitions or standardization difficult.

SOLUTIONS:

     1.   Common agreement on definitions from the constituencies
          (e.g., Congress, IG, GAO, OMB, Headquarters) need to be
          made.

     2.   Definitions need to be frozen for a period of time.

     3.   Definitions need to be revisited on a regular basis.

PRIORITY PROBLEM #6:  There is no linkage between financial
systems and program systems.  The group used the JFMS and CERCLIS
systems to discuss this problem.

ROOT CAUSES FOR PRIORITY PROBLEM #6:  Root causes for priority
problem #6 are presented in the order in which they were
presented by the group. - Because of our desire to address more
priority problems/ we did not attempt to prioritize root causes
for this problem.

     1.   Built as independent systems to deal with individual
          problems

     2.   No common goals

     3.   Incompatible formats

     4.   Did not consider financial data when designing CERCLIS

     5.   Too difficult to bring together

     6.   No resources to link

     7.   Inadequate preplanning

     8.   Needs changed over time- accountability more priority
          over tracking.
                              IV-33

-------
                                                  APPENDIX IV
SOLUTIONS:
     1.   Up front communications, planning
     2.   Consider all directions you can go (for development)-
         . CERCLIS/Finance/Contracts (diversity of users)
     3.   Anticipate future needs
     4.   Do not build one big system, use modules that will be
          easier to change
     5.   Need own technically competent people to build (govt.
          function - primarily design to be able to communicate
          during development)
                              IV-34

-------
                                                  APPENDIX  IV

                    FOURTH FOCUS GROUP SUMMARY

PROBLEMS IDENTIFIED

1.   Systems were not designed to do what they are currently
     asked to do, and an unwillingness to be forthright about it.
     For example, few of our data bases interact.  (52 Votes)

2.   Data quality is not known.  There is no comparability
     mechanism to determine whether data supporting EPA's
     mission-critical information systems is accurate or
     inaccurate.  (29 Votes)

3.   Difficulties in addressing cross-media pollution problems.
     (7 Votes)

4.   Significant cost overruns and delays in developing and
     implementing information systems.  (7 Votes)

5.   Development of duplicate systems.  (3 Votes)

           ROOT CAUSES  (LIMITED TO  TOP 2 PROBLEM  AREAS)

Problem 1

Systems were not designed to do what they are currently asked to
do, and an unwillingness to be forthright about it.  For example,
few of our data bases interact.

Root Causes for Problem 1

1.   Lack of an "Information Systems Champion" and accountability
     for information systems:

          Lack of top management understanding of the importance
          of information systems to the mission;  attention to
          information systems activities;  and commitment to
          supporting information systems.

          Top management officials not accountable for
         • information systems management.

          Senior Management does not understand the strategic
          value of IRM.

          Lack of knowledgeable, experienced, and forceful
          leadership in the IRM arena.

          Lack of commitment to enforce standards.
                              IV-3 5

-------
                                             APPENDIX IV

Inadequate planning and budgeting process for information
systems:

     Lack of a overall Agency business plan with long range
     goals, objectives, approaches- to meet goals and
     objectives, and associated costs.

     Budget cycle not consistent with technology development
     activities.  System development projects are not line
     item budget items--funding comes from program areas.
     When program budgets get cut, the system development
     projects suffer.  For example,  Integrated Contract
     Management System (ICMS) .was a 5-year project and is
     currently floundering due to program budget cuts.

     o    Management lacks staying power in the budget
          process.
     o    Inconsistent funding.
     o    Support services are the first item to go.

     Administration, Congressional,  and EPA top management
     turnover change policies and priorities of information
     systems management.

     Management unwilling to redo big systems when
     information requirements change.               '

     Changes in technology overcome information systems--
     systems become obsolete.

     Technology becomes dated with delays.

     Computer support not tied into performance measurement
     (i.e., success/accomplishments/economic benefit).

     Technology activities not tied to mission.

     Difficulty in using raw data for meaningful reports.

     Lack of an overall Agency business plan which ties
     budget with IRM needs.

     Limited resources invested by OIRM in planning, policy,
     and oversight.

     End users are not being considered in developing and
     maintaining information systems.  The role of States is
     not being recognized.  Systems are not user friendly. .
                         IV-36

-------
                                                  APPENDIX IV

3.   Lack of built-in flexibility and integration of Agency
     information systems.

          No Agency requirements for multi-media approach to
          pollution prevention and enforcement.  For example:  2
          States are moving away from a single medium approach
          for pollution prevention.  One State is integrating
          information regarding the air, water, and permit
          activities of the top 400 facilities.

          No centralized standardization of data elements

          Limited initiatives supporting data integration/manage-
          ment.  Good examples are the Integrated Task Force
          Monitoring (NOAA, EPA, and USGS); GIIS (DOD, DOT, EPA)
          involving the standardization of analytical data
          definitions on Superfund Federal facilities; and the
          Global Position System.

          Lack of consistent data dictionaries (e.g., multiple
          data dictionaries for common data elements with
          different data definitions).

          Lack of system development standards.

4.   Unrealistic expectations for information from Agency
     information systems.

          EPA is seen as being on the cutting edge of technology,
          which is not true.  This raises false expectations for
          EPA in the eyes of Congress,  States, and employees.

          Management is not forthright with top management and
          Congress regarding the inability of our current systems
          to support current requirements.

          o    Even when requirements are well-defined in the
               beginning (e.g., RCRIS-$millions to define
               requirements; best example of collaborative effort
               with States and other users to define
               requirements) new requirements may be difficult to
               implement and result in complaints of not meeting
               user needs.  .

          Extreme difficulties are encountered in projecting
          expectations for information.

5.   End users not adequately consulted.

          Difficult to identify and communicate information

                              IV-37

-------
                                                  APPENDIX IV

          system projects with all customers with vested interest
          to get their involvement and contribution.

          EPA is not recognizing the role of the States in
          implementing and maintaining information systems.

6.   Insufficient qualified information system personnel staffing
     due to budget and personnel policy constraints is not
     competitive with the private sector.

Recommendations for Problem 1

1.   Establish an information systems "Champion" (i.ef, Chief  .
     Information Officer) as a separate Assistant Administrator
     position (or equivalent).   The person selected to this
     position must have good management, agency background, and
     technical qualifications.   This person must be bold enough
     and skillful to describe the lack of systems capabilities
     when appropriate.  This position requires access to the
     Administrator and requisite power and authority to execute
     its responsibilities.  We recommend that the Administrator
     look to the private sector for a model and/or recruitment.

2.   Develop an overall Agency business plan with long range
     goals, objectives, approaches to meet goals and objectives,
     and associated costs.  Business plan should include high
     level costs including a line item for information systems.
     The Chief Information Officer needs to be involved in this
     process.

3.   Clearly define Agency-wide information systems needs.
     Identify information systems requirements needed to meet the
     Agency's missions.  These needs should reflect the
     requirements of the end users.  Key customers should k>e
     involved in this process.

4.   Develop a strategic information systems plan which includes
     detailed costs.  Identify tasks to be completed annually
     with the funds.

5.   The strategic information systems plan should .assure that
     systems are flexible enough to deal with changing program
     needs and changing technology.

6.   Update Agency-wide information systems requirements and the
     strategic information systems plan annually with associated
     costs of changes.  Factor in technology changes.

7.   Information systems "Champion" must represent the strategic
     plan implementation in the budget.

                              IV-38

-------
                                                  APPENDIX IV

8.   Develop personnel systems that allow recruiting and
     adequately compensating people for jobs they do, rather than
     for their credentials.

Problem 2

Data quality is not known.  There is no comparability mechanism
to determine whether data supporting EPA's mission-critical
information systems is accurate or inaccurate.   (29 Votes)

Root Causes for Problem 2

1.   Lack of focus and attention to data administration/
     management-.

          Most data administration/management related activity is
          on a system by system basis as opposed to an Agency-
          wide basis.

          No data audit process by programs exists to. ensure that
          data elements/definitions within individual systems is
          consistent.

2.   Changes in data definitions by Headquarters program offices
     on a year to year basis, (e.g., CERCLIS--changed definition
     of remedial investigation/facility study (RI/FS) for Federal
     facilities in 1993).

3.   Differences in interpretation of data definitions between
     offices (e.g., site definitions).

4.   Changing legislation and requirements.

          New laws and regulations come out every year which
          require changes to systems which are not funded.
          Systems can't keep up with the changes because of lack
          of funding and delays due getting contracts in place.

          For example: The Agency's inability to make required
          changes to the CERCLIS Accomplishment Report for 1993
          due to contract problems and delays.

     Lack of resources and priority for data administration/
     management.

          When budget cuts occur, ADP support services is one of
          the first things to go, which reduces system
          maintenance.
                              IV-3 9

-------
                                                  APPENDIX IV

5.   Complexity of design of systems.

          Program systems so large and complex that ADP and
          program knowledge is required to use the systems.

          Systems are not user friendly.

          Difficult to keep up with changing systems.

          For example:  PCS information is updated by the States
          through interfaces with related States systems.  Any
          changes to PCS affects the States systems and
          interfaces.

          Even report writer software is too complex.

6.   Users of information not educated as to the capabilities and
     contents of the systems (e.g., IFMS/CERCLIS, IFMS/DOCKET).

7.   Lack of adequate training in some cases.

          Changes in systems and constant turnover of personnel
          require annual training.

8.   Lack of interface between IFMS and CERCLIS  (i.e., CERCLIS
     obligation data not updated from IFMS).

          System design problems in both systems-preclude easy
          interface.

9.   Technical personnel gathering data have little or not vested
     interested in data gathered in some cases (e.g., CERCLIS).


Recommendations for Problem 2

1.   Data management must be considered as a "core" program
     rather than a secondary function within each program with
     the same weight as other program functions.

2.   Senior management (i.e., Administrator)  must be educated as
     to the strategic importance of systems/data and of the
     impact of programmatic changes to systems.

3.   Establish management and staff accountability for data
     (i.e., ownership of systems and data).

4.   Trade information systems contractor support for more FTEs.
     (Contractors have no vested interest in information
     systems).

                              IV-40

-------
                                                  APPENDIX IV

5.   Provide more funds, people, equipment, and software for data
     administration and management.

6.   Increase Regional involvement in legislation/rule making
     process.  Regions have a better understanding of the
     impacts.

7.   Redesign systems to allow easier adaptation to change, and
     control by Regions.

8.   Allow Regions more control over their aspects of systems.

9.   Change budget process to establish direct budget line items
     for information systems activities in order to reflect
     accountability for resource allocation for information
     systems and Regional portion of that allocation.

10.  Survey Regions and Headquarters to determine the actual
     level of resources expended for data management and systems
     support.

11.  Simplify the contracting process to avoid delays and to
     quickly respond to changing requirements.  For example:
     (a) 4-6 months to execute a delivery order on an existing
     contract; (b) 9 months to get an 8A contract in place; (c)
     2-3 years to get an IRM contract in place is unacceptable.

12.  Allocate more information systems FTEs to the regions.

13.  Establish Agency-wide directories of systems and data
     (information).

14.  Educate the users of the information from systems (not
     system users) on the capabilities,  contents,  and limitations
     of the information systems.

15.  Develop standard data definitions across information systems
     with common data elements.
                              IV-41

-------
                                                       APPENDIX V
                     SUMMARIES OF INTERVIEWS
BACKGROUND:  This Appendix presents the transcripts of the series
of interviews conducted on special emphasis areas (such as data
integration) and with particular people-—two Regional SIRMOs.
The transcripts are detailed narratives of the meetings.
                               v-i

-------
                                                       APPENDIX V
                SUMMARY OF HEADQUARTERS INTERVIEW
                    SUBJECT:   DATA INTEGRATION
PROBLEMS IDENTIFIED

1.  EPA is deficient in the area of data integration (DI):

          internal links among EPA mainframe systems are not
          there
          external links to other agencies, states, etc. are not
          there

There is no unified user interface in the area of performance
indicators, environmental quality, other data fields.

2.  EPA staff cannot always get information they need and they
have, over time, stopped asking many important questions because
we simply don't have the data to provide the answer or can't get
it out of the systems we have.  Decisions based on limited
amounts of information becomes a general course of business.
This practice is then defined as the acceptable level.

For Example:

RCRIS/CERCLIS — Reconciliation is difficult between
RCRIS/CERCLIS and permit writing is constrained by information
availability.  Its difficult for Superfund and RCRA systems to
answer questions asked now.  Five and a half years ago different
questions were asked.  Systems were not designed for the
questions being asked now.  It is not possible to anticipate the
types of questions that will be asked.

PCS — Provides a pointed set of information on standards versus
discharges but doesn't bring in wetlands, population impacts,
ecosystems, proximity (of contamination) to the well supply.

GIGS— Congressional questions originally asked, are no longer
asked because information is not available.

CIS — Programs resist because data is entered into separate
systems.  CIS users showed management that the system had
incorrect data that showed Superfund facilities in the middle of
the Atlantic Ocean and at the North Pole.

STORET cannot show how much contamination is coming out into a
body of water.
                               V-2

-------
                                                       APPENDIX V
3.  There is a pervasive perception that EPA information systems
have no data integrity.  Information systems have primary and
secondary users.  The primary use of Agency data is successful,
i.e., CERCLIS helps SCAP preparers provide good data to
Headquarters staff using it.  The problem in 01 is secondary use.
While primary users of CERCLIS have the information they need,
secondary users seeking the following information will not find
answers through CERCLIS:  what is the nature of the hazard; is it
a PCP site; where is the hazard located, what is the degree of
the threat posed by certain chemicals, and sites, etc. are not
successful because this information was not included in the
system requirements during development.  This creates the
perception that data are bad—even when the system works for the
primary user.  The problem surfaces especially when the Congress
and others attempt to select information the system never
collected and never intended to provide.

ROOT CAUSES

1.   Systems independently designed and managed lack both EPA and
     Government-wide data standards in many areas such as
     organisms, naming conventions, data architecture.  Without
     standards system managers will continually be putting bad,
     unintelligible information into employee's personal
     computers.
     TQM may provide correction, however the competent people
     involved are not tuned in to the protection of the
     environment.  The culture doesn't foster meaningful
     discussions to determine commonality, instead each have
     their own life, no common vision.

2.   EPA has not thought strategically.  The Agency culture does
     not value data:  it is of no use; its importance is
     minimized.  There is parochial focus on data, even down to
     ecosystem level.  Data/information are not considered
     strategic resources.  IRM is considered an overhead activity
     rather than a key element of EPA business for the purpose of
     enforcement and monitoring.  Too much impetus behind current
     system development has been on bean counting:, how many
     permits, inspections, enforcement activities, not the
     environmental data needed by users.

3.   The statutory framework promulgated by Congress doesn't
     address cross media.  Program managers are driven by a
     single Act, a requirement for reporting on activities.
     Statutes compartmentalize and drive the Agency in that
     direction, mandating bean counting rather than real
     environmental results and DI~.  This orients managers toward

                               V-3

-------
                                                       APPENDIX V
     collecting specific information.  EPA has not addressed this
     problem as other agencies have successfully.  EPA was not
     aggressive in writing regulations to facilitate integration.
     There is very little movement afoot to go in a integrated
     direction like TRIS and CIS.  Five years ago resources to
     support CIS were pulled from all over.  Five FTE's were
     identified from each Region.  The outcry was such that
     resources were pulled back from GIS.  The Agency lacks
     resources and funding for FTEs which increased its
     dependence on contractors.

     The DI problem is not unique.  USDA historically had the
     problem...everyone is under the gun to meet daily
     operational needs...there is no slack to look at the big
     picture.  USDA has been successful because they have less
     dependence on contractors.

4.   Administrations don't emphasize IRN:  Nineteen teams were
     assembled to conduct the National Performance Review at EPA.
     None addressed information management as the keystone —
     strategic importance in the plan.  Pollution Prevention is
     addressed but the importance of IRN was missed.  This
     continues the pattern — neglect IRM.  The OE reorganization
     is not emphasizing DI as one of the objectives.

5.   It is still acceptable not to integrate data—there was no
     penalty to be paid for not wanting integrated permits...it
     was, however, not acceptable not to meet the* mission or not
     to bean count or not to get out the permits.  Systems were
     built to do specific things and are still doing them.  The
     data quality questions are highlighted but no harm came to
     EPA.

     EPA missed an opportunity to move forward on DI.  TRIS, an
     Information Program ordered by Congress, is hailed as one of
     EPA's successes, yet the Agency hasn't expanded lessons
     learned here to other systems.  STORET has only 500 users
     while it should have 5000 users.  Office of Water is taking
     solid looks for providing for secondary users and seeking
     standards for water data.

6.   Information systems have primary and secondary users.  If a
     system under development for a primary user has no need for
     latitude and longitude data, it is not collected.  Nobody
     can enforce other programs to meet needs of another office
     where the one office does not control the resources of the
     other. Current trends of demographics analysis that deal
     with population at risk now need boundary data and latitude

                               V-4

-------
                                                       APPENDIX V
     and longitude.  At present, some Agency information systems
     are required to provide geographical location.  The Agency
     still has not defined its business in terms of latitude and
     longitude of information.  There is risk involved  (resource
     and technical risks) to projects for incorporating
     integration to other projects.  Program managers believe,
     "If I have to alter my plans to handshake with you I
     introduce increased resources and risk; I won't do this
     unless someone makes me.1*  There is no requirement for
     programs to consider secondary uses during systems
     development.   People may or may not be
     rewarded for extra risk and effort, therefore programs
     usually avoid the risk by ignoring DI.

RECOMMENDATIONS

1. Top management needs to mandate DI to enable EPA to:

          better assess risk and target areas needing action
          implement a risk ranking program to provide a
          systematic method to compare threats, and rank and
          evaluate how we are doing against threats
          measure outcomes
          meet the Administrator's 4 priorities which cut across
          all media:

          o Ecosystem protection
                                                a

          o Pollution Prevention

          o Environmental Justice/Equity

          o Partnership building

          revise the Agency culture such that people can and will
          work together more on DI projects by encouraging
          developers to consider other office's needs

          recognize that DI takes resources, creates risks, that
          may complicate one office's systems
          development/operation efforts

          integrate DI requirements into the day-to-day business,
          i.e., enforce environmental equity into the process by
          mandating the equity requirement be met before issuing
          permits
                               v-5

-------
                                                       APPENDIX V
2.   Elevate OIRM in the organization.  Establish a data
     administration center, central group for DI, and a CIO~a
     strong General Patton type with 'status/stature/authority.
     Solving the problem is a matter of the vision and will of
     top management to:

          Build sound IRM processes in the Programs

          Build strong IRM programs in the Program areas

          Put the existing resources and logistics to work under
          strong leadership

          Develop and enforce data standards

3.  Sound thinking about IRM needs to be built into the thinking
of program personnel—raise the strategic role of information in
protecting the environment:

          Consider an alternative to the IRM steering Committee -
          -high level managers don't deal with issues

          Program officials on the Committee should have more at
          stake

          Assistant Administrators don't know ADP/IRM lower l«v«l
          issues.  They don't have authority over it either.

          EPA should follow the path taken by FAA, Patent and
          Trademark, NWS, SEC: view information as a core
          business line; institute a major modernization effort.
          Revisit lessons learned during one of EPA's major
          efforts in ORD and OW, EMAP and STORET modernization;
          determine if a link exists between systems.

4.  Link in to the data highway bandwagon and open all non-CBI,
non-enforcement sensitive, non-privacy act information to the
public

     would identify data quality problems
     would require move toward continuous improvement in data
     quality by making EPA more accountable for data quality (can
     we do it without.creating fear within EPA?)  provide TQM
     amnesty—line it up so that data managers will not suffer
     repercussions.  Be prepared for chaos.
     would create greater demand for DI
     ensure the customer knows where to go for information—set
     up an 800 number

                               v-6

-------
                                                       APPENDIX V
5.  Create consistency, standards and naming conventions. This is
the critical path to success.  Review the case in point, TOXIC
loadings model — the Great Lakes Project — which is dealing
with multiple data standards, for example, how much chlorine
loadings?

(Also important is the) Institutional ability to use DI.  Do the
users have the ability to ask the questions that force DI?
What are the questions that demand DI in'order to be answered and
who's asking them?  The public's naive questions require DI. (How
are we achieving) ecosystem protection data/pollution prevention
data/where are the integrators.  We have no background, no
experience.
                               v-7

-------
                                                       APPENDIX V
                SUMMARY OF HEADQUARTERS INTERVIEW
              SUBJECT:  OFFICE OF AIR AMD RADIATION
PROBLEMS AND ROOT CAUSES

LACK OF CREDIBLE BASIC MANAGEMENT INFORMATION ABOUT WHERE AGENCY
FUNDS, INCLUDING SUPERFUND MONEY, ARE BEING SPENT AND HOW, AND
THE RESULTING ACCOMPLISHMENTS

     This previously identified problem is not considered an
issue in the AIRS program.  It was felt that they have a
"complete financial structure" and can identify where funds go,
how they are distributed, and what they are used for.  This is
possible because the entire AIRS system budget is contained in
one Branch budget.

DIFFICULTIES IN ADDRESSING CROSS-MEDIA POLLUTION PROBLEMS

     "Root causes" preventing information integration include
both programmatic and philosophical issues.  An example of a
programmatic difficulty is related to the "lack of support" that
FINDS receives.  This lack of support has resulted in programs
being reluctant to participate in the program and a "lack of
trust" by the programs.  Although the FINDS program is a good
idea, OIRM is "dropping the ball" by not properly supporting it.

     Philosophical differences are related to the different
"power groups" outside the Agency (i.e., OAR/NOAA 6
OW/COE/USFWS/USGS) involved in the process.  The fact that there
is not a lot of commonality between programs leads to serious
difficulties in cross-media integration.  The recently
reorganized Office of Enforcement will have a hard time
addressing cross-media enforcement issues because of the lack of
commonality between programs.  There is also a lack of
compatibility between programs.

     Recent budget cuts are significantly setting back long-term
efforts at data integration.

SIGNIFICANT COST OVERRUNS AND DELAYS IN DEVELOPING AND
IMPLEMENTING INFORMATION SYSTEMS

     They did not believe that cost overruns and delays are
necessarily problems (in the AIRS program).  Cost overruns are of
a continuing nature and are necessary because information systems
continually need modification.  For example, they recently
received $300,000 for IRM support" to implement new requirements

                               V-8

-------
                                                       APPENDIX V
related to the Clean Air Act authorization  (caused expansion in
users, requirements, etc.).

     Lack of centralized, unified data management guidance is
resulting in Regions and States "going their own way."
Additionally, Agency programmatic managers have not "bitten the
bullet" on identifying what information will be needed to run
programs in the future.  This has resulted in no base budget
being established to develop the information systems necessary to
support future program operations.  Small policy decisions can
change systems in a significant way.

     There is no base budget to maintain software.  As a result,
50% budget cuts translate to putting developed systems on the
shelf.

DATA INTEGRITY PROBLEMS WITH EPA'S MISSION-CRITICAL INFORMATION
SYSTEMS

     Data integrity problems stem from the fact that there is no
consensus on who needs what information.  For example, some
Regions rely totally upon the States to provide information while
other Regions do not.  Additionally, managing the quality of data
received from the States is a problem.  Data quality standards
are not always established and if they are established they are
not enforced.   There is an inconsistency in data quality
standards.  For example, the Agency was not able to force the
State of California to enter zip code information into the AIRS
system because they are not convinced that the information is
correct.  Instead, the State enters XXXXX for the zip code.

DEVELOPMENT OF DUPLICATE SYSTEMS

     There is a belief that use of the mainframe is the best way
for EPA to do business.  However, there is tremendous pressure on
the programs to develop PC-based systems.  For example, Region 4
has developed their own PC-based AIRS system for use in the
Region and the States within the Region.  This system was
developed because Region 4 did not like AIRS (they found it hard
to use).  Duplication exists because systems are developed in a
Regional office to meet a specific need and then are given to
other offices or States for use because systems.  These systems
are then viewed as "Agency" systems without any Agency
endorsement.  They believed that a formal process would not help
the situation but only add another layer of bureaucracy.

     Nationally, many State systems duplicate EPA systems and
this duplication causes problems."  These systems are often not

                               V-9

-------
                                                       APPENDIX V
integrated.  It may be impractical to "force" one system (ex.
AIRS) on everybody, but data standards would certainly help.
AIRS is working with the ANSI standards committee to develop
national standards.  The problem will likely get worse because of
increasing demands for information (ex. recent new Clean Air Act
requirements).  The problem is further aggravated by the current
push against centralization and the push to empower States and
local governments and reduce oversight (i.e., NPR).  The
reorganization of the OE and movement of the air enforcement
function to OE may lead to further fragmentation over time (as
has happened with permit systems)..  In addition, "Title V" forces
have created a need for more systems and more money to be
provided to the states.

     One of the participants is chairman of an OAQPM data
management workgroup (task force) and the workgroup will produce
an "action oriented paper" around April 1994 addressing data
management issues.  The document is part of a long-term plan with
the Department of Defense to identify similar systems to combine
and integrate data.  The ability to integrate has been impaired
by recent budget cuts.

EXPOSURE OF AGENCY'S FINANCIAL PAYMENT SYSTEMS TO UNNECESSARY
ACCESS RISKS

     This issue is not believed to be a problem in the AIRS
program.

ADDITIONAL PROBLEMS

     One problem they have been dealing with for some time
(twenty-plus years) was the problem of data confidentiality.
There are cases in which data requirements are promulgated by EPA
and States refuse to provide the data because the data is
considered confidential by state statute and not confidential
(i.e., subject to FOIA requests) by Federal statute.  One
participant has been unable to get the Attorney General and
Office of General Counsel to make a determination as to the
confidentiality of some data in these cases.  In addition, there
are different standards of Confidentiality for industry and
States resulting in data integrity problems (certain data is not
reported).  In addition, vulnerability problems sometimes arise
when the level of security for particular data varies between
pieces of legislation.

     A problem area in hardware and software compatibility  is
related to the direction the Agency is taking technologically.
It is felt that with the new client-server technology, NDPD/OIRM

                               V-10

-------
                                                       APPENDIX V
has not established a new hardware/software direction for the
Agency.  Hardware standards are not being built and national
priorities are not being established  (e.g., future
telecommunications needs).  User's needs are not being solicited.

     0ARM'S consideration of cutting one of two NDPD mainframes
in response to FY94 budget cuts would be "catastrophic for
States11 and EPA would "lose a customer base we'll never regain".
NDPD used to roll out new equipment all the time but that real
future planning by one of the NDPD branches stopped 2-3 years
ago.  It was also felt that user funding (fee for service) isn't
really the answer because (1) the program's budgets don't have
money to pay for the services in the first place, and (2) the
program's budgets are also being cut.  User funding would be very
disruptive in the short run.

     The State/EPA Data Management program (SEDM) was felt to be
an excellent program in which State and EPA personnel (at the
working level) were able to work together on small multi-media
types of projects.  The participants in this project were very
enthusiastic about the work being done and the project
facilitated good communication between the Agency and States.
They believed this program was cut last year (beginning of FY93)
because of budget cuts.

     It was observed that States use AIRS in one of two ways,
directly or indirectly.  Direct users input data into AIRS and
extract data from AIRS for their own use.  Indirect users input
data into AIRS because they are required to provide the data, but
do not extract data from the system and feel this is an onerous
request.  When Regional officials support a system like AIRS, the
States in that Region are more likely to cooperate than when the
Regional office does not support a system (States are more likely
to go their own way).

ORGANIZATIONAL ACCOUNTABILITY

     Senior management recognizes the data collection function
but does not always recognize this function as a component of the
broad data management process.

     They believe there have been problems resulting from the
termination of the TOSS contract.  The termination of TOSS has
lead to the dispersion of system development activities and this
has resulted in less standardization.  They  feel the MOSES
contract is not a popular development vehicle because although it
can be used to develop a system, the programs cannot use it the
support the system.

                              V-ll

-------
                                                       APPENDIX V
     OIRM showcasing products, introducing tools, and providing
models in the past was appreciated.  There is concern that there
will be a shift, on OlRM's part, toward enforcement and away from
support.

     One organization has had problems getting personal computers
off the Agency contract.  Some PCs have been on order for "over a
year."  The problem seems to be that NDPD does not get the
support they need from 0AM.  Another office felt that buying
equipment off the contract does not result in paying the lowest
prices for the equipment.  They "get laughed at" when they buy
equipment at such high prices.

     Regarding the FMFIA process, it was felt that material
weaknesses have no relationship to resources.  For example, one
office can be performing alright with few resources and have a
need for additional resources while another office can have a lot
of resources, report a material weakness, and get additional
resources.  In fact, reporting an issue as a material weakness
often results in additional resources (i.e., failure breeds
success).

PLANNING AND RESOURCES

     They believed that OIRM's policy development process does
not always get all programs' input and buy-in, creating a "lose--
lose" situation.  Currently, programs are not plugged into this
process.

     IRM functions are currently taking a double cut in budget
cutbacks.  For example, program cutbacks result in cuts to
available program IRM resources in addition to cuts in OARM,
resulting in cuts to available non-program IRM resources on which
the programs depend.  The programs have very little say in OARM
discretionary budget cuts even though these cuts directly affect
program, operations.  The program office has dealt with severe
cuts for the past two years.  Part of the problem is related to a
lack of understanding, by the programs, of* the non-discretionary
portion of OARM'S budget.

     Communication does not always follow a proper process and
procedures nor is there awareness of sensitivities.  For example,
members of an audit team met with AIRS representatives to discuss
AIRS system maintenance.  However, communication problems in the
OAR chain-of-command resulted in the audit team taking some time
to identify the correct group of people to interview.  It was
felt that OIRM does not always keep programs adequately aware of
upcoming responsibilities.  For example, the office received a

                              V-12

-------
                                                       APPENDIX V
notice stating that a certain system security plan was due.  They
noted that they had little warning that the plan was due and
pointed out that the memo (re: installation security) did not
identify, or give guidance on, specific security issues to be
addressed (i.e., PC security or data security).

     Too many SIRMOs are part-time.  OAR does not have the
resources to staff the SIRMO position full-time.  OSWER was used
as an example of a better way to staff a SIRMO function.

     The Agency is not adequately planning and funding
information systems.  There is a feeling that the Agency is
behind in the client/server environment and there isn't enough
"muscle" in the hardware "architecture."  Some efforts have been
made to examine the direction the Agency should take.  Part of
the problem is that IRM does not get much visibility.

DATA INTEGRATION/DATA MANAGEMENT

     FINDS is a great idea that has "not hit the mark."  Some
problems exist with the reliability of the data maintained by
FINDS.  For example, some States refuse to use the EPA ID
(assigned by FINDS) as a primary identifier (although that was an
original intention) because of data quality problems.  It was
believed that Regions/system owners need incentives to clean up
data.  The Agency cannot implement policy and then "walk away." •
They felt that this is happening with the locational policy.
Apparently, several States are not willing to commit to
implementing this policy and nobody in EPA Headquarters is
forcing it.  They believed there are many complicated issues
involved with the FINDS program but that the whole process is
further complicated because of the lack of continuity within the
FINDS program.  It was pointed out that FINDS promotes linkage
rather than integration.  In addition, policies are not enforced
through Grants or by giving technology to States for free
provided States put resources into these projects.  The
fundamental issues of what data is needed in National Air
repository both now and in the future have not been addressed.

QUALITY ASSURANCE AND PROTECTION OF DATA

     Quality assurance was looked at during the FMFIA process and
a data quality plan is being developed.  However, the level of
budget cuts is making things difficult.  There are some
weaknesses.  For example, emissions (air) does not have data
quality objectives or plans but needs to develop them.  However,
Clean Air Act amendments have caused an overload and many things
are not getting done.  There has'been a fairly steady erosion in

                              V-13

-------
                                                       APPENDIX V
maintaining quality in data that is being collected.  Quality
assurance "falls by the wayside" when other requirements  (ex.
court-ordered deadlines or top management's "hot" initiatives)
are a higher priority.

COMMUNICATION

     It was felt that OIRM does not clearly identify and state
priorities, communicate these to everyone, provide training, or
solicit input on the effects of their efforts.  It was felt that
OIRM loses credibility when they ask for information and never
follow up.

SOLUTIONS

SIGNIFICANT COST OVERRUNS AND DELAYS'IN DEVELOPING AND
IMPLEMENTING INFORMATION SYSTEMS

     The Agency needs centralized, unified data management
guidance promulgated by Headquarters.

     Need national consistency in data (ex. emission inventory).

DATA INTEGRITY PROBLEMS WITH EPA'S MISSION-CRITICAL INFORMATION
SYSTEMS

     Data quality standards need to be enforced.
                                                 •
DEVELOPMENT OF DUPLICATE SYSTEMS

     OIRM needs to explore the issue of duplicative systems and
"official1* Agency endorsement of systems for use by other offices
and States to accomplish program missions (without adding
bureaucratic layers).

     Establish data standards for national systems for use by
States in designing systems.

ADDITIONAL PROBLEMS

     OIRM should establish a Bulletin Board System (BBS) to
provide advice on purchasing hardware and software to help solve
compatibility problems.

     Looking into the future in terms of planning and providing
long-term commitment and stability to major national systems was
emphasized as being very important.


                               V-14

-------
                                                       APPENDIX V
     Bring back the  (SEDM) program.

ORGANIZATIONAL ACCOUNTABILITY

     The Senior IBM  Steering Committee should move from an
advisory role to a decision making role and have strong technical
subcommittees providing input and advice to senior decision
making personnel.

     Senior management needs to buy into the data management
process (including the data collection activity).  There needs to
be a data management plan which would address needs from today to
5 or 10 years out.

     It is "critical" that top management '(political appointees)
understand and approve of data collection/systems development
efforts.  Programs need to "close the loop" with senior
management on data requirements (need top level concurrence) and
that commitment must be received up front.  "Don't create another
bureaucracy".  Senior management buy-in needs to be informational
(i.e., understanding and agreement) versus "getting a signature
in the right blank."  Senior management should share
accountability based on buy-in but buy-in need not be
"memorialized" in the form of a document.  Another "paperwork
exercise" is not needed.

     The SIRMOs should act as a "clearinghouse of information"
across their programs.  In this capacity, the SIRMOs would be in
a position to reduce the duplication of information systems.
OIRM should have a "fostering" role in systems development
projects.  Helplines could be used to support development
activities.  It is "impossible" for OIRM to track development
activities and a lot of development work is being missed.

     There needs to  be better communication with the procurement
office, Office of Acquisition Management (OAM), NDPD, and the
program offices related to buying equipment off Agency contracts.

PLANNING AND RESOURCES

     OIRM needs to get programs together and agree on policies
and standards.

     Better communication needs to be established within the
Agency concerning IRM matters.  Communication needs to follow a
proper process and procedures and be aware of sensitivities.
                               V-15

-------
                                                       APPENDIX V
     Move toward integrating PCs and mainframes in a
client/server environment to help the Agency "catch up" in the
IRM arena.  The Agency needs a centralized organization (NDPD) to
take the lead to make this happen.

    ' EPA needs to "put some muscle" into the hardware
architecture.  There needs to be better centralized leadership to
address common technical problems not "what are you doing in
AIRS."

DATA INTEGRATION/DATA MANAGEMENT

     Develop incentives for Regions/system owners to clean up
data in FINDS.

     OIRM needs to foster importance of policy through better
policy guidance.

     A National Air repository is needed and the fundamental
issues of what data is needed both now and in the future need to
be addressed.  A long term plan needs to be developed.

QUALITY ASSURANCE AND PROTECTION OF DATA

     Emissions (air) needs to develop data quality objectives or
plans.  To accomplish this, they need the full support of senior •
management (Deputy Assistant Administrator (DAA) accountability).

COMMUNICATION

     OIRM's priorities need to be clearly stated and communicated
to everyone.  In addition, OIRM's role needs to be more
facilitating, showcasing, and helping not just compliance.  OIRM
needs to ask if their efforts are helping.  They need to identify
priorities and provide training.

OTHER COMMENTS

     This was the "first time in 20 years anybody asked us these
questions" and we are very pleased to have been included in the
review.
                               V-16

-------
                                                       APPENDIX V
                SUMMARY OF HEADQUARTERS INTERVIEW
               SUBJECT:  SYSTEM DEVELOPMENT CENTER
PROBLEMS AND ROOT CAUSES

MANAGEMENT CONTROL

*    Change of people  (turnover of Contract Officers  (COs) and  .
     Project Officers  (POs)) hasn't helped.  Has led to
     inconsistencies on EPA's side.

DIFFICULTIES IN ADDRESSING CROSS-MEDIA POLLUTION PROBLEMS.

*    Agency not created "holistically."
*    When Agency not organized "holistically" info, and data can
     not be.  There have been efforts to improve/combine data but
     unless EPA is organized "holistically1* EPA cannot adequately
     address this.
*    System Development Center (SDC) not hearing a need for this.
     Only need for this (from big picture) is in the mind of top
     management.
*    Continues to be problems in complexity and difficulties in
     interpreting data (ex. Superfund can not agree on what a
     site is).
*    Difficulties in trying to add structure to .things that My
     not be able to be structured.
*    GATEWAY/ENVIROFACTS info, is not designed to be "meaningful1*
     when viewed together.  Have not looked at public needs.

ORGANIZATIONAL ACCOUNTABILITY

*    SDC sees spotty SIRMO involvement (weak link - pressures
     from other organizational roles they have).
*    Most Delivery Order Project Officers (DOPOs) need to be
     prodded for performance measures in DOs.
*    Programs had some problems factoring some areas such as
     product assurance into their process.

PLANNING AND RESOURCES

*    High turnover of CO's/PO's (MOSES).
*    Expertise of DOPOs mixed (not consistent).  Hard to get all
     expertise in one DOPO.
*    Hard to get away from treating on-site contractors as staff.
*    Frequent conflict -pressure to get something done vs. doing
     it right.


                              V-17

-------
                                                       APPENDIX V
*    Lack of knowledge of implications of this pressure and
     balance of requirements vs. good IRM on staffing.
*    Info, gathering in regulations not involved enough with IRM.
*    No consistency from year-to-year for IRM involvement up
     front.

QUALITY ASSURANCE AND DATA PROTECTION

*    Inadequate systems documentation on systems coming into SDC,
     hard to maintain, don't always know what changes will
     affect.
*    Can not always get back to sources.
*    SDC process forces policies, but there is a limit to what
     can be enforced when no funds are available.

SYSTEMS DEVELOPMENT LIFE-CYCLE

*    Most systems need constant enhancement, mostly because of
     regulation changes (change in information needs).
*    There is an unwillingness to do some things that are
     policies (i.e., documentation, life-cycle planning).

COMMUNICATION

*    DOPOs without background in IRM not knowledgeable or want to
     avoid it.  OSWER people generally aware of IRM guidance.
*    IRM management not "plugged in" or not strong enough to deal
     with Steering Committee.
*    Re: Policy, Procedures, Standards - Some programs are good,
     some bad.

SOLUTIONS. IMPROVEMENTS. AND ACCOMPLISHMENTS

MANAGEMENT CONTROL

*    MOSES process was designed to improve. SDLC
*    Management structure (both in SAIC contractor and EPA)
     developed to oversee work in SDC helps ensure things are
     done right (i.e., EPA oversight in development of project
     plan through negotiation and revision of project plan).
*    Audits have made a difference on program side.  IG reports
     have gotten program office management's attention.  Program
     offices trying to change including more management control
     (esp. in RCRIS) and IRM decision making.  Superfund is also
     improving/Govt. approves all changes - due in part to
     structure/management of contract (forces controlled change).
     Have reviews of projects once per month focussing on
     schedule/technical issues.

                               V-18

-------
                                                       APPENDIX V
PRODUCT ASSURANCE

*    Product assurance on each project (including configuration
     management and software QA) has paid off, although there
     have been complaints on added cost.
*    Provides better control over changes.  Contractor now
     provides recommendations, but EPA meets with all principles
     and agree on changes.
*    Getting better at prioritizing changes and selecting those
     that can be implemented within budget.  Better handle on
     where money is spent.

DATA MANAGEMENT

*    This summer's Superfund data collection was a success
     because of central consolidation of data management in SDC.

DIFFICULTIES IN ADDRESSING CROSS-MEDIA POLLUTION PROBLEMS

*    GATEWAY/ENVIROFACTS is a step toward pulling together
     data/info, that exists.  May be a front end to public
     access.  Pulling data from different systems points out
     differences.
*    Part of GATEWAY/ENVIROFACTS is also an effort to standardize
     data elements through data modeling to see what data is
     there.

SIGNIFICANT COST OVERRUNS AND DELAYS IN DEVELOPING AND
IMPLEMENTING INFORMATION SYSTEMS

*    MOSES process is trying to preclude this by planning and
     through development of a good project plan up front.
*.    However, MOSES is not dealing with entire Agency (only
     Water/OSWER/some ORD - big programs).  No acid rain work.
     Regions do their own thing.

DATA INTEGRITY PROBLEMS WITH EPA'S MISSION-CRITICAL INFORMATION
SYSTEMS/DATA MANAGEMENT

*    Data integrity rests more with programmatic area.
*    RCRIS - States are putting more resources into data
     integrity.
*    PWSS - States anticipating needs to have data integrity
     because they are involved in process.
*    Currently doing data management at SDC.
*    Working towards a better data definition standards in IMDA.
*    Efficiency & quality improvements have been noted by SDC
     centralization.  A 40% reduction accrued when geographically

                               V-19

-------
                                                       APPENDIX V
     separated functions (i.e., hotline problems and software
     maintenance/development) were moved to a central location
     (same system with geographically separated functions when
     functions needed close communication—RCRIS).
*    CERCLIS also becoming centralized.
*    You get cross fertilization of experiences between different
     systems when co-located.
*    OERR has benefitted from sharing data when moved to same
     location as other OSWER systems.

DEVELOPMENT OF DUPLICATE SYSTEMS

*    SDC set up to try and avoid this through its process.

EXPOSURE OF AGENCY'S FINANCIAL PAYMENT SYSTEMS TO UNNECESSARY
ACCESS RISKS

*    No financial systems in SDC other than IFMS which is an
     evolving picture.

ORGANIZATIONAL ACCOUNTABILITY

*    Documentation approvals seen at higher levels (OSWER, ow,
     OAR).  Other offices are improving.
*    Need an IRM organization (centralized) budget to support
     Agency.  NDPD is the only unifying force via architecture.
*    Contractor estimation process records changes and effort in
     projects for better estimates in future.
*    Use of IEF (Computer Aided Software Engineering ((CASE)
     tool) helps identify performance measures.
*    Govt. side of SDC providing better guidance for IGCE adding
     product assurance to cost estimation items (product
     assurance mandatory).

PLANNING AND RESOURCES

*    No sign that qualifications & training getting worse (not
     necessarily getting better either).
*    Meetings conducted to educate DOPOs on SDC process.
*    Would like to see more OIRM/IMDA participation in process.
*    SDC has self-discipline with its existing structure - i.e.,
     separate facility, "off-site" meetings, etc. therefore no
     day-to-day direction.
*    SDC setup forces better planning.
*    Certain things  (functions) should be government function.
*    If resources are not available, reduce what is being done or
     reduce requirements to be met..


                               V-20

-------
                                                       APPENDIX V
*    More emphasis on IRM, skills, life-cycle training, support,
     and continue to make sure DOPOs understand contract issues.
*    Need programmatic pressure to make projects better.
*    Need mechanism to address requirements in time period vs.
     violation of good IRM.
*    IRM needs to get "plugged in" to Congress/bills.
*    When Congress does "how to1* legislation, EPA/IRM needs to be
     involved or they should not be so specific.
* -   Need better OMB interface on information collection forms,
     complete with time frames and implications on systems.
*    Need to identify system life but technology mainly the
     driver in system life, plus NDPD support of technology
     (80286 computers are still being supported).
*    EPA needs FTEs to do certain functions such as LAN support
     on-site.

QUALITY ASSURANCE AND DATA PROTECTION

*    As work is done on "inherited" systems (existing systems
     coming into the SDC), documentation improves.  All new
     releases are well documented.
*    Data standards are being used when applicable and adhered
     to.
*    Contractor personnel have signed confidentiality statements.
*    Risk analysis being done more frequently by program offices
     and are also done during development.
*    SDC supporting Agency security program.
*    Considering adding cost for security as mandatory item much
     like what is done for QA.

SYSTEMS DEVELOPMENT LIFE CYCLE

*    Public Water Service Supply (PWSS) effort started with an
     ISP which identified systems to develop first.  The effort
     included Regions and States for requirements.  The use of
     CASE helped to get immediate feedback and "buy in" with "on
     the spot" documentation.  Emphasis was on requirements with
     validation.  Also looked at public and local govt. needs.
     Able to develop system more quickly (esp. through - Rapid
     Application Design (RAD) of subsystem identified in ISP).
     Has gone to pilots in States and getting good feedback.
*    Need more.emphasis on flexibility within policy, procedures,
     and standards.
*    Need to identify what level of mandatory policy can be
     enforced no matter what, ensure policies can be implemented
     (some policies cannot be implemented, some are not
     applicable).
                               V-21

-------
                                                       APPENDIX V
*    Need understanding of policies, standards, and procedures
     plus budget to implement.
*    May be useful to have an O&M SDC as well as a development
     SDC.  Would be difficult to support an SDC at each Region.
*    Need some mechanism to determine if a system should be
     scrapped.

COMMUNICATION

*    Need exposure to policies, procedures and standards through
     training.
*    Keep up pressure (audits).  Enforcement of policies is
     crucial.

Note:     This interview session included key EPA SDC management
          personnel as well as key contractor SDC management
          personnel.                      •            .
                               V-22

-------
                                                       APPENDIX V
                SUMMARY OF HEADQUARTERS INTERVIEW
                       SUBJECT:  8UPERFUND
Problems Identified
       Have had some problems with definitional interpretation.
       The CERCLIS information system interface with the Agency's
       financial system has not been reestablished following
       replacement of the system with IFMS.
       The Office of Information Resources Management does not
       have control over the Agency's IRM program.
       The Agency has not adequately defined commonality between
       information systems.
Root Causes
       The Office of Information Resources Management does not
       enforce IRM policies, procedures, and standards.
       Agency management does not understand the cost of
       information.
       The Office of Information Resources Management does not
       have a clear mission statement.
Recommendations
       Recruit and place technical professionals in the program
       offices.  These professionals need to be EPA FTEs.
       Make available a core staff of IRM professionals  (Agency
       FTEs with technical expertise).
       Provide technical training (e.g., systems management
       training).
       Make end-users responsible for data quality.
       Take steps to monitor data quality.
       The Office of Information Resources Management needs to
       enforce IRM policies, procedures, and standards if they
       develop them.
       System changes need to have buy-in from Headquarters and
       the Regions.  Need to build consensus.
       Resist asking Regions to collect information that they do
       not use or need.
       Need to get data off of the mainframes and in the hands of
       users and managers.
       Need to conduct IRM planning up front.
       Need to link the IRM plan to the budget.
       Control of IRM needs to take place at the program level.
       IRM support services should be provided as part of a
       program-level mission support contract.

                               V-23

-------
                                                       APPENDIX V
Additional Comments
       Superfund information systems provide the information
       necessary to manage the program.
       Many of the core CERCLIS utilities were originally
       developed in the Regions.
       Senior management support is critical.
       Support from the OSWER Information Management Staff has
       been valuable.
       Don't see a need to merge information systems that contain
       unique information.
                              V-24

-------
                                                       APPENDIX V
                SUMMARY OF HEADQUARTERS INTERVIEW
                SUBJECT:  PERFORMANCE MEASUREMENT


The Review Team met with representatives from the OPPE strategic
Planning and Management Division and OARM Financial Management
Division.  The purpose of the meeting was to discuss the Agency's
accomplishments reporting capabilities and its efforts to
implement the "Government Performance and Results Act of 1993."

STARS

Strategic Planning and Management Division is responsible for
running STARS.  STARS is the Agency's official accomplishments
reporting system.  An OPPE official explained that 97% of the
data in STARS system comes from other Agency databases (CERCLIS,
RCRIS, etc.) and that the data relates primarily to "activities .
that the Agency performs."  The system contains a significant
amount of enforcement and Superfund data.

Past administrations have used information reported by the system
(in quarterly reports) to question AA's and RA's about
accomplishments.  However, the immediate previous and current EPA
Deputy Administrators have not used STARS information heavily.
Mr. Habicht was briefed on the STARS information before visiting
a Regional office, but did not "question" Regional management
about information reported in STARS.  OPPE officials believe that
the current low usage relates to the fact that program management
has not yet been fully established or that it does not fit the
management style of the new administration.

STARS II

The STARS II system, which is currently in the conceptual stage,
will get closer to identifying outcome related information
(versus purely "bean counting" accomplishment data).  OPPE
believes this would be an improvement on the current system.
However, whether or not the system will receive outcome
information from Agency systems (as STARS now receives "bean
counting" accomplishment information) is not known at this time.
System planners do not envision the system including many
environmental indicators.  Primary reasons for developing STARS
II are to link resources to accomplishments and to have
representation from all program offices.  Targets/commitments are
set on an annual basis.  Target setting for FY 1994 is going on
now.  Targets are locked at the end of April.
                               V-25

-------
                                                       APPENDIX V
STARS II documentation to date consisted of briefings given  (in
the August tineframe).  This is the latest information available
(and it may be slightly out of date -particularly from the
standpoint of incorporating the NPR results).

PERFORMANCE MEASUREMENT

A case study is currently being performed using the Chesapeake
Bay program.  The purpose of the case study is to identify
obstacles, stumbling blocks, etc.  An additional purpose is to
tie together resources, output, and outcomes.  The problem with
this case study is that State resources are not being accumulated
as part of the project (which means that the measurement of
accomplishments per resource will be overstated).

The Agency anticipates participating at some level of response to
the Government Performance and Results Act of 1993 (6PRA).  At
this time, however, EPA is not participating in any pilot
project.  A GPRA workgroup is drafting a letter to OMB expressing
interest in a GPRA pilot project.  The pilot area should be
identified in the near future.

The STARS II effort would be a performance, measurement project
and is expected to tie into the budget system (link to IFMS
and/or RMIS) and would report GPRA results (including
accomplishments compared to goals).  STARS II is expected to tic ••
planning, budgeting, and accomplishments together.  Information
needs to be tied into measurement as well, but at this point the
Agency is still grappling with conceptual issues.  No software
tools were being used to identify available sources of the data
needed for STARS II at the time of this meeting.  The review tea*
pointed out that it is during this conceptual stage that software
tools are needed for identification of information sources.

While more than 90 percent of STARS data is submitted from
program information systems, OPPE does not perform data quality
work.  However, there is a discrepancy resolution process whereby
the Regional "numbers" and Headquarters "numbers" are printed and
compared.  Differences are formally hashed out at the Assistant
Administrator and Regional Administrator level.  If no agreement
can be reached, the Headquarters number is used.  Discrepancies
are usually the result of timing differences and are negotiated
between the Regional Administrators and Assistant Administrators.
                               V-26

-------
                                                       APPENDIX V
                  SIRMO INTERVIEW SUMMARY No. 1


This SIRMO emphasized that much information is available, but
that a typical manager must usually rely on an intermediary
person to get the information.  The ideal is for managers to have
readily accessible and more user-friendly systems that they can
use directly.  The SIRMO acknowledged Headquarters resource
constraints preclude making systems more user-friendly and
believes national systems provide useful information but more is
needed.  National systems that are designed to answer
Congressional questions rarely meet Regional needs.  There
usually is little or no Regional ownership of the data being fed
into the national systems.

This Region has or is in process of automating many
administrative systems:  training, contracts management, travel,
and procurement.  The SIRMO discussed advances in the Agency: the
goals project and THIS as success stories, and stated that she
believes the Agency is "on the cusp" of beginning to really use
its information well.  Possible obstacles contributing to any
inability to provide basic management information through the
Agency's information systems were discussed.

OBSTACLES

Lack of a distinct program element for IRM in the Regions leads
to inadequate funding for IRM related FTE resources stemming from
the current practice of lumping this requirement with all other
resource management issues.  Outmoded software and the need for
cultural change in which managers would be seen as end users,
able to retrieve information without an intermediary were
described as obstacles.  The Headquarters EPAYS system's
inability to interface with this Region's on-line training
system, requiring Regional staff to re-key training data in batch
to EPAYS, is a lost efficiency.

RECOMMENDATIONS

Establish an Headquarters IRM official organizationally placed
above all Assistant Administrators.

Establish an Agency standard relational database management
system for all applications.

Continue culture change that fosters significant user involvement
in system development projects.


                               V-27

-------
                                                       APPENDIX V
Ensure all system development projects have clear understanding
of customers' and users' needs.  For example, repeat the recent
effort where Regional programmers went to Headquarters to work on
ICMS.

Revise FTP resource acquisition requirements which apply equally
to purchases of diskettes and toner cartridges as well as major
purchases of much greater value.
                              V-28

-------
                                                       APPENDIX V
                  SIRMO INTERVIEW SUMMARY Mo. 2
This SIRMO indicated that generally data  (environmental, permit,
water quality, etc.) is not easy to get at and does not serve
purposes of what users (public/local government) want or need it
for.  By way of example he said that often they dial an "800"
number, get a wrong number, nobody answers it, and when they do
get through it's not what they thought or wanted.  The SEDN
program was a good program and was allocating funds to enhance
State capacity but this was dropped in FY 92 or 93.   A Task
Force, started under Administrator Reilly, found that States do
not have tools to do the job.  A Steering Committee is
implementing the Task Force report, with support from the
National Performance Review and Cabinet status bill for EPA.
Initiatives are integrally involved in data management.
                           X                           ".
Management often relies on staff and often assumes that things
are taken care of, especially when they have good staff.  If you
have good staff (under the SIRMO functions) you tend to devote
less time to the SIRMO function.  He estimated that he spends
about 25% or less on SIRMO duties because he has a very good
staff.  He felt his recent detail to Headquarters was very useful
because it helped to improve communication and relationships with
key Headquarters management.

The SIRMO expressed concern over whether we really protect data
and have the right people handling data.  He felt that additional
work was needed in this area.

Regarding resources, he will get the client (program manager) to
provide people before committing his resources.  After work is
initiated, the client is trained to provide continuity and reduce
need for his resources on a continuing basis.  He felt that
planning is pretty good in the Region.  He sends out call memos
at the beginning of each FY and mid-way through to determine what
IRM support is needed.  There could be benefit from long-term
planning, however.

OBSTACLES

*    The public and local governments are frequently not
     identified as users up front in the process.

*    Lack of the ability for management to quickly communicate
     issues.

*    Not enough attendance by SIRMOs at ARA meetings.
                    i
                              V-29

-------
                                                       APPENDIX V
*    People (program managers) not always willing to commit
     resources for developing systems/applications.

*    There is a tendency to add on to systems and not enhance
     what is there.  Quality data is best when everybody has to
     access to do their job.

RECOMMENDATIONS

*    state Data Capacity - one current effort trying to get
     better communication, with minimum investment, via e-mail
     between States/Regions/Headquarters to more quickly identify
     and respond to issues.

*    Management needs to be more proactive in getting information
     from staff and ensuring that things are done.

*    If you have senior staff substitute in some of the IRM
     meetings, it is critical that you have good communication
     both before and after the meeting.

*    Need assessments to determine how best to handle data/who
     should handle data.

*•    Need to commit/invest time and dollars to get good data.
                              V-30

-------
                                                  APPENDIX VI

        AGENCY PARTICIPANTS IN FOCUS GROUPS AND INTERVIEWS

                           HEADQUARTERS


Office of A^pjnlgtration and Resources Management
     Office of Information Resources Management

Jeff Byron, Chief, Information Systems Management Branch,  Program
     Systems Division
Steve Young, Information Systems Management Branch, Program
     Systems Division
Barbara Jarvis, Program Systems Division, Systems
     Development Center
Margarite Shovlin, Program Systems Division, Systems
     Development Center

     Office of the Comptroller

Debbie Ingram, Financial Management Division

     System Development Center
          SAIC

Charlie Stringfellov, Program Manager, MOSES Contract
Tom Thomason, Assistant Program Manager, OSWER Projects

Office of Enforcement
     Office of Compliance Analysis and Program Operations

Bruce Rothrock, Chief, Information Management Branch

Office of Policy. Planning and Evaluation
     Office of Strategic Planning and Environmental Data

Phil Ross, Director, Environmental Statistics and
     Information Division
Chapman Gleason, National Environmental Statistics Branch
Sue Priftis, Strategic Planning and Management Division

Office of Water

Michelle Killer, Director, Communications and Information
     Management Staff
Robert King, Office of Wetlands, Oceans and Watersheds
Phil Lindenstruth, Office of Wetlands, Oceans and Watersheds
Dela Ng, Office of Wastewater Enforcement and Compliance
Larry Weiner, Office of Ground Water and Drinking Water
                               VI-l

-------
                                                  APPENDIX VI

Office of Solid Waste and Emergency Response

Thomas Sheckells, Director, Office of Program Management, Office
   of Emergency and Remedial Response
Michael Cullen, Director, Management Systems Staff, Office of
     Program Management
Linda Boornazian, Deputy Director, CERCLA Enforcement
     Division, Office of Waste Programs Enforcement
Myra Galbreath, Chief, Information Management Branch, Office of
     Solid Waste

Office of Air and Radiation
     Office of Program Management Operations

Kelly N. Spencer, Acting Director, Resource Management Staff,
      Office of Program Management Operations
Reginald Slade, Resource Management Staff

     Office of Air Quality Planning and Standards
     Research Triangle Park. NC

Robert G. Kellam, Acting Director, Technical Support Division
John Bosch, Chief, National Air Data Branch
David Mobley, Chief, Emission Inventory Branch
Andrea Kelsey, National Air Data Branch

Office of Prevention. Pesticides and Toxic Substances

George Bonina, Deputy Director, Information Management Division
Ruby Boyd, Information Management Division
Thomas Hooven, Office of Program Management Operations
Jim Willis, Deputy Director, Environmental Assistance Division

Office of Research and Development

Allen Johnson, Information Systems Staff
Clifford Moore, Chief, Information Systems- Staff

Focus Group Facilitator

Steven Smith
                              VI-2

-------
                                                  APPENDIX VI

                             REGION 2


Office of Policy and Management

Robert A. Messina, Chief, Information Management Branch
Barbara J. Pastalove, Chief, Planning and Evaluation Branch
Jo-Ann Velez, Financial Management Branch

Office of Regional Counsel

Janice Dudek

Air and Waste Management Division

Helen S. Beggun, Deputy Director

Water Management Division

Robert Vaughn, Chief, Water Permits and Compliance Branch

Emergency and Remedial Response Division  •

William J. McCabe, Deputy Director for New York/Caribbean
     Programs
Doug R. Garbarini, Deputy Director for New Jersey Superfund
     Branch I
Vincent J. Pitruzzello, Chief, Program Support Branch
Richard C. Salkie, Associate Director for Removal and Emergency
     Preparedness Programs

Environmental Services Division

Dr. Barbara M. Metzger, Director

National Enforcement Investigations Center

Richard Herman, Office of Criminal Investigations

Focus Group Facilitators

Allan Sommerman
Peter Brandt
                               VI-3

-------
                                                  APPENDIX VI

                             REGION 4


Office of Policy and Management

Donald J. Guinyard, Assistant Regional Administrator
     for Policy and Management
William. A. Waldrop, Jr., Deputy Assistant Regional Administrator
     for Policy and Management
Jack Sweeney, Chief, Information Management Branch
Randall Davis, Information Management Branch

Waste Management Division

Joseph Franzmathes, Director
Richard Green, Associate Director for Superfund and Emergency
     Response
James Kutzman, Associate Director for RCRA and Federal Facilities
Elmer Akin, Director, Office of Health Assessment
Franklin Hill, Director, Office of Management Support
Doug Lair, Chief, Emergency Response and Removal Branch
Robert Jourdan III, Chief, North Superfund Remedial Branch
Douglas Mundrick, Chief, South Superfund Remedial Branch
Doug Murdock, South Superfund Remedial Branch
Jon Johnson, Chief, Federal Facilities Branch
H. Kirk Lucius, Chief, Waste Programs Branch
Jim Miller, Waste Programs Branch
Eddie Wright, Waste Management Division

Water Management Division

James Scarbrough, Chief, Water Permits and Enforcement Branch

Air. Pesticides and Toxics Management Division

Chester Wakamo, Director

Focus Group Facilitators

Lila Koroma
Annie Godfrey
                               VI-4

-------
                                                  APPENDIX VI
                             REGION 9

Office of Policy and Management
Nora L. McGee, Assistant Regional Administrator
  for Policy and Management
David S. Mowday, Deputy Assistant Regional Administrator
  for Policy and Management
David C. Henderson, Chief, Information Management Branch
Angle commisso, Information Management Branch
Hazardous Waste Management Division
David B. Jones, Chief, Remedial Action Branch
Michael T. Feeley, Chief, Permits and Solid Waste Branch
Betsy Curnow, Chief, Case Development Section
Tom McMenamin, Superfund Program Management Team
Water Management Division
Carey Houk, Data Base Administrator
Jon Merkle, Senior Environmental Scientist
Air and Toxics Management Division
Ed Snyder, Acting Chief, Compliance and Oversight Section
Focus Group Facilitator
Becky Tudisco
                              VI-5

-------
                                                  APPENDIX VII
             )IG REPORTS.  GAP
               REPORTS.  AND CONGRESSIONAL TESTIMONY
Note: The codes following each document date  (e.g., OIG-A-V1)
have been assigned uniquely to each report and testimony,  and
provide a cross-reference to individual recommendations  in
Appendix IX, column entitled "Reference/Page."


EPA INSPECTOR GENERAL REPORTS

1.   Report on Special Review - ADCR IBM Mainframe Password
     Exposure, Report No. E1NMGO-15-0023-0400003, dated
     December 21, 1989.  (OIG-A-V1)

2.   Report on CERCLIS Reporting, Report No. E1SFF9-15-0023-
     0100187, dated March 12, 1990. (OIG-B-V1)

3.   Review of the Fiscal Year 1988 Superfund Report to  Congress,
     Report No. E1SFF9-11-0015-0100227, dated March 28,  1990.
     (OIG-C-V1)

4..   Flash Audit Report - Disclosure of User Passwords on  EPA' s
     IBM 3090 Computer Mainframes, dated May 7, 1990.  (OIG-D-V1)

5.   Report on Special Review - CERCLIS Post-Implementation
     Evaluation, Report No. E1SFGO-15-0020-0400019, dated
     June 14, 1990. (OIG-E-V1)

6.   Report on Special Review - Hotline Complaint Concerning the
     Office of Research and Development's Modeling and Monitoring
     Tracking System,  Report No. E1NBGO-15-0038-0400037, dated
     September 24, 1990. (OIG-F-V1)

7.   Flash Audit Report - Vulnerability of Sensitive Payroll and
     Personnel Files on the National Computer Center (NCC) IBM
     3090 Computer System,  dated September 26, 1990. (OIG-G-V1)

8.   Review of the Fiscal Year 1989 Superfund Report to Congress,
     Report No. E1SFFO-11-0018-1100026, dated October 18,  1990.
     (OIG-H-V1)

9.   Integrated Financial Management System: Managing
     Implementation of the New Accounting System, Report No.
     E1AMFO-11-0029-1100153, dated March 29, 1991. (OIG-I-V1)

10.  Significant Savings Possible by Increasing IBM 3090 Computer
     Operations Efficiency, Report No. E1NMBO-15-0021-1100152,
     dated March 29, 1991.   (OIG-J-.V1)

                              VII-1.

-------
                                                  APPENDIX VII

11.  Improvements Needed in EPA's Resource Access Control
     Facility (RACF) Security Software, Report No. ElNMBO-15-
     0027-1100151, dated March 29, 1991.  (OIG-K-V1)

12.  Annual Superfund Report to the Congress for Fiscal 1990,
     Report No.  P1SFFO-11-0032-1100385, dated September 16, 1991.
     (OIG-L-V1)

13.  Special Review of EPA's Major Information Systems, Report
     No. E1RMG1-15-0041-1400061,. dated September 30, 1991.  (OIG-
     M-V1)

14.  Special Review, on Follow-up of CERCLIS Reporting and Post-
     Implementation, Report No. E1SFG1-15-5001-2400027, dated
     March 27, 1992. (OIG-N-V1)

15.  CONTRACT MANAGEMENT: EPA Needs to Strengthen the Acquisition
     Process for ADP Support Services Contracts, Report No.
     E1NMF1-15-0032-2100300, dated March 31, 1992. (OIG-0-V2)

16.  Flash Report on Mainframe Access Control Weaknesses at the
     National Computer Center, dated April 17, 1992. (OIG-P-V2)

17..  SOFTWARE INTEGRITY: EPA Needs to Strengthen General Controls
     over System Software,  Report No. E1NMF1-15-0055-2100591,
     dated September 22, 1992. (OIG-Q-V2)

18.  COMPUTER SYSTEMS INTEGRITY: EPA Must Fully Address
     Longstanding Information Resources Management Problems,
     Report No.  E1NMF1-15-0032-2100641, dated September 28, 1992.
     (OIG-R-V2)

19.  Annual Superfund Report to the Congress for Fiscal 1991,
     Report No.  P1SFF1-11-0026-2100660, dated September 30, 1992.
     (OIG-S-V2)

20.  Special Review of EDP Internal Controls for Selected
     Pesticide Revolving Funds' Information Systems,  Report No.
     E1EPP2-15-7001-3400043, dated March 31, 1993. (OIG-T-V2)

21.  Special Review of Allegations Regarding Copyright
     Infringement Within the. Office of Communications,  Education,
     and Public  Affairs, Report No. E6AMG3-15-0071-3400042, dated
     March 31, 1993. (OIG-U-V2)

22.  Consolidated Report Regarding Fiscal 1992 CERCLIS Data,
     Report No.  E1SFF3-11-0016-3100392, dated September 29,
     1993). (OIG-V-V2)
                              VII-2

-------
                                                  APPENDIX VII  •

GENERAL ACCOUNTING OFFICE REPORTS

23.  SUPERFUND: A More Vigorous and Better Managed Enforcement
     Program Is Needed, Report No. GAO/RCED-90-22, dated
     December 14, 1989. (GAO-A-LIB)

24.  HAZARDOUS WASTE: EPA's Generation and Management Data Need
     Further Improvement,  Report No. GAO/PEMD-90-3, dated
     February 9, 1990. (GAO-K-V1)

25.  FINANCIAL AUDIT: EPA's Financial Statements for Fiscal Years
     1988 and 1987, Report No. GAO/AFMD-90-20, dated March 16,
     1990. (GAO-B-LIB)

26.  PUBLIC ACCESS: Two Case Studies of Federal Electronic
     Dissemination, Report No. GAO/IMTEC-90-44BR, dated May 14,
     1990. (GAO-B-V1)

27.  GEOGRAPHIC INFORMATION SYSTEMS: Status at Selected Agencies,
     Report No. GAO/IMTEC-90-74FS, dated August 1, 1990. (GAO-C-
     VI)

28.  DISINFECTANTS: EPA Lacks Assurance They Work, Report No.
     GAO/RCED-90-139, dated August 30, 1990.  (GAO-L-V2)

29.  DISINFECTANTS: Concerns Over the Integrity of EPA's Data
     Bases, Report No. GAO/RCED-90-232, dated September 21, 1990.
     (GAO-C-LIB)

30.  PESTICIDES: EPA Could Do More to Minimize Groundwater
     Contamination, Report No. GAO/RCED-91-75, dated April 29,
     1991. (GAO-D-LIB)

31.  HAZARDOUS WASTE: Data Management Problems Delay EPA's
     Assessment of Minimization Efforts, Report No. GAO/RCED-91-
     131, dated June 13,  1991. (GAO-E-LIB)

32.  ENVIRONMENTAL ENFORCEMENT: Penalties May Not Recover
     Economic Benefits Gained by Violators, Report No. GAO/RCED-
     91-166,  dated June 17,  1991. (GAO-F-LIB)

33.  TOXIC .CHEMICALS: EPA's Toxic Release Inventory Is Useful but
     Can Be Improved, Report No.  GAO/RCED-91-121, dated June 27,
     1991. (GAO-M-V2)

34.  WASTE MINIMIZATION:  EPA Data Are Severely Flawed, Report No.
     GAO/PEMD-91-21, dated August 5, 1991. (GAO-G-LIB)
                              VII-3

-------
                                                  APPENDIX VII

35.  PESTICIDES: Better Data Can Improve the Usefulness of EPA's
     Benefit Assessments, Report No. GAO/RCED-92-32, dated
     December 31, 1991.  (GAO-N-V2)

36.  FOOD SAFETY: USDA Data Program Not Supporting Critical
     Pesticide Decisions, Report No. GAO/IMTEC-92-11, dated
     January 31, 1992. (GAO-H-LIB)

37.  INFORMATION RESOURCES: Summary of Federal Agencies'
     Information Resources Management Problems, Report No.
     GAO/IMTEC-92-13FS, dated February 13, 1992.  (GAO-D-V1)

38.  ASBESTOS REMOVAL AND DISPOSAL: EPA Needs to  Improve
     Compliance With Its Regulations, Report No.  GAO/RCED-92-83,
     dated February 25, 1992. (GAO-I-LIB)

39.  WASTE MINIMIZATION:  Major Problems of Data Reliability and
     Validity Identified, Report No. GAO/PEMD-92-16, dated
     March 23, 1992. (GAO-J-LIB)

40.  ENVIRONMENTAL ENFORCEMENT:  EPA Needs a Better Strategy to
     Manage Its Cross-Media Information, Report No. GAO/IMTEC-92-
     14, dated April 2, 1992. (GAO-E-V1)

41.  ENVIRONMENTAL ENFORCEMENT:  Alternative Enforcement
     Organizations for EPA, Report No. GAO/RCED-92-107, dated
     April 14, 1992. (GAO-0-V2)

42.  SUPERFUND: Problems With the Completeness and Consistency of'
     Site Cleanup Plans,  Report No. GAO/RCED-92-138, dated
     May 18, 1992.  (GAO-P-V2)

43.  WATER POLLUTION MONITORING: EPA's Permit Compliance System
     Could Be Used More Effectively, Report No. GAO/IMTEC-92-
     58BR, dated June 22, 1992.  (GAO-F-V1)

44.  Perceived Barriers to Effective Information Resources
     Management: Results of GAO Panel Discussions, Report No.
     GAO/IMTEC-92-67, dated September 1992. (GAO-G-V1)

45.  PESTICIDES: Information Systems Improvements Essential for
     EPA's Reregistration Efforts, Report No.  GAO/IMTEC-93-5,
     dated November 23, 1992. (GAO-H-V1)

46.  Information Management and Technology Issues, Transition
     Series, Report No. GAO/OCG-93-5TR, dated December 1992.
     (GAO-I-VI)

47.  Environmental Protection Issues, Transition Series, Report
     No. GAO/OCG-93-16TR, dated December 1992. (GAO-J-V1)

                              VII-4

-------
                                                  APPENDIX VII

48.  Superfund Program Management, High-Risk Series, Report No.
     GAO/HR-93-10, dated December 1992.  (GAO-A-V1)

49.  HAZARDOUS WASTE: Much Work Remains to Accelerate Facility
     Cleanups, Report No. GAO/RCED-93-15, dated January 19, 1993.
     (GAO-Q-V2)

50.  ENVIRONMENTAL ENFORCEMENT: EPA Cannot Ensure the Accuracy of
    • Self-Reported Compliance Monitoring Data, Report No.
     GAO/RCED-93-21, dated March 31, 1993. (GAO-R-V2)

51.  SUPERFUND: EPA Actions Could Have Minimized Program
     Management Costs, Report No. GAO/RCED-93-136, dated
     June 1993. (GAO-S-V2)
                                           EPA MANAGEMENT REPORTS

52.   National Archives and Records Administration - Records
     Management in the Environmental Protection Agency, dated
     February 19, 1992.  (EPA-A)

53.   EPA IRM Compliance Strategy Task Group Report, dated
     October 1, 1992.  (EPA-B)

54.   Analysis of Materiality of Weaknesses in the United States
     Environmental Protection Agency's Information Resources
     Management Program, dated December 14, 1992 (prepared by
     Federal Sources, Inc.), (EPA-C)

55.   Written comments by Paul Wohlleben, Acting Director, Office
     of Information Resources Management, on strengthening EPA's
     IRM program, dated May 19, 1993. (EPA-D)

56.   Financial Management Status Report and .Five-Year Plan, dated
     July 30, 1993.  (EPA-E)
CONGRESSIONAL TESTIMONY

57.  SUPERFUND:  Current Progress and Issues Needing Further
     Attention,  GAO testimony before the Subcommittee on
     Oversight,  Committee on Ways and Means, U.S. House of
     Representatives, Document No. GAO/T-RCED-92-56, dated
     June 11, 1992.  (T-GAO-A)

58.  SUPERFUND:  Problems With the Completeness and Consistency of
     Site Cleanup Plans Two, GAO testimony before the
     Subcommittee on Investigations and Oversight, Committee on
     Public Works and Transportation, U.S. House of

                              VII-5

-------
                                                  APPENDIX VII

     Representatives, Document No. GAO/T-RCED-92-70, dated
     June 30, 1992.  (T-GAO-B)

59.  SUPERFUND: EPA Needs to Better Focus Cleanup Technology
     Development, GAO testimony before the Subcommittee on
     Investigations and Oversight, Committee on Public Works and
     Transportation, U.S. House of Representatives, Document No.
     GAO/T-RCED-92-92, dated September 15, 1992. (T-GAO-C)

60.  Nomination of Carol M. Browner, Hearing before the Committee
     on Environment and Public Works, U.S. Senate,  on June 11,
     1993. (T-EPA-D)

61.  Why EPA Should be a Cabinet Department, Statement by
     Administrator Carol Browner to the Committee on Governmental
     Affairs, U.S. Senate, on February 18, 1993. (T-EPA-E)

62.  Creation of a Department of the Environment, GAO testimony
     before the Committee on Governmental Affairs,  U.S. Senate,
     Document No. GAO/T-RCED-93-6, dated February 18, 1993. (T-
     GAO-F)

63.  Testimony of Administrator Carol M. Browner before the
     Subcommittee on Oversight and Investigations,  Committee on
     Energy and Commerce, U.S. House of Representatives, on
     March.10, 1993.  (T-EPA-G)

64.  Testimony of EPA Inspector General John C. Martin before the
     Subcommittee on Legislation and National Security, and the  '•
     Subcommittee on Environment, Energy and Natural Resources of
     the Committee on Government Operations, U.S. House of
     Representatives, on March 29, 1993. (T-OIG-H)

65.  Management Issues Facing the Environmental Protection
     Agency,  GAO testimony before the Subcommittee on Legislation
     and National Security, and the Subcommittee on Environment,
     Energy and Natural Resources of the Committee on Government
     Operations, U.S. House of Representatives, Document No.
     GAO/T-RCED-93-26, dated March 29, 1993. (T-GAO-I)

66.  ENVIRONMENTAL PROTECTION: EPA's Actions to Improve
     Longstanding Information Management Weaknesses, GAO
     testimony before the Subcommittee on Legislation and
     National Security, and the Subcommittee on Environment,
     Energy and Natural Resources of the Committee on Government
     Operations, U.S. House of Representatives, Document No.
     GAO/T-IMTEC-93-4, dated March 29, 1993. (T-GAO-J)
                              VII-6

-------
                                                  APPENDIX  VII

67.  SUPERFUND: Progress, Problems, and Reauthorization  Issues,
     GAO testimony before the Subcommittee on Transportation  and
     Hazardous Materials, Committee on Energy and Commerce, U.S.
     House of Representatives, Document No. GAO/T-RCED-93-27,
     dated April 21, 1993.  (T-GAO-K)

68.  Testimony of EPA Inspector General John C. Martin before the
     Subcommittee on Legislation and National Security,  and the
     Subcommittee on Environment, Energy and Natural Resources of
     the Committee on Government Operations, U.S. House  of
     Representatives, on May 6, 1993. (T-OIG-L)

69.  Creation of a Department of Environmental Protection,  GAO
     testimony before the Subcommittee on Legislation and
     National Security and the Subcommittee on Environment,
     Energy, and Natural Resources, Committee on Government
     Operations, U.S. House of Representatives, Document No.
     GAO/T-RCED-93-39, dated May 6, 1993.  (T-GAO-M)

70.  Testimony of Administrator Carol M. Browner before  the
     Subcommittee on Superfund, Recycling and Solid Waste
     Management of the Committee on Environment and Public  Works,
     U.S. Senate, on May 12, 1993.  (T-EPA-N)

71.  Testimony of EPA Inspector General John C.. Martin before the
     Subcommittee on Superfund, Recycling and Solid Waste
     Management of the Committee on Environment and Public  Works,
     U.S. Senate, on June 10, 1993. (T-OIG-0)

72.  SUPERFUND: EPA Action Could Have Minimized Program
     Management Costs, GAO testimony before the Subcommittee on
     Superfund, Recycling, and Solid Waste Management, Committee
     on Environment and Public Works,  U.S. Senate, Document No.
     GAO/T-RCED-93-50, dated June 10,  1993.  (T-GAO-P)

73.  Testimony of EPA Inspector General John C. Martin before the
     Committee on Governmental Affairs,  U..S. Senate, on June 22,
     1993.  (T-OIG-Q)

74.  SUPERFUND: Little Use Made of Techniques to Reduce Legal
     Expenses, GAO testimony before the Subcommittee on
     Transportation and Hazardous Material, Committee on Energy
     and Commerce, U.S.  House of Representatives, Document  No.
     GAO/T-RCED-93-60, dated June 30,  1993.  (T-GAO-R)

75.  ENVIRONMENTAL PROTECTION:- EPA Faces Formidable Challenges
     Managing Water Quality Data, GAO testimony before the  Sub-
     committee on Clean Water, Fisheries and Wildlife, Committee
     on Environment and Public Works,  U.S. Senate, Document No.
     GAO/T-AIMD-93-2, dated Augus.t 5,  1993.  (T-GAO-S)

                              VII-7

-------
                                                  APPENDIX VIII
                     APPROACH AND METHODOLOGY
I.   Project Plan

     The initial draft of the project plan was developed by the
OIG team members in September 1993.  OIRM and OSWER team members
were selected and joined the team in October 1993.  When the
entire team was assembled, the project plan was reviewed, changed
and agreed to by all team members.

     The project plan identified the review objectives,
background, scope and methodology, location of the project work,
preliminary project work, and milestones and associated time
frames.  The following description of the approach and
methodology documents in more detail the methodology used.

II.  Initial Field Work

     This phase was completed by OIG staff prior to assembling
the entire team.

     A.   Collection of Previous Reports

          Copies of prior OIG and GAO reports addressing the IRM
     and Superfund programs, and OIG and GAO position papers and
     interview summaries for ongoing reviews were obtained and
     reviewed.  Copies of any relevant documents, reports, or
     studies (e.g., task force reports, contractor studies,
     Superfund reauthorization studies, testimonies,  etc.)
     directed at resolving problems with the Agency's IRM
     program, especially within Superfund, were also obtained and
     reviewed.  Throughout the rest of the review, as additional
     documentation was identified it was reviewed and the
     references were added to the master list.   A complete list
     of documents used in performing this review is contained in
     Appendix VII.

          The scope of the review was limited to audit reports
     and studies completed since fiscal 1990 so that the most
     current issues would be addressed.

     B.   Consolidation and Analysis of Previous Report
          Recommendations

          OIG staff analyzed the assembled documents and listed
     all the recommendations to identify root causes,  problems,
     issues and concerns, which were then prioritized and

                             VIII-1

-------
                                                  APPENDIX VIII
     categorized into subject areas.  This effort resulted in an
     initial listing of 7 categories.  Five statements of problem
     issues and concerns were developed, and 23 root causes were
     identified and distributed among the 7 categories.  From
     this analysis a two-page summary entitled "Previously
     Identified Root Causes for IRM Problems" was developed.

          A spreadsheet of all previous recommendations was
     developed based on the assembled documentation.  These
     previous recommendations were then grouped by the identified
     root causes.  The spreadsheet further identified the office
     responsible for implementing each recommendation, the Agency
     response, any DIG comments, a document reference and page,
     additional or associated causes, and a priority (high,
     medium, or low).  The recommendations were initially
   .  prioritized by consideration of the relationship to the root
     cause(s), issue/concern, and mission; breadth of impact
     (Agencywide vs. system-specific); relationship to Superfund;
     whether it was a recurring issue (severity of the problem
     area); currentness of the report; complexity of the
     recommendation; potential cost; and timing of scheduled
     implementation..

III.  Previous Recommendations' Priority Determination and
     Refinement of Initial Field Work Results

     Once the joint DIG/Agency review team was assembled, it
reviewed, revised and approved the documents prepared to that
point.  Where possible, additional documents were identified and
obtained.  For example, Superfund officials initiated a
significant data collection effort during the summer and related
activities were continuing in the Superfund program in
preparation for Congressional reauthorization.  The review team
collected and analyzed relevant data for inclusion in the review
process.

     A second review of the listing of problem issues/concerns
and root causes resulted in increasing the problem issues/
concerns statements to six and redistributing root causes among
five rather than seven categories.  The two page summary,
"Previously Identified Root Causes for IRM Problems," was
revised.

     Finally, the priorities assigned in the spreadsheet of
recommendations were reviewed.  A refined ranking was based on:
the relationship to root cause(s), issues/concerns, and mission;
the breadth of the recommendation's impact (addressing a broad
IRM problem area); the relationship to Superfund; whether the

                              VIII-2

-------
                                                  APPENDIX VIII
issue is a single or recurring issue; and the currency of the
report.

IV.  Development of Legislative Issues

     After initial analysis and focus group sessions, high  level
IBM-related legislative issues were developed by the team.  These
issues were relevant to Superfund reauthorization and/or
legislation to elevate the Agency to Departmental status.
Throughout the project, periodic briefings were given to key
Congressional staff, which included these issues.

V.   Identification of and Notification to Participants

     Headquarters and Regional IRM and program senior staff were
invited to participate in Total Quality Management  (TQM)-style
focus groups and interviews to validate the problems, root
causes, and solutions identified by the review team analysis.

     Our goal was to obtain input from a cross-section of the
Agency and focus as much as possible on Superfund, as
specifically identified in the Congressional request.  Superfund
emphasis provided the framework for selection of Regional offices
and led to inclusion of a separate Headquarters interview with
Superfund system managers.

     Selection of Regions was based on Superfund budget
information, the nature and variability of- Superfund work in the
Regions, review team size, time and travel constraints, and
participant impact considerations.  Three Regional offices were
selected for focus group sessions: Region 2 (New York, NY),
Region 4 (Atlanta, GA), and Region 9 (San Francisco, CA).

     Headquarters organizational representation focused on
obtaining a cross-section of the Agency programs: (1) SIRMOs from
all program offices, including selected systems managers
recommended by the SIRMOs, were identified to take part in a
focus group session to provide feedback from a programmatic
perspective; (2) inclusion of OPPE was designed to address
performance measurement and Agency strategic planning; (3) a
"Data Integration" session was designed to obtain feedback on
specific data integration issues; (4)  the Systems Development
Center (SDC) session was selected for feedback on system
development life cycle issues related to major information
systems, with a focus on the Superfund program area; and (5) a
separate teleconference with Air program officials was held
because the AIRS major system program component is located in
Research Triangle Park, NC.

                             VIII-3

-------
                                             APPENDIX VIII
A.   Identification of Participants

     Identification of individual participants was based on
the criteria of obtaining a management cross-section of
programs, focusing on major IRM initiatives and automated
systems, and talking to as many SIRMOs as possible.  The
overriding constraints were keeping focus group sessions to
a workable size of 12 to 15 people, and keeping interviews
to 6 people or less (optimally 2-3).  These group sizes were
suggested optimums based on individual team member
experience and discussions with experienced focus group
facilitators.

     All SIRMOs in the selected Regions were interviewed
separately to reduce impacts on their time.  To obtain
management perspective and programmatic overview, the focus
groups were comprised of Regional Division Directors/Deputy
Directors and Branch Chiefs.  In addition, the IRM Branch
Chiefs and other knowledgeable staff from each "Region were
invited when possible.  Participation was weighted toward
the Superfund program.

     All Headquarters SIRMOs were invited to participate in
a focus group, along with a major information system manager
from their office.  However, the SIRMOs from the Office of
Enforcement participated in a separate interview session,
and the SIRMO from the Office of Air and Radiation attended
a teleconference.

     OPPE participation included individuals involved in
performance measurement and the STARS system.  The "Data
Integration" session participants were based on leadership
in the two largest data integration initiatives,
Envirofacts/Gateway (OIRM) and IDEA (OE).  The Superfund
session participants were selected from CERCLIS system/
program managers.  Finally, the SDC interview participants
were selected from both EPA and the contractor.

B.   Notification and Refinement of Participants

     In advance of the meetings, each participant was
notified in writing and received a standard information
package, including the summary of previously identified root
causes for IRM problems,'to help them prepare to
participate.  Each Assistant Regional Administrator (ARA)
(who function as the Regional SIRMOs), Headquarters SIRMO,
and "special group" participant was contacted by the project
manager or a team member to=confirm participation.  We

                        VIII-4

-------
                                                  APPENDIX VIII
     encouraged the ARAs to review the participant list to ensure
     the participants were the best representatives of the
     Region's IRM and Superfund efforts and to replace invitees
     where appropriate.

VI.  Conduct of Interviews and Focus Groups and Verification of
     Feedback

     A.   Planning and Preparation

          Team member participation in focus sessions,
     interviews, and teleconferences was based on team size and
     individual* team member schedules.  Each team member was
     assigned to a "sub-team11 consisting of one OIG staff and one
     Agency staff member.  Each sub-team was responsible for
     setting up and attending one Regional focus group.
     Attendance at the Headquarters focus group and interviews
     was determined primarily by availability during the
     scheduled time (some sessions were held concurrently) and
     expertise in the subject matter.  At least two team members
     attended each session.

          Team members attended each focus group session to
     provide introduction and background, answer questions, and
     clarify and record results.  During the interviews and
     teleconferences,  team members also asked questions to ensure
     all areas of concern were addressed.

          The focus group sessions were split into two half-day
     sessions, based on discussions with experienced facilitators
     who indicated that was the minimum time necessary to arrive
     at meaningful results while minimizing impacts on management
     resources.

          The list of "Previously Identified Root Causes for IRM
     Problems" provided guidance to the focus group facilitators
     as a basis for obtaining feedback.  The sessions were to
     verify the problem areas and root causes, and allowed the
     participants to reject them and articulate their own
     perceived problems and associated root causes.  Finally,  the
     participants were asked to identify solutions for the root
     causes.  This approach was discussed with each facilitator
     and was used throughout the focus group sessions.

     B.   Conduct of Focus Groups, Interviews and Teleconferences

          Review team members and facilitators discussed and
     planned all sessions in advance.  All sessions used the

                             VIII-5

-------
                                             APPENDIX VIII
notification package summarizing previously identified root
causes for IRM problems as the basis for obtaining feedback.

    The focus group sessions at the Regional offices and
Headquarters were moderately structured.  Although the
sessions were conducted under the same framework, some
flexibility was allowed based on determinations of whether
the sessions were progressing enough to elicit sufficient
feedback.  One focus group had the benefit of two
facilitators, which allowed use of smaller groups to address
more root causes and solutions.  In that focus group, the
results of the small groups were always discussed and
verified in the larger group.  The other focus groups had
one facilitator and the process was conducted in one large
group.

     The first focus group session was evaluated both during
and subsequent to its conclusion to provide feedback on
positive and negative aspects of the process for subsequent
facilitated sessions.  All focus group sessions were
evaluated throughout, to continually improve the process and
ensure that sufficient feedback was obtained.  Although
minor details of methods may have varied between facilitated
sessions (based on facilitators' individual experiences and
training), all sessions were conducted in a generally
consistent framework as follows.

      After discussing the previously identified root
causes, each of the problems was either verified as relevant
to the participants or rejected.  The participants were then
invited to identify additional problems.  These problems
were prioritized by multi-level voting.  There was usually a
distinct break point in voting scores that determined the
highest priority problems.

     Addressing the problems in priority order, the
participants identified root causes for each,  once .all the
root causes were identified, they were likewise prioritized
by multi-level voting.  The highest priority root causes
were further discussed to determine solutions.  The choice
of where to break the list of high priority root causes was
discussed between the review team members and the
facilitator(s) to determine what was achievable in the
remaining time.  At the end of the session the results were
summarized.

1     Interviews and teleconferences were shorter and
consisted of a more free-form flow and expression of

                        VIII-6

-------
                                                  APPENDIX VIII
     problems, root causes and solutions.  Team members
     functioned to clarify statements, ask questions to maintain
     the pace of the interview, and address specific areas of
     concern.  Although useful feedback was obtained, this method
     made analysis of results more difficult, and relative
     priorities of problems, root causes and associated solutions
     could not be determined.

     C.   Verification of Feedback

          All session results were typed and returned to the
     individual participants for verification.  Participants were
     given at least one week to respond with additions or
     corrections.  They were not required to respond if no
     additions or errors were noted.

VII. Summary and Synthesis of Results

     The review team evaluated the results of the focus groups.
Comprehensive assessments of the previously identified root cause
subject area headings were completed, and all results were
organized into additional or revised subject areas as identified
in the focus groups.  Also, the analyses determined whether these
subject areas were valid and logical for categorizing problems
and solutions.  Additional subject areas were identified, such as
the area of data management.

     The logical progression of the analysis led to consolidating
the problems, root causes, and solutions statements from all
sessions.  Once consolidated statements were developed, the
problem statements were matched to root causes.

     The consolidated lists of solution statements were ranked
and placed into clusters of related solutions.  The review team
ensured consistency and integrity in the report by tracing back
all the solution statements to their respective root causes,  and
the root causes to their respective problem areas.  In this
manner, the team ensured that all. root causes were addressed by
the solutions, and that the solutions were fully responsive to
the root causes.  As a final analysis, the review team identified
the solutions that cut across multiple problem areas, and used
this analysis to assist in refining the presentation of the
recommendations in the report.

VIII.Follow-up on Previous Recommendations

     Using the priority list of previous recommendations, members
of the Agency's IRM community were asked to provide status

                              VIII-7

-------
                                                  APPENDIX VIII
updates for the high priority recommendations.  These members of
the IBM community were identified based on the responsible
offices for each recommendation, identification in automated
reporting and tracking systems, and the review team's subjective
judgment of which individuals within the offices would have
direct awareness of the current implementation status.

     Existing audit reporting and tracking systems were used
where possible to generate the most current reports, and the
scope of the request for status information was kept tightly
focused.  When automated tracking reports were not sufficient or
had not been recently updated, the respondents were contacted and
asked to provide:

     r, STATUS (is the recommendation completed, in progress, or
          not started?),
     • DATE (date completed or target date for completion), and
     • COMMENTS (a few sentences describing the effort and/or
          upcoming milestones).           •

     Because of the breadth of prior recommendations, the request
for status information went to many organizations.  These
included OPPTS,  OSWER, OE, Office of Acquisition Management,
Office of the Comptroller, NDPD, and all divisions and staffs of
OIRM.  In most cases, electronic mail was used to convey the
initial request, with follow-up for clarification by phone, fax, .'
or face-to-face discussion.

    The review team formatted all responses similarly, and
arranged them into the matrix contained in Appendix IX.  Where
timing and review team resources allowed, there was additional
follow-up to obtain missing dates and to resolve any confusion
about particular comments.  Finally, responses were sent back to
the respondents for verification.
                             VIII-8

-------
                                                       APPENDIX IX
                  IMPLEMENTATION STATUS OF PRIOR
                  HIGH-PRIORITY RECOMMENDATIONS
BACKGROUND:   This  appendix  contains  information  gathered from
selected members of the Agency's IRM community, chosen because of
their personal,  hands-on knowledge  of EPA's  responses  to prior
audit recommendations.   The responses are  based  on the official
status  of  corrective actions,  as  tracked  in  various management
audit tracking  systems,  and  include additional  details obtained
directly  from those  who implemented  the recommendations.   The
responses,  tend  to be  more  detailed  and  specific than  those
available  through  the   formal  audit  implementation  tracking
channels.

The  recommendations  highlighted  in this appendix  come from  a
variety of OIG, GAO, and Agency reports issued since 1990.  These
high-priority  recommendations  are  a  small  subset  of  all  the
recommendations contained in the documents listed  in Appendix VII.
These particular recommendations were prioritized  for  followup
because the  review team judged them to  be  especially important.
Further details about how the review team  established the relative
priorities of prior recommendations are provided "in Appendix VIII.'-
                               IX-1

-------
               Audit Recommendation

    Notify Hie mar caimunily of the new NDPD IBM
    Mainframe Security Policy which specifically addresses
    generic User-Ids (SYSTEM), shared User-Id, User
    Support authentication issues, and several production
    control security issues.

    Source program and load modules will reside In a
    centralized Horary.

    Sgnofffrom the EPA program office (report owner)
    will 6e required for all report specifications, reports
    library documentation, and test/sample reports.

    Sgnofffrom the EPA program office (report owner)
    will be required fat all report modifications prior
    to reinstatement to the production reports'menu.
M  Any change* to source code will be recorded within
.1   the program in the form of comments. Additionally,
    a report change tog must be updated each time a
    report Is modified.
Responsible
AA-OARM
    When a change is made to one program report
    developers will consult the reports librarian to
    see If related or affected programs need to be
    changed.

    OEM/USDS will review reports usage analysis and
    soli fit user comments to Identify all reports critical
    to end of the year reporting and FJ90 planning.  These
    reports will be given highest priority for review and
    correction. As problems an identified, users will
    be informed.

    All reports not Identified as critical to end of the
    year reporting or mOplaming will be removed from
    At CBU3JS National Reports lienu and made
    available only through a special menu on the production
    system until such time as each report can be verified,
    tested and released onto the National Kepons Uenu.
    As part of this complete aitdti o/CEKCUS reports.
    reports will also be Identified for deletion or combination.
ID
Reference/Pane

OIG-A-V1/3
'Action
Status
                                                     Date    Comments
AA-OSWER     OIGB-V1/14
AA-OSWER     OIG B V119
Completed     Various   NDPD has published user memos, brochures, and other
                         documents describing the NDPD security policy. RACP. etc.
                     Completed      Dec-89  OSWER has established the reports librarian function and
                                              revised the reports development procedures.  All five of the
                                              following audit recommendations have been incorporated
                     Completed      Dec-89  into the  reports development procedures.
AA-OSWER     OIGB-V1/14     :  Completed       Dec-89  as above
AA-OSWER     OIG-B-V1/14       Completed       Dec-89 as above
AA-OSWER     OIG-B-V1/14       Completed       Dec-89 as above
AA-OSWER     OIG-B-V1/17       Completed       Dec-89 This was completed by establishing the reports librarian.
AA-OSWER     OIGBV1/17       Completed      Dec-89  This was completed by establishing the reports librarian.
                                                                                                                           •o
                                                                                                                           11
                                                                                                                           z
                                                                                                                           D
                                                                                                                           X
                                                                                                                           M
                                                                                                                           X
                                                                         |l| NOTE:  See Appendix Vn for Ml ciutio* oTccferace.

-------
          Audit Recommendation

Activate the KACF option PROTECTALL.

Require that a complete assessment o/CEKCUS
software bt performed as soon as feasible.

Require that evaluations be made of the four areas
discussed herein: data management, change controls.
data bait integrity, and security.

Require the Director of the OSWER Information
Management Staff to include the requirement for
Independent testing and verification procedures
In the performance of system evaluations.

Require that RA CF training be developed and made
mandatory for system and account managers. .
 Immediately Initiate a review of PAK and TAPP to
 determine the appropriate access levels for
 Individual users and eliminate access to those
 users that do not have an absolute need.

 Promulgate formal EPA guidance regarding the system
 decision process during the development and
 implementation of large information systems.
Responsible .
D-NDPD
AA-OSWER
HI
Reference/Page
OIG-D-V1/2
OIGE-V1/8
Action
Status
Compli
Compli
AA-OSWER    OIGE-V1/6
AA-OSWER    OIGE-V1/6
DOIRM
DOIRM
AA-OARM
 Involve all user groups in developing and implementing  AA-OARM
 the Agency's financial management system, and
 document that their needs and priorities were
 considered In deciding the direction and plans for the
 system.
OIGG-V1/3
OIG-G-V1/3
OIG-I-V1/18
 Establish procedures to coordinate the update of the
 Mainframe capacity report, master facility plan, and
 budget trading system report to continually reflect
 current workload trends and revised requirements
 in all three documents.
D-OARM/RTP  OIG JV1/10
                   Completed
                    Completed
Completed
Completed
In progress
                OIG-I-V1/32     .   Completed
                    Completed
               Date    Comments

               Dec-93 The RACP PROTECTALL option was implemented in
                       December 1993.
                Jun-90 There was a joint O1G/OSWER/OBRR decision not to
                       perfonn an assessment of CERCLIS while System 2000
                       was the DBMS since the Agency has made a decision
                Jun-90 to move all S2K users to new platforms and assessment of
                       CERCLIS in the current environment was a moot point.
Dec-89 OSWER/IM has established an ongoing delivery order
        for IV&V using the MOSES contract, which provides for
        independent validation and verification of all systems .
        upon request.

Dec-93 NDPD account managers were trained by 12/92. then NDPD
        trained other account managers by 12/93. In addition, all
        RACP security administrators received mandatory training.

Nov-93 Review of profiles and access levels
        is complete, with ongoing reassessment to eliminate
        access for users who do not have an absolute need.
 Jan-94 revised SLCM policy drafted 6/93. review of draft in progress,
        green border .review forthcoming, procedures & stds to be
        developed, related changes have also been made to draft
        revised charter for IRM Steering Committee

 Sep-90 Two formal groups,  exist to manage IPMS.
        There is an executive management group and a systems
        management group.  Both ensure that user needs are
        documented and addressed in deciding the directions for
        the system.

 Sep-92 NDPD updates the reports on a quarterly basis at a minimum
        The Mainframe Capacity Report information is used for the
        Master Facility Plan and the  Budget tracking system.
                                                                                                                                  •o
                                                                                                                                  •o
                                                                                              x
                                                                                              H
                                                                                              X
                                                                   |l| NOTE: See Appendix VII for foil duiio* of Kfarace.

-------
           Audit Recommendation

Establish annual reviews of user-developed'
applications to determine which applications should
be updaleU and. based on the results of those reviews.
lake appropriate action.

Perform a formal cost-benefit analysis to determine
which major applications should be rewritten to
increase their performance and reduce overall costs
and. based on the review results, take appropriate
action.

Establish production control configuration
management and quality assurance procedures for
the user community, which Includes the incorporation
of a limited access central data set for all jobs
placed in production.

Charter the task force to conduct an analysis of
.data/lie and table creation, storage, and retention
to determine If efficiencies can be gained by
eliminating processing steps.

Provide guidance and training to account managers
regarding their roles and responsibilities for
controlling the issuance of access authorities.

Determine the IBM mainframe accounts that process
highly sensitive data and develop  retpdranatts for
protecting resources under those accounts.
 Olve account managers the primary responsibility to
 control issuance of access authorities (CREATE.
 GKPACC. etc.) for users assigned to their accounts.
 Reduce and maintain the number of users with
 SPECIAL OPERATIONS. AUDTTOK. and
 ALTEK access authorities to an absolute minimum.

 Activate the audit trail features of the RACF control
 mechanism for highly sensitive accounts.
                HI
Responsible     Reference/Pane

D-OIRM        OIGJV1/13
D-OIRM        OIG-JV1/13
D-OARM/BTP  OIG-J-V1/28
D-OARM/RTP  OIG JV1/26
D-OIRM
D-OARM/RTP
COMPTROLL
D-OIRM
D-OARM/RTP
D-OIRM
D-OARM/RTP
D-OIRM
D-OARM/RTP
D-OIRM
D-OARM/RTP
OIGKV1/12
OIGKV1/19
OIGKV1/12
OIGK-V1/12
OIG-K-V1/12
                    Action
                    Status

                    Closed
                    Closed
                    Completed
                    Closed
             Completed
             Completed
             Completed
             Completed
                             Dale    Comments

                             Peb-92 This recommendation Tails outside the scope of NDPD's
                                     mission.  The audit was closed by OIG on 2/2/92.
                                     Periodic reviews of the Agency's applications are required.
                                     per Agency Directive 2100.

                             Peb-92 This recommendation falls outside the scope of NDPD's
                                     mission.  The audit was closed by OIG on 2/2/92.
                              Jan-91  Implementation of the job scheduling package
                                     JOBTRACK satisfied this recommendation.
 Jan-91  The objectives of this recommendation will be achieved
         through implementation of the job scheduling package.
         During this process, the JCL will be reviewed and any
         processing inefficiencies can be addressed.

Nov-93  All RSAs have been trained regarding these issues.
         RACP Security Administrator's Guide was completed 11/92.
         All NDPD and other account managers have been trained.

Apr-93  NDPD has trained account managers how to protect
         highly sensitive accounts with RACP. However. NDPD does
         not have the resources to analyze each application running
         on the NCC mainframes.  Directive 2I9S establishes the
         requirements for protecting automated information resources.

Dec-93  Through the decentralization project. NDPD is training
         account managers to assume the recommended
         responsibilities.

Mty-93  The number of users with high level system RACP authorities
         has been reduced to the minimum level required to maintain
         security and service level goals.
                                                                                                                       o
                                                                                                                       n
             Completed      Dec-93 All RACF audit trail features have been activated for
                                     sensitive accounts.
|1| NOTE: See Appendix VII for fell ciMioo of reference.

-------
          Audit Recommendation

Turn on Iht RACF OPERAUtVTfeature to monitor
activities of users assigned Ike SYSTEM
OPERATIONS authority.

Revoke the ALTER authority of those users assigned
highly sensitive accounts who do not warrant this
authority and maintain a minimum number of men
with this authority.

Establish and Implement a plan to phase In the RACF
option PROTECTALL with milestone dates for
completing the plan and all phases of the
Implementation.

Identify and delete account users In 'revoke ' status
who no longer warrant system access.

Plan and implement a method of periodically reporting
to account managers the status of users with RA CF
profiles assigned under the mainframe accounts for
which they an responsible.

Determine what highly sensitive files In Agency
financial systems and systems processing privacy
data would obtain madnum security by utilizing
the RACF audit trail and/or ERASE features.

Develop a RACF Implementation plan, security
objectives, and quality assurance procedures. Include
milestone dates for each of these components.
Enhance Agency software to provide for complete
and accurate transfer and audit trail of data between
the Contract Payment System and IFUS.
 Perform a software assessment to determine if
 CERCUS can be altered to provide man flexibility
for information retrieval.
Responsible
HI
Reference/Page
D-OIRM        OIGKV1/12
D-OARM/RTP
DOIRM        OIG-K-V1/12
D-OARM/RTP
O-OARM/RTP  OIG-KV1/18
D-OARM/RTP  OIGKV1/18
D-OIRM        OIG-K-V1/18
D-OARM/RTP
COMPTROL
D-OIRM
D-OARM/RTP
OIGK-V1/18
 D-OARM/RTP  OIG-K-V1/24
AA-OARM     OIGL-V1/8
 AAOSWER    OIG-N-V1/33
 AA-OARM
Action
Status          Date    Commenu

Completed      Jun-92  OPERAUDIT is on and is being monitored on a weekly basis.
                   Completed     May-93 The number of users with high level system RACP authorities
                                          has been reduced to the minimum level required to maintain
                                          security and service level goals.
                   Completed     Nov-93 Implementation Strategy Tor RACP Features completed 6/91,
                                          subsequently revised. PROTECT ALL fully implemented
                                          on 11/30/93.
                   Completed      Jan-94 RACP security admins, now receive reports identifying users
                                          in revoke status, and are responsible for USBRID management.

                   Completed      Dec-93 All account managers received RACP training.  The training
                                          empowers the account managers to obtain RACP profiles
                                          as needed.
Completed     Nov-93 NDPD has trained account managers how to protect sensitive
                       files. However. NDPD does not have the resources to analyze
                       each mainframe application.
                    Completed      Jun-91 The Implementation Strategy for RACP Features was
                                          completed on 6/91 and subsequently revised.  The
                                          implementation plan and procedures are completed and are
                                          being implemented.

                    Completed      Jun-91 Data are transmitted from CPS to IPMS via an automated
                                          nightly cycle.  Bach morning, personnel review a reject report
                                          to address any transactions that did not successfully reach
                                          IPMS. On at least a monthly basts, an automated CPS/IPMS
                                          reconciliation report is run to verify account balances
                                          between CPS and IPMS.
                    Completed     Mar-93 The software assessment report was published by the
                                          •Systems  Development Center in March of 1993.
                                                                                                                                   3'
                                                                                                                                   •u
                                                                                                                                   •n
                                                                                                                                   s
                                                                                                                                   u
                                                                 |l| NOTE: See Appendix VU fat Ml ciMkM of itfeteace.

-------
              Audit Recommendation
                (I I                  Action
Responsible     Reference/Page      Status
Dale     Comments
    Discontinue hard-coding of parameter values subject
    to change into the source code and replace them with
    a table that may be read in from a data file unless
    It can be shown through a cost/benefit analysis
    that hard-coding is the more efficient and mart cost
    effective approach.                    .  .

    Establish procedures to Include essential
    reasonableness, completeness, and edit checks in
    programs to preclude the reporting of erroneous data.

    - Sgnoffupon the completion of report
    programing with a certification that a 3-way check
    and comprehensive documentation review hat been
    completed.
AA-OSWER    OIGNV1/33       Completed
AA-OARM
AAOSWER    OIGNV1/20       Completed
AA-OSWER    OIGN-V1/27       Completed
Mar-93  Reports using hard coding of parameter values are being
         phased out at major revision steps of the system.
 Feb-93  This is addressed by the draft report writing manual, published
         in February of 1993.  The manual is scheduled to be
         completed in final during PY94.

 Feb-93  as above
    - Insert comments into the source code and modify
    the CERCUS reports Horary to reflect changes when
    report* are examined for inclusion in the National
M  Keports library!
X               •                   .
 I
    -ferfoma 3-way check of consistency between the
    Report Specification Form, the CERCUS Reports
    Library, and )he source code upon completion of
    report programming;

    Require IFUS. FHD. and CERCUS officials to work
    together to establish accuracy in IfVS/CERCUS
    rdauidata.

    Correct the specific deficiencies identified In this
    finding.

    Develop error reports to separately capture
    Inaccurate/incomplete CERCUS transactions for all
    reports in production.

    Take appropriate actions to eliminate the specific
    deficiencies identified in the five reports discussed
    in Alt finding.
AA-OSWER    OIG-N-V1/27       Completed
                                                                                                    Feb-93  as above
AA-OSWER    OIGN-V1/27       Completed      Feb-93  as above
AA-OSWER    OIGN-V1/14      Completed
AA-OARM
AA-OSWER    OIGNV1/20      Completed
 Sep-92  1PMS. PMD. and CERCUS representatives have worked
         together to improve data accuracy.
         An automated interface to export data from IPMS to CBRCLIS
         has been created, though the interface is not fully operational.
 Sep-92  as above
AA-OSWER    OIGNV1/20      Completed      Mar-93  Production of error reports began in March of 1993.
AA-OSWER    OIG-N-V1/27       Completed      Sep-92  The reports have all been corrected or archived.
                                                                                  •o
                                                                                  •o
                                                                                  M
                                                                                                                                     X
                                                                                  H
                                                                                  X
                                                                      |l| NOTE: See Appendix VII for full citalio* of refereace.

-------
              Audit Recommendation             Responsible
                                                             HI
                                                             Reference/Page
x
i
-4
                                                AA-OSWER     OIGNV1/27
                                             AA-OSWER    OIG-N- V112 7
   Develop written procedures and controls regarding
   report programing documentation requiring
   CERCUS report owners to:

   — Perform a comprehensive review of report
   programming evaluating adherence to selection
   criteria standards and docanenlalion standards
   prescribed by the National Bureau of Standards
   guidance on software maintenance:
Establish procedures to accumulate costs on ADP       AA-OARM     OIG-O-V2/30
support services contracts for: (I) ADP equipment:
(2) proprietary software; (3) maintenance services;
(4) ADP Services; and (5) ADP support services to
ensure that the Agency obtains required approvals
fromGSA.

Establish senior DOPO/technical manager positions     AA-OARM     OIG-O-V2/22
with ADP technical skill requirements commensurate
with the still levels needed to technically monitor the
development of ADP DOs and deUverabtes, or
consider the use of independent evaluation contractors
to perform the technical reviews of DOs.

Perform a study to determine the appropriate          AA-OARM     OIG-O-V2/22
procurement office staffing to administer the current
and proposed ADP support services contracts.
Add to Ou Agency's contract management program     AA-OARM      OIG-O-V2/22
AgencywUe mandatory and formal IKtt standards,
project management guidance, and SDH requirements.
to efficiently manage ADP support services contracts
and information resources.
    Separate the duties of Government employees who       AA-OAAM     OtO P V2/7
    perform system operations and system security
    activities.
Action
Status

Completed
Date     Comments

Mar-93  This b addressed in the draft report writing manual, first
         published in February of 1992, and scheduled to be made
         final in PY94.
Completed      Sep-92 This is also part of the draft report writing procedures manual.
                                                                                     Completed      Jun-93  OIRM has issued interim policy requiring the accumulation
                                                                                                             of costs by PIP resource categories as defined by the PIRMR.
                                                                                                             This requirement is also included in OIRM's draft PIP
                                                                                                             Resources Acquisition Manual.
Completed       Oct-93 Por all national IRM contracts, OIRM performs technical
                        evaluation of delivery orders containing significant IRM
                        content or issues.  Upon request, OIRM also provides DOPOs
                        technical assistance in preparing DOs and evaluating contractor
                        performance.  Use of GSA zone contracts also ensures
                        independent review of contract delivery orders.

Completed       Sep-93 OAM conducted workload analyses in support of the PY9S
                        budget submission. This analysis addressed current and
                        projected resource needs for properly managing existing and
                        projected ADP contracts for the Agency.

Completed      Aug-91 The IRM clause was added  to the EPAAR in August of 1991.
                        As of 8/92, all current contracts involving PIPS were modified
                        to reflect the EPAAR IRM requirements.  The APDS system
                        puts the IRM clause automatically into all new solicitations
                        for PIP resources  as a mandatory clause.  In addition,
                        IRM training for OAM is currently in development by OIRM.

Completed       Apr-92 As of 4/92. the full-time NDPD Security Officer began
                        reporting to the Director. NDPD. regarding all security matters.
                                                                                                                                                                                       •o
                                                                                                                                                                                       PI
                                                                                                                                                                                       o
                                                                                                                                                                                       M
                                                                                                                                                                                       x
                                                                       |l| NOTE: See Appendix VO for full ciuioa of rcfeieoce.

-------
          Audit Recommendation

Add RACFprotection to the following: all CPS
data sets; ADCR budget and obligation Jala; and
all Mills program, budget, and budget reconciliation
data.
Responsible
(I |                  Action
Reference/Paee      Status
AA-OARM      OIG-P-V2/6        Completed
Date    Comments

Nov-93 All data sets running on the NCC IBM compatible mainframes
        are protected by RACP.
Based on the specific job responsibilities, redact
the lumber of users with; SPECIAL* OPERATIONS,
and AUDITOR attributes; ALTER access to data sets
controlling RACF: ALTER access to operating system
data sett; ALTER access to IFHS and UFA f; and
CKPACC capabilities and CREATE authority In
several groups.

Develop performance standards Jar system software
maintenance, security administration, and DASD
management and require the contractor- to develop
and doeiment specific procedures to meet the standards.

Kased on specific individual job responsibilities.
reduce trie number of users with:
a. SflClAL, OPERATIONS, and AUDITOR attributes;
b. ALTER access to data sets controlling RACF:
c. ALTER access to operating system data sets; and
d. OtPACC capabilities and CREATE authority in
  several groups.

Develop information security policies, standards,
and procedures for data protection, Including:

a. determining the level of protection reauired
under RACF:

b. developing RACF profiles which provide protection
wUlt maintaining adequate Internal controls suck
at separation of duties: and

c. develop and publish standards of performance
far Information security, with specific attention to
utiUiation of RACF to protect sensitive data on
me IHH systems.
AA-OARM     OIGPV2/7        Completed
                                   May-93 The number of individuals with high level system RACP
                                           attributes has been reduced to a minimum.
AA-OARM     OIGQV2/53      Completed
AA OARM     OIG-Q-V2M5       Completed
                                   Completed
AA-OARM     OIGQV2/14   '   Completed
AA-OARM     OIG-Q-V2/14      Completed
AA-OARM     OIGQV2/14      Completed
                                    Jun-93  NDPD developed an MVS procedures manual (6/93) and a
                                            DASD management plan (2/93)  that cover the issues in this
                                            recommendation.
                                   May-93 The number of individuate with high level system RACP
                                         .  attributes has been reduced to a minimum
                                   Nov-92  NDPD created the Application RACP Security Administrator's
                                            Guide, which addresses many of the following topics:

                                   Nov-93  NDPD provides comprehensive  recommendations on these
                                            topics as part of training other organizations to become RACP
                                            aware. Agency directives 219S  and 2197 are also
                                   Nov-93  responsive to these recommendations.  Training for almost
                                            all of the Agency accounts was completed by 11/93.
                                    Peb-93  NDPD has complied with item c) through the
                                            NDPD .Operational directives.
                                                                                 •o
                                                                                 •0
                                                                                 tl'
                                                                                 '£
                                                                                 rj
                                                                  |1| NOTE:  SeeAppoiduVUfarfallciUtiomofrrfere.*.

-------
             Audit Recommendation

   Strengthen NCC security policy to define:
   a. controls over sensitive application data;
   b. Controls over access to operating system data sets
     to include requirements for minimum access authority;
   c. controls over powerful RACFprivileges; and
   d. division of responsibilities for RACF administration.

   Update written policies, standards, and procedures for
   systems software activities to include proper APF
   administration and SVC installation. Hu policies
   should Include the requirement to preserve IBtt system
   integrity, and follow vendor guidelines for maintaining
   system Integrity.

   Issue a requirement that all support services contracts
   Incorporate performance requirements that meet the
   intent of OMB A-76 performance-oriented work
   statement and quality assurance turvtUance plan.
Responsible
AA-OARM
X
 I
\D
in
Reference/Pace
OIG-Q-V2/15
   Separate the duties of Government employees who       AA-OARM      OIG-Q-V2/14
   perform system operations and system security activities.

   Establish a Ugh level IRtl Steering Committee         DA-EPA        OIG-R-V2/19
   which acts as a decision-making body for significant
   OUt activities, headed by DA or DSO with members
   being senior executives with authority to commit
   offices to action.

   Formally designate a DSO in accordance with the       A-EPA         OIG-R-V2/19
   PRA at the AA level.

   Delegate the authority and responsibilities for all       A-EPA         OIG-R-V2/19
   the OUI functions to the DSO in accordance with
   the PRA. and clearly define any redelegations.

   Establish a clear chain of command under the          DA-EPA        OIG-R-V2/19
   DSO for all [Kit activities, especially between
   OOtH and NDPD.
Action
Status

Completed
AA-OARM     OIG-Q-V2/24       Completed
AA-OARM      OIG-Q-V2/53       In progress
                                     Completed


                                     Completed
                                     Completed

                                     Completed
Date     Comments

 Peb-93  The NDPD security policy has been updated to reflect
         the OIG recommendations.
                                     Jun-93  NDPD has developed the MVS Standards and Procedures
                                             Manual, which addresses these issues.
                                     Jul-93  The TOSS Task Force reviewed all TOSS SOWs to ensure
                                             the use of performance-oriented requirements. This
                                             requirement is also spelled out in the PIP Resources
                                             Acquisition Manual, to be issued as final in 2/94. NDPD
                                             requires that all support services contracts comply with
                                             Circular A-76 requirements, including those regarding
                                             performance-oriented work statements and QA surveillance.

                                    Apr-92  As of 4/92. the full-time NDPD Security Officer began
                                             reporting to the Director. NDPD. regarding all security matters.

                                    Dec-92  Steering Committee charter has been revised to reflect
                                             AA/OARM as chairman. Further refinements to charter
                                             and Committee membership in progress. AAs scheduled
                                             to meet with the AA/OARM in January  '94.
                                    Dec-92  AA/OARM was formally designated as Agency's Senior
                                             Official for IRM in Delegation 1-84

                                    Dec-92  Delegation 1-84 redelegates specific portions of the IRM
                                             program to OPPB.
                                     In progress     TBD
                                             Further revisions to Delegation 1-84 have been drafted
                                             to establish a clear chain of redelegations to the
                                             office level.
                                                                                                    tn
                                                                                                    O
                                                                                                    n
                                                                                                    x
                                                                                                    M
                                                                                                    X
                                                                       |l| NOTE: SeeAppettfcxVUforMlciuUoaofreferace.

-------
          Audit Recommendation

Develop a comprehensive Agencywide oversight and
enforcement progran which focuses on software
quality and the system development life cycle and
which at a minimm should Include...
a provision of training on Information system quality
assurance.
Responsible
AA-OARM
                                                            HI
                                                            Reference/Page

                                                            OIG-R-V2/58
Action
Status

In progress
    Establish a formal. Agencywide. Integrated planning
    process for the direction, coordination.' and control
    of IRM activities and resources that will provide
    management involvement and accountablity at all
    levels, which at a minimum should include the:

I   a. Development and implonentallon of an action plan
O  to accomplish Agencywide mission-based
    bottom-up OUi planning.

    b. Establishment of an evaluation and review process
    for program offices' IHH mission-based plans to
    ensure the plans support a consolidated Agencywide
    mission-based UUI plan.

    c. Integration of the responsibilities for OUi
    planning and budgeting.
d. Modification of the methodology for OOI planning
to include dear policies and procedures for linkage
of the planning and budgeting processes.
                                            DA-EPA
                OIG-R-V2/46
                                                                               In progress
Date    'Comments

 Jul-93 The MOSES program has presented briefings on the System
        Devel. Ctr. product development process, which describe the
        product assurance process during development of deliverables.
        A 3-day class on managing software requirements and the role
        of product assurance during a software project has been pre-
        sented twice.  Another 3-day class, more focused on product
        assurance, is being considered for presentation during FY94.
        These offerings are being considered for training components
        of a comprehensive Agencywide oversight and enforcement
        program.
        OIRM has also established the oversight and QA function in
        the Oversight and Compliance Support Staff in OIRM/MES.
        which ensures new acquisitions comply with EPA policies.

 Jan-94 The IRM Planning Group was established to develop these
        functions. In addition at a Jan. '94 meeting of the AAs.
        they agreed to directly work on an IRM Strategic Plan
        for the Agency.
                                            DA-EPA
                                             DA-EPA
                OIG-R-V2/46
                OIG-R-V2/46
                                            DA-EPA
                OIG-R-V2/46
                                            DA-EPA
                OIG R-V2/46
                                                                               Completed
                                                                               In progress
                                                                                In progress
                                                                                In progress
                Apr-93 The action plan was completed in April of 1993.
                        currently being implemented.
                                                       His
                Nov-93 Formal procedures are under development.  Responses to
                        the 1993 integrated data call for IRM planning and budgeting
                        are currently being reviewed and will be part of the Agency
                        IRM Strategic Planning process.

                Dec-92 The revised charter of the Executive Steering Committee for
                        IRM integrates these responsibilities at a high level, at did
                        formal designation in December 1992  of the AA/OARM
                        as EPA's Senior Official for IRM.

                May-93 The formal procedures are currently in development.  The
                        Agency's first integrated IRM planning and budget data call
                        was issued in 1993, and is currently being analyzed.
                                                                                                                                                                                   •u
                                                                                                                                                                                   v
                                                                                                                                                                                   w
                                                                                                                                                                                   z
                                                                                                                                                                                   o
                                                                                                                                                                                   M
                                                                   |l| NOTE: See Appendix VU lot fell cilatkm of icfcreace.

-------
               Audit Recommendation            Responsible

     Develop a comprehensive Agencywide oversight and    AA-OARM
     enforcement program which focuses on software
     quality and the SDLC and which at a minimum should
     Include the:

     Development and implementation of an action plan to   AA-OARM
     accomplish Agencywide quality assurance for major
     information systems.
               HI
               Reference/Pane

               OIGR-V2/57
                   Action
                   Status
                            Date    Comments
                OIGRV2/58
                   In progress     TBD
                   In progress     TBD
                                    The Systems Development Center was created to emphasize
                                    the development of high quality information systems. Under
                                    EPA direction, the MOSES contractor das developed a
                                    quality assurance program and has created a separate QA
                                    group within the SDC, reporting directly to the program
                                    manager. Product assurance procedures are being developed.
                                    documented and practiced, and will be refined over the
                                    life of the MOSES contract. It is OIRM's intention that such
                                    procedures will be broadened for adaptation to other EPA
                                    IRM contracts.
x
 i
     Establishment of an oversight and enforcement
     function to be responsible for the overall Information
     systems quality assurance program to Include
     Independently reviewing and evaluating major
     information systems.
                                                  AA-OARM
                OIG-RV2/58
                   Completed
      Fomaliie and prioritize a plan for developing and
      revising policies, standard!, and procedures which
      addresses the Issues presented in this finding, which
      also Include Ike following actions:

      Review existing IRM guidance documents and
      incorporate them as necessary into IRM policies.
      standards, and procedures under Directive 13 IS.

      tmedwlely issue temporary directives for informal
      guidance and standards as set forth in Directive
      1315 on critical OUi guidance documents until
      green bonier review can be performed.

      Develop additional comprehensive, formal.
      authoritative. IKtl policies, standards, and
      procedures which would cover all minimum Federal
      and EPA IRM requirements.
                                                  DA-EPA
DA-EPA
DA-EPA
DA-EPA
                OIG-R-V2/3S
OIG-R-V2/35
OIG-R-V2/35
OIG-R-V2/35
                    Completed
             Completed
                             Jul-93 The oversight and quality assurance function has been
                                    established in OIRM/MES and implemented by the Oversight
                                    and Compliance Support Staff. This Staff ensures new
                                    acquisitions comply with EPA policies for information systems
                                    design, development, and maintenance.  In addition,
                                    OIRM assures compliance and adherence to appropriate
                                    lifecycle practices for projects run under the MOSES
                                    contract for national systems development. Finally, starting
                                    in PY94, OIRM will initiate systems reviews to monitor
                                    compliance with Agency and Federal IRM policies & stds.

                             Jun-93 A comprehensive IRM policy workplan has
                                    been developed and is being used as the basis for prioritizing
                                    revision of IRM policies, standards, and procedures.
                Jun-93 OIRM coordinated with NDPD and SIRMOs to review
                       existing guidance documents and formalize a plan for devel-
                       oping and revising policies, standards, and procedures.
                       The plan identifies subject mailer experts and sets priorities.
Completed      Apr-93 OIRM developed a list of candidate policy documents to be
                       issued as temporary directives, had this list reviewed by the
                       Agency IRM community, and had a set of 8 critical IRM
                       policy documents approved as formal directives on 4/30/93
             Completed      Jun-93  Two additional policies recently completed include a policy
                                     on access to computer equipment by the disabled, and a
                                     policy on use of electronic signatures within EPA.
           .   .                       Additional policies have been drafted on topics such as
                                     telecomm.. background investigations for IRM contractors,
                                     HW/SW standardization, and systems lifecycle mgmt.
|l| NOTE: See Appendix VII for fullciudo. of reference.
                                                                                                 t»

                                                                                                 o
                                                                                                 M
                                                                                                 X
                                                                                                 1-1
                                                                                                 X

-------
              Audit Recommendation
Responsible
    Updating and establishment of dear policies, standards.  AA-O ARM
    procedures, and guidelines on Information systems
    quality assurance and incorporation of them into
    format EPA directives.
    Establish and maintain a central repository for         DA-EPA
    IRM policies, standards, procedures, and guidance.
    Include segments on managing electronic records       A-EPA
    In Agency records management training sessions, m
    addition, develop presentations for OUI officials
    and program staff responsible for electronic records
    systems that inform them of their responsibilities In
    relation to electronic records, particularly In terms
H  of building in maintenance and disposition at the
 I   system development phase, and of creating and
to  maintaining appropriate documentation.
    Revise and expand the guidance for creation.           A-EPA
    maintenance and use, and disposition of electronic
    records already in place to mate It more complete
    and current, hi particular, Incorporate, as appropriate
    and In a farm applicable to EPA. the guidance In
    NARA's regulations on electronic reeordkeeping.
    found at 36 CFR 1234. and In NARA 's handbook
    entitled "Managing Electronic Records. • Ensure
    that records maintained on all types of systems an
    Included. As part of the EPA directive system, this
    guidance should be desstminated to all records managers
    and administrators and OUt staff,  both in headquarters
    and the field. Clear assignment of responsibility
    for developing and maintaining documentation for
    electronic records should be included tn the directive.
ni
Reference/Page

OIG-R-V2/58
Action
Status
Date     Comments
                OIG-R-V2/35
                EPA-A/52
                EPA-A/51
In progress      Jun-94  Draft of system lifecycle mgmt. policy is in pre-green bonier
                         review.  Procedures and guidelines will be developed to
                         complement the policy.  The current system design and
                         development guidance was issued as temporary Agency
                         directive 2182 in April 1993.

Completed       Peb-93  Copies of documents in the IRM policy document inventory
                         are maintained in OIRM/1MSD. OA/MOD retains the central
                         repository of all official Agency directives and is currently
                         in the process  of examining all directives as part of the
                         administration's streamlining initiative.

Completed       Dec-93  An electronic records segment has been included in the
                         standard records management training given to program staff.
                         We made many presentations to IRM branch chiefs, SIRMOs,
                         etc.. on electronic records management 'and developed records
                         disposition schedules for most information systems listed in the
                         Agency's information systems inventory.  We featured elec.
                         records mgmt. concerns in the 4/92 issue of INPOACCESS,
                         and included other articles on elec. records mgmt. policy in
                         other  issues.  We plan to develop a 2-hour training session for
                         all records managers on electronic records for PY95.

Completed       Dec-93  We included revised wording on managing electronic records
                         in the revised version of Ch. 10 of the IRM  Manual  (records
                         management) scheduled to go through green border  review
                         in PY94. We  completed a study of existing policy, procedures.
                         and guidance on electronic records thai indicates areas where
                         our policies need improvement. We plan to revise Ch. 8 of
                         the Agency Records Management Manual (Directive 2160)
                         during PY94,  and prepare it for informal review during PY9S.
                         We distributed copies of NARA's "Managing electronic
                         Records" to all records officers and SIRMOs.
                                                                                                                                         •o
                                                                                                                                         •o
                                                                                                                                         B
                                                                                                                                         D
                                                                                                                                         M
                                                                        |1| NOTE: See Appeodi* VU for full ciutioa of reference.

-------
          Audit Recommendation

Develop a process to ensure /KM considerations get
addressed during development of proposed regulations,
and information collection requests QCRs) by
engaging OiRH participation In the Agency Steering
Committee and associated regulatory workgroups.
A integral component of the review should focus on
whether the collections are In compliance with
established Agency IRM policy and standards.
Responsible
D-OIRM
HI
Reference/Page

EPAB/18
                                                                                Action
                                                                                Status
               Date     Comments
In progress     TBD
Establish more formally designated roles and
responsibilities for all members of the Agency IRM
community, reflected in performance standards.
Require AA/RA-tevtl reviews of all office/regional
not organiuaional structures and require formal
designation of authority and responsibility for all
positions In these structures.
Develop a model of a recommended Kit
organiuaional unit in program offices.
 Revisit and redefine SIKMO concept. FDs. and
 perfomunce standards In light of the SIKMO's
 threefold role: Information conduit, approval
 authority, and technical resource.
                                             D-OIRM
                EPA-B/9
                    under
                    consideration
               TBD
                                             D-OIRM
                 EPA-B/9
                    under
                    consideration
                                             D-OIRM
                 EPA-B/9
                    In progress
                                             D-OIRM
                 EPA-B/9
                     In progress
OPPE is leading an effort to revise the regulatory development
process. The existing process did not emphasize up front
analysis, or training, tools, and guidance for workgroup chairs
or participants regarding IRM issues. OARM reps, have
been actively involved with the workgroups developing
proposals for the new process.  Our approach is to. push for
strong upfront guidance for workgroup chairs and participants.
coupled with an evaluation program that allows us to determine
where additional guidance or tools are needed.  We will
maximize the impact of our experts by providing lists of
contacts on specialized issues to workgroups upfront so that
they can access experts on an as-needed basis.

The model IRM program study will define approp. rotes and
responsibilities for the SIRMOs. and in  so doing will show
the relationships between the various elements of the IRM
community. More attention would need to be focused on
central  IRM's specific rotes and responsibilities for adequate
perf. standards to be developed for the entire community.

Decisions on implementation of the model IRM program will
probably  result in  review of AA-tevel IRM structures and
responsibilities. Top management support will be required to
formally designate authorities and have  them implemented.
There is a model for the Regions that appears to be working,
but this has not been addressed in the program office model
IRM  study.
                May-94 The model will include recommended IRM organizational
                        unit placement, staffing, and functions for program offices.
                        The model will be drafted by the end of January,
                        and will be shared with the IRM community in Spring.
                        after review and approval by OIRM/1MSD.

                May-94 In essence, the SIRMO concept gels defined in the model
                        study.  Responsibilities of the IRM unit in program offices
                        will be defined by function, thus these could serve as the
                        basis for job descriptions and functional statements.
               TBD
                                                                                                                                                                                   •o
                                                                                                                                                                                   n
                                                                    |1| NOTE: See Appendix VII for fall ciu«io« of reference.

-------
x
           Audit Recommendation         .    Responsible

Establish a scalable, flowcharted review and approval   D-OIRM
protest for information system development projects.
focusing the review on projects involving Ugh dollar
and/or mission critical systans. Inherent in this
recommendation Is the need to clearly define the
thresholds of projects requiring OUUt/OTUtO review.

Define criteria and thresholds for system              D-OIRM
development/enhancement efforts requiring joint
OnOt/SIRHO formal review/approval. Jnis includes
new system development initiatives and/or major
changes to existing systems. Joint review and approval
at designated Intervals will involve review of the
life cycle products and documented decisions la
proceed or not proceed with the proposed work, and
conclude with the formal decision to retire an
information system.  It should be noted that the
review of the project management plan for all system
development and enhancement efforts should address
the need to comply with IRM policies and standards.

Providt tailored IRM training for various groups.       D-OIRM
based on levels of need and areas of responsibility.
Target groups' Include:             '•

   System development Delivery Order Project
   Officers and Work Assignment managers;
                                                                 HI
                                                                 Reference/Page

                                                                 EPAB/15
             Action
             Status
Date     Comments
             In progress     TBD
                                                                 EPAB/18
              In progress     TBD
         This review and approval process is described in the draft
         system lifecycte management policy: Thresholds
         are clearly spelled out. as are the reviews by SIRMOs,
         central IRM offices, and the Executive Steering Committee
         for IRM.
         Criteria and thresholds for system development and/or
         enhancement efforts are clearly spelled out in the
         draft systems lifecycte management policy.
         The project management plan is one of the most
         important documents referenced in the policy.
                                                                 EPA-B/13
              In progress     Various  Dates for specific mining courses vary.
                                                                                     Completed
                                      The Sys. Devel. Center established under the MOSES
                                      contract is required to operate based upon standard operating
                              Jan-94 procedures which describe how projects receive technical
                                      direction from DOPOs. limits on such direction, and controls
                                      over the deliverable development process.  The contractor has
                                      briefed DOPOs on how work flows through the SDC & more
                                      briefings are scheduled.  We are  considering making the
                                      briefing mandatory training for MOSES DOPOs. (Other
                                      components may be determined by OAM in addition to the
                                      current general DOPO training requirements.) In addition
                                      over IS DOPOs attended • 3-day course on the importance
                                      of clearly stating requirements, controlling changes to
                                      req'ts, and the interaction between requirements, design.
                                      the software being developed, and quality assurance.
      Contractors working on system development
      dealing with issues that are EPA-spedflc; and
                                                                                  In progress
                              Jan-94 Beginning in Jan. 1994. SAIC will be providing its MOSES
                                      contract employees with information security training. These
                                      sessions will continue until all MOSES contract employees
                                      have received mining.
|1| NOTE: See Appendix VII for fell ciuikw of rtfeeace.
                                                                                                                                                                                        PI
                                                                                                                                                                                        SS
                                                                                                                                                                                        D
                                                                                                                                                                                        M
                                                                                                                                                                                        X

-------
           Audit Recommendation
(continued...)
  frogran staff developing refutations involving
  information collection.
Responsible
01
Reference/Page
Action
Status

Completed
Date     Comments

 Jan-94  Briefings will be offered in January to program managers at
         the office director, division director, and branch chief levels on
         the Paperwork Reduction Act. its goab and requirements.
         the ICR process, and how to comply with the Act.  Direct
         training is also provided to reg. workgroups upon request
         to supplement what they learn from reading the ICR manual.
Decrease EPA's dependence on contract tervicet.   .
increase Ike ratio of EPA to contract penonnei. and
Increase the number of Agency IHit FTEs by initiating
a conceited effort to plan for. justify and request a
significant Increase in Federal IRU FTEs.

Require that Agency programs develop mission-eased
IRU plans which will clearly Identify IRU expenditures
in their annual budget and operating plans.

Develop specific policies and processes to ensure that
Agency IRM systems developed under any EPA contract
comply with Federal and EPA IRU policies and
standards. Make certain that Utt work under these
contracts which Is considered incidental to the support
of the contract effort is only for the contractor's
internal use and will not be transferred to Agency
personnel or programs.

Require that all EPA contract, grants and Interagency
agreements must Include provisions for compliance
with EPA and other Federal OUl policies.

Review all Agency IRU policies, standards, and
guidances at regular intervals to identify gaps and
ensure they are in line with current Federal
requirements and support the Agency's primary goals
and objectives.

Establish a standard, formal process for issuance.
periodic review, and maintenance of IRU policies,
standards, procedures, and guidance which shall
include review by program and regional office staff
having functional OUI duties and responsibilities.
                                              D-OIRM
                 EPAB/20
                     In progress     TBD
D-OIRM
D-OIRM
EPA-B/6
EPA-B/6
                        EPA's actions to correct a declared material weakness in IRM
                        planning will support the development of Agency and program
                        IRM plans that can provide solid justification for requesting
                        additional FTEs. Efforts are abo underway to investigate
                        possibilities for converting IRM contractor funds to PTEs.
In progress     Nov-93  Formal procedures to require this are being developed.
                         Program offices were required to submit budget, mission, and
                         acquisition information for the first integrated data call.

Completed     Aug-91  The IRM clause of the EPAAR requires that systems
                         developed under any Agency contract must comply
                         with Federal and EPA IRM policies and standards.
                         In addition, the APDS system puts boilerplate language
                         for IRM automatically into all new solicitations.
                         for PIP resources, as a mandatory clause.
 D-OIRM
 D-OIRM
 EPA-B/18
 EPA-B/5
 Partially
 Completed
 Completed
 DOIRM
 EPA-B/5
 Completed
TBD    All contracts must include provisions for IRM compliance.
         Grants and lAGs are now addressed on a case-by-case basis,
         and may not be readily amenable to a comprehensive solution.

May-93 All policy documents were reviewed to identify gaps during
         creation of the IRM policy workplan
 lun-93 The Agency's formal green border review process is the
         standardized issuance process. The IRM policy workplan
         provides the framework for periodic review and
         maintenance of the IRM policy documents.
                                                                                                                                           •o
                                                                                                                                           TJ
                                                                                                                                           M
                                                                                                                                           §
                                                                     |I|NOTE: See Appendix VU for full citation of reference.

-------
              Audit Recommendation

    CrealeAnainialn an effective online system to
    disseminate all KM pollciej (including cross-
    references, purpose. Issuing organization, points
    of contact, review dale, and compliance requirement}).

    Asses* user needs for the IDEA system, including
    analytical capabilities; develop a formal test plan;
    properly lest existing software; and document the
    IDEA system design and software before developing '
    additional software for the system.
Responsible
D-OIRM
HI
Reference/Page

EPA-B/11
Action
Status

In progress
Date    Comments

 Sep-93 Many IRM policy documents are now disseminated
        electronically via NDPD's EPADOC CD-ROM.  Options
        for on-line distribution are being investigated
        and piloted by OA/MOD.
AA-OE
GAO-E-V1/13      In progress     Sep-94
a
 I   Address data qualify problems In the FINDS

-------
              Audit Recommendation

    Develop an Agencywide information systems
    architecture that explains the tincture of and
    communications among the Agency's information
    resources that are needed to achieve its single- and
    cross-media mission.
X
Responsible
A EPA
HI
Reference/Page

GAOE-V1/14
                                   Action
                                   Status

                                   In progress
    Complete the Agency's cross-media strategy by
    developing policies and guidance'and instituting
    management procedures to plan, coordinate, and
    budget for crom-media information resources and
    activities.
                                                A-EPA
                GAOEV1/13      In progress
Date    Comments

 Sep-94 The IMDA Group is leading the effort to develop an info.
        architecture for EPA. with initial concentration on a data
        architecture. The IMDA group is taking an approach based on
        the Zachman Framework for Information Systems Architecture.
        John Zachman has given a presentation to OIRM and program
        office staff.  Work is underway to specify the expected content
        of the architectures for EPA and to prepare a development
        plan. Since this is an expensive task requiring involvement of
        many EPA staff, an education effort has been started by
        inviting a nationally recognized speaker to present the
        business case for architectures to EPA staff.   Also in
        preparation for this work, several EPA staff  have attended
        Zachman and related seminars on information system
        architectures. Planning and education efforts will continue in
        FY94, with architecture development being initiated in PY9S.

 Jan-94 The IMDA Group's work to develop an Agency data
        administration policy is supportive of this recommendation.
        The draft policy will begin green border review during PY94.
        During PY93 work was initiated to define conceptual,
        logical, and physical level models for EPA and to develop
        quality metrics for them. Procedures for reviewing these
        models are  currently under development.  Recent ankles also
         indicate that the White House Office of Environmental Policy
        and OMB are discussing incorporating ecosystem planning
        issues in budgeting, and a possible exec, order on ecosystem
        management that would encourage interagency. state, and
         federal cooperation. Admin. Browner has created a sr. mgmt.
        workgroup on ecosystem protection, which will present
        recommendations to Browner by 3/15.
     Strengthen OPf'i conformanci with federal guidance
     and generally accepted practices for automated systems
     development so that OFP'i information systems are
     consistently planned, developed, and enhanced. As
     part of this effort. OFF should ensure that the
     pesticide information needs of all users Involved In
     administering and managing EPA's pesticide
     reregistmtion process are defined and Unlud to an
     overall program management plan.
AA-OPPTS     GAO-H-V1/8
                                    Completed       Jul-93  Developing a multi-year OPP/OCM strategic planning process
                                                            and Unking this process with OPP's budgetary process in PY92
                                                            and PY93 has created a cross-divisional path for IRM project
                                                            planning and resource allocation and is a major step toward
                                                            ensuring overall program office mission goab and objectives
                                                            are reflected in OPP's IRM activities.  In addition, this planning
                                                            provides IRM project managers with somewhat more  realistic
                                                            and timely information about expected mission timetables and
                                                            resources for developing full-scale IRM project plans. This
                                                            greatly facilitates system Ufecycte management and contributes
                                                            to more fully coordinated and controlled systems development.
                       III NOTE: See Appendix VTI for hUcit*io« of reference.
                                                                                                                     •o
                                                                                                                     M
                                                                                                                     X

-------
              Audit Recommendation

    Establish data managtmenl policies and implement a
    plan with milestones, for resolving OPP systems'
    data integrity problem.
Responsible
ID
Reference/Page
AA-OPPTS     GAO-H-V1/8
Action
Status

Completed
S
H"
00
    As OPP moves toward systems Inttfration activities.
    ensure that requirement* analyses, feasibility studies.
    and cost/benefit analyses are conducted to support
    OPP's automated systems solutions.
                                                AA-OPPTS     GAO-H-V1/8
                                   Completed
                                   and
                                   Ongoing
Date    Comments

 Jul-93 The central issues in OPP's data mgmt. process are being
        addressed in two major integration efforts that follow the goals
        in the OPP/OCM strategic plan. These 2 teams span organiz-
        ational lines and involve both clients and developers. The
        teams reflect  data environment lines, with core enterprise
        data addressed in one team and activity tracking data in the
        other.  The data management issues that led to a decision to
        launch full-scale integration efforts are these:
        * core data in our systems is being derived from multiple
          sources - leading to data validity problems
        * core data in our systems is variously denned, both from a
          system design and from a logical use standpoint —  leading to
          data reliability and interpretation problems.
        * multiple update streams and non-synchronous update timing
          across systems leads to data validity problems.
        All these issues as well as system architectural issues are
        being addressed in  a standard controlled systems lifecycte
        management approach, with full-scale documentation and an
        OPP-wide data dictionary.

 Jul-93 In 1993. OPP integrated three different pesticide reregistration
        systems (ALISS, SMARTS, and DCI system) into one system
        called the Chemical Review Management System (CRMS).
        For this system, a requirements analysis and a feasibility study
        were completed prior to developing the system. The cost/
        benefit analysis is in progress and will be completed shortly.
        The other systems lifecycle documents were prepared in
        accordance with EPA Directive 2182, on systems design and
        development.  The  office plans to follow Agency policy and
        standards as it moves forward with other integration projects
        in  1994.
                                                                                                                                                                                   §
                                                                                                                                                                                   o
                                                                      III NOTE: SceAfpeadUVUforfollcitMioaofRfeftoce.

-------
              Audit Recommendation

    Take appropriate steps to enhance its information
    system development process and fully ensure that
    data collection efforts complement each other and
    support the program mission. Specifically, a
    comprehensive data collection plan should be developed.
    Steps should be taken to improve the assignment
    of responsibilities for planning and directing the
    development of Information system components by
    increasing the authority of the central coordinating'
    office to develop data collection efforts and ensure
    consistency. Finally, the lift cyde management system
    should be refined to ensure the complete and
    detailed analysis and documentation of each stage
    of the cyde for major system components.

    Ensure that stale data collection and quality control
    efforts receive fully adequate support and include
    specific Indicators related to data collection and
X  verification in the Agency's mechanism for
I-1  monitoring state performance.

    Amend federal recordteeping and reporting
    regulations so that states art required to collect
    and provide standard data elements in a
    disaggregated form and hazardous waste handlers are
    required to provide sufficiently detailed data.

    Ensure that the toxic chemical release inventory
    reporting system complements other hazardous
    waste data collection efforts so that the data it
    provides on toxic chemical concentrations can be
    used to dietr maximum potential.
                                             Responsible
HI
Reference/Page
                                                                                      Action
                                                                                      Sums
                                             AA-OSWER     GAOKV1/39      Completed
                                             AA-OSWER     GAOK-V1/103    Completed
                                             AA-OSWER     GAO-K-V1/103    Completed
                                              AA-OSWER    GAOKV1/103    Completed
                                              A-EPA
Ensure that the new accounting system and property
control system (I) provide accurate and reliable
financial and management control records (including
the type of asset, date of acquisition, cost estimated
useful life, applicable depreciation data, physical location, and
 identity of custodial officers) to account for and control property assets
 and (2) contain a common data elemental or tnttrface(s)
to permit the reconciliation of accounting and property systems data.
 GAOBLIB/19      In progress
Date     Comments

 Oct-92  The Agency prepared an amendment to the biennial report
         regulations that would have ensured that all stales collected
         consistent data on hazardous waste generation and mgmi.
         This amendment would have required consistent data collection
         and would have enabled more consistent national reporting of
         the data. However. EPA has decided to defer the rule until
         after the 1993 and 1995 data collection cycles are complete.
         Per discussions with Slates and other interested parties. EPA has
         hypothesized that collection of consistent data will not require
         a change to the regulation, but rather nation-wide use of a
         consistent form for data collection and system for data entry.
         In 1987,18 states were using the BR forms.  In 1991,47 states
         used the BR form. The Agency has followed all important
         lifecycle management procedures.

 Oct-92  The Agency developed improved biennial reporting data
         entry and report software for the states which improved slate
         data collection and data quality control efforts. The Agency
         developed an amendment to the federal recordkeeping and
         reporting regulations so that slates would be required to collect
         and provide standard data elements, and hazardous waste
 Oct-92  handlers would be required to provide sufficiently detailed
         data.  However. EPA has deckled to defer the rule until
         after the 1993 and 1995 data collection cycles are complete.
         Per discussions with Suites and other interested parties. EPA has
         hypothesized that collection of consistent data will not require
         a change to the regulation, but niher nation-wide use of a
 Oct-92  consistent form for data collection and system for data entry.
         In 1987.18 states were using the BR forms. In 1991, 47 slates
         used the BR form.
         The Agency undertook a year-long study of possible
         linkage between BRS and TR1.  The result of this effort was to
         show a less than 25% overlap in faculties reporting.

 Sep-93  PMSD and PMD convened a quality action team (QAT)
         during PY93 that examined requirements for a combined
         property and financial accounting system, and recommended
         acquiring an additional module of commercial software (the
         PPS fixed assets module). The final decision on acquiring the
         module is still pending.
                                                                                                                                                                                     •o
                                                                                                                                                                                     •a
                                                                                                                                                                                     M
                                                                     |l| NOTE: See Appendix VUfor full ctUUo* of reference.

-------
   EPA  OFFICES WITH MAJOR
     IRM RESPONSIBILITIES
                                 APPENDIX X
           ADMINISTRATOR
        DEPUTY ADMINISTRATOR
OPPE
      OARM
      (DSO)
       OIRM
OARM
 RTP
OARM
 Cinn
                   NDPD
             IRMD
              X-1

-------
                                                      APPENDIX XI
                     STRENGTHENING IRM AT EPA

            OFFICE OF INFORMATION RESOURCES MANAGEMENT

                Issue Paper Updated February,  1994


Background:  This Issue Paper was provided by OIRM officials to
show the reader the aggregate, collective efforts the IRM
community is undertaking to improve information management at the
Agency.  It presents a summary of many of EPA's current IRM
improvement initiatives, and gives a brief description of the
context in which they were developed.
                              XI-1

-------
                                                      APPENDIX XI

Introduction

     The purpose of this paper is to present the IBM improvement
program and our overall strategy for addressing weaknesses that
have been identified with EPA's IBM program.  The paper is
characterized as "strengthening1* the IRM program, not building
anew, since many of the components for a solid IRM program are in
place,.and some have been judged in the past to be excellent.

Historical Perspective

     The EPA IRM program parallels the structure and operating
philosophy of EPA as a whole - centralized planning, policy, and
oversight, and decentralized implementation.

     This modus operandi served the Agency fairly well in
providing systems to support the numerous stand-alone
environmental statutes and programs.  To support this
arrangement, OIRM invested limited resources in planning, policy,
and oversight, and applied the bulk of its resources to services
that supported implementations, or to implementations directly.

     Two factors began to emerge in the late 1980's that have
made the traditional EPA approach to IRM less effective.  One,
the Agency's business has been informally changing to be more
oriented to cross-media and integrated approaches.  Using the
traditional EPA model and emphasis, IRM has not been able to
support this new business very well.  Second, several statutes  .
and implementing regulations focused on improving Federal IRM
were issued or amended (Paperwork Reduction Act,  Computer
Security Act, A-130).  These authorities require a stronger
central IRM presence in EPA.  Despite these two factors, EPA OIRM
has had difficulty disengaging from its services and
implementation work and reorienting resources towards the
strongly central activities of planning, policy,  and oversight.
Recent budgetary pressures have lead to reduced OIRM budgets, and
have further hampered this reorientation.

Accomplishments

     Despite the effect of several areas of concern, and a
historical IRM strategy that may not provide the basis for
providing strong support to the Agency's strategic direction
towards integrated,  cross-media analysis, the IRM community has
achieved a number of notable accomplishments.  Below are listed
some selected, representative examples:

     o  Implemented the Systems Development Center to improve
        EPA systems through development methods and discipline.


                              XI-2

-------
                                                       APPENDIX XI

         EPA received  an award from "Government Computer News"
         recognizing this effort.

     o   Installed a supercomputer  to  support  environmental
         research at EPA's new facility at Bay City Michigan.

     o   Operated the  National Computer Center at a level of
         efficiency, as  adjudged by industry expert Nolan, Norton
         and Company,  in excess of  10% more cost effective than
         the norm for  data centers  of  like size*

     o   Established,  out of base,  a unit dedicated to
         establishing  the framework for data management in EPA,
         and developed a set of initial standards.  This key
         activity is establishing the  "rules"  required  to support
         data integration across EPA programs  and with  the outside
         world.

     o   Established disaster  processing capabilities for three
         key information systems in EPA, to enable EPA  to continue
         to  operate as an organization if the  data center
         experienced an  incapacitating event.  EPA received an
         award from "Government Computer News" recognizing this
         achievement.

     o   Developed and installed in a  number of offices  in HQ and
         the Regions the Office Forms  Facilitator (OFF)  suite of
         systems to enhance the productivity of EPA offices.  A
         January 1994  study of  OFF  usage in typical office
         settings documented significant productivity gains and
         efficiency improvements for the administrative processes
         automated by  the  system.

     As  an  overall- indication  that the EPA IRM community has
managed  the overall program quite well, Congressman Jack Brooks,
former Chairman of the  House Government Operations Committee,
was quoted  in "Government Computer News" several years ago as
indicating  that EPA was  one of a select few agencies "doing a
good job" in acquiring  and using computer and communications
systems  and services.

Areas of Concern

     The OIG has repeatedly identified IRM as a candidate for
consideration by the  Senior Council on Management Controls as a
material weakness.  GAO has also identified IRM weaknesses as
contributing to problems the Enforcement and Pesticides programs
have encountered in performing their missions.
                               XI-3

-------
                                                       APPENDIX XI

      In  reviewing  the work and  conclusions  of  such oversight
 organizations,  OIRM has  organized the primary  IRM concerns into
 five  areas:   1)  IRM Planning -  This  area  includes creating a 5-
 year  IRM plan and  integrating that plan into the  budget process;
 2) Formal IRM Policies - This area focuses  on  formalizing through
 the green border process a number of policies/standards/etc,  with
 Agencywide impact,  developing additional  policies to  cover some
 identified gaps, and improving  communications, training,  and
 assistance for  those affected by  the policies; 3)  IRM Security -
 This  area deals with creating a comprehensive  security program
 which assures cost-effective protection of  sensitive  Agency
 information,  by focusing on training and  oversight, covering a
 few policy gaps, and improving  the mainframe environment at NCC
 by enhancing  certain general controls; 4) IRM Quality
 Assurance/Oversight - This item focuses on  enforcement of IRM
 policies pertaining to system development,  operations,  and
 maintenance,  and evaluating systems' effectiveness; and 5)  IRM
 Contracting - This  area  deals with formalizing controls to ensure
 IRM contracting is  conducted efficiently  and in accordance with
 Federal  regulations,  both  during  the pre-award phase  of acquiring
 Federal  information processing  (FIP) resources and throughout
 contract administration.

      OIRM has examined and evaluated these  five areas  in
 accordance with EPA's material weakness criteria.  On  the basis
 of these analyses,  OIRM  acknowledged that all five areas  have
 certain  weaknesses.   In  the case  of IRM Planning and IRM
 Security, OIRM deemed them to be  sufficiently material  to b*
 reported to the President  (see the Administrator's December 1992
 FMFIA letter).   In  the case of IRM Policies, IRM Quality
 Assurance/Oversight,  and IRM Contracting, OIRM deemed the
 weaknesses to be serious,  but not of sufficient importance to
 report to the President, and has  declared these weaknesses at the
 Agency-level,  corrective  action  plans have been developed and
 are being implemented.

 The Improvement Program

      OIRM is  vigorously  addressing, in cooperation with the
 Agency IRM community, the underlying circumstances that have
 resulted in these areas  of  concern by working in five key areas:

      1.  Revising the OIRM  Emphasis -  Historically, OIRM has
 been  substantially  invested in activities related to implementing
 systems.   This emphasis has diverted resources from the
 traditional functions of a  Federal IRM organization of planning,
 policy,  and oversight/review.

     The lack of emphasis on these stewardship functions of
policy, planning, and oversight has contributed to certain

                              XI-4

-------
                                                      APPENDIX XI

weaknesses in the IRM program.  To address this matter we are
taking several actions:

     o  We.declared IRM planning a material weakness, immediately
        bolstered our planning function, and established an
        action plan to create a 5-year IRM plan for EPA that is
        linked to the budget.  The first planning cycle will be
        completed in February 1994.  In addition, with our
        support, the Assistant Administrators gathered in January
        1994 and initiated work together on IRM strategic
        planning.  At the same time, a subgroup of the National
        Advisory Council on Environmental Policy and Technology
        was used to engage the Agency's external stakeholders in
        IRM strategic planning activities.

     o  Out of base resources, we placed senior, experienced IRM
        staff at the lead of our information security program.
        We have developed, and are supporting,  a network of
        information security officers for information technology
        installations across the Agency.  We have launched a
        solid security awareness training program and ensured
        that every data set on EPA's mainframe has been
        evaluated, with a security decision made by its owner,
        regarding the appropriate level of protection to be
        implemented via the Resource Access Control Facility.

     o  We have increased attention to the IRM policy area.   To
        date, we have inventoried all Agency IRM policies,
        standards, procedures, and guidance,  created a
        prioritized listing of additional policies needed,  and
        formally issued a number of key IRM documents as
        temporary Agency directives.   We recently established
        Agency policies on access to computer equipment by the
        disabled and on use of electronic signatures within EPA.
        We are currently revising the Agency's  systems life-cycle
        policy and guidance,  telecommunications policy,  and
        information security-related policies.   We are developing
        anew EPA hardware/software standards and policy,  a
        comprehensive data administration policy,  and a policy
        requiring appropriate suitability investigations for IRM
        contractors.  In the spirit of the National Performance
        Review,  we are focusing our policies more on desired,
        measurable outcomes* (the "what",),  and less on specific
        procedures (the "how").

     o  We established an organization to review IRM
        acquisitions,  major systems,  and IRM organizations  in the
        Agency.   This organization has comprehensively reviewed,
        to ensure inclusion of relevant IRM controls,  all
        requests for procurement. (RFPs)  for FIP resources and all

                              XI-5

-------
                                                      APPENDIX XI

        delivery orders for OIRM-provided national  IBM  contracts.
        We have implemented, at our Systems Development Center
        under the MOSES contract, a very thorough set of
        operating principles, policies, and procedures  that
        ensure good contracting practices, improved planning,  and
        useful, high-quality deliverables.

     o. To more clearly identify the functions that need to be
        •performed in EPA to ensure good IBM is practiced, last
        Fall we initiated an IBM business planning  effort.  After
        the functions of IBM, both for central organizations and
        the programs/regions, are clearly established,  OIBM will
        examine its current organization to determine if any
        changes are necessary to support the central
        responsibilities.  If so, a formal reorganization will be
        proposed.

    • 2.  Assuring senior—level Engagement la IRM — We formalized
the delegation from the Administrator to the Assistant
Administrator for OABM establishing the AA as EPA's Designated
Senior Official for IBM (the "DSO", as required by the  Paperwork
Reduction Act and OMB Circular A-130), and we are working to
clarify related redelegations of responsibility and authority.
We have strengthened the IBM Steering Committee by ensuring
direct DSO leadership, raising the level of seniority of
Committee members, and engaging the Committee directly  in key
decisions on IBM planning, contracting, policies, and major
systems activities.  The Committee's new charter is presently  in
the approval process.

     As one of its first tasks in fiscal 1994, the Committee is
presently focusing the Assistant Administrators' and Begional
Administrators' attention on creating a strategic IBM vision.
The Committee is obtaining the views of the Agency's key external
stakeholders by soliciting input on IBM strategies from the
National Advisory Council on Environmental Policy and Technology
(NACEPT).

     3.  Strengthening IRM Components in the Programs — IBM is
patterned after the organizational culture in EPA - decentralized
implementation and operation.   We are actively working with the
SIRMOs in the Programs to strengthen that position,  and the
support available to organizations performing or
utilizing IBM in the Programs.   This work is primarily in the
form of a Model IBM Program study.  After completing the study in
March 1994,  we will take steps to formally delegate appropriate
functions to the programs/regions, and to create proposed
organizational structures, position descriptions, etc.,  for the
SIRMO organizations.


                              XI-6

-------
                                                       APPENDIX XI

     We  are also working to  strengthen IRM components  in certain
programs by increasing their awareness and knowledge of  IRM
issues.  This effort will focus on periodic  internal
communications, technical assistance, and formal training.   For
example, we recently provided IRM training,  which was  very  well
received, to 42 Contract Officers and Specialists within the
Headquarters Office of Acquisition Management and the  Contracts
Management Division in Cincinnati.

     4.  supporting Data Int•oration - We are involved in this
area in  at least five ways.  First, we have  dedicated  resources
to "set  the rules" by which  data are defined and managed.   We
established the Information  Management/Data  Administration
program  in 1992 towards this end.  This includes EPA data
policies which provide the links to integrate data, like the
Facility ID and the Locational Data Policy.  Setting and
enforcing these rules is critical to success in this area.
Second,  we are working to improve the utility of mechanisms that
are in place to support integration, like the OIRM-operated
FINDS.   An improved version  2.0 of FINDs was released  in late
1993, and we plan to implement version 3.0 during FY94.   Third,
we have  proposed a major initiative to build an integrated
environmental database for EPA called "ENVTROFACTS," and to
provide  initial capabilities focused on selected geographic
initiatives.  This "data warehouse" is envisioned as an  interim
approach to enable better integration and public access,  until
key program systems undergo modernization and reengineering.

     Fourth, we are working on this same effort to develop  access
tools (GATEWAY) to support analysis using ENVIROFACTS data.
Fifth, we are working in concert with other  Federal agencies to
ensure GATEWAY, ENVIROFACTS, and other information locators are
consistent with national and international efforts towards
environmental data integration.   We are also actively engaged in
positioning EPA for successful contributions to the information
superhighway,  via Internet.

     5.   Fixing specific Deficiencies - We are aggressively
addressing the many specific recommendations made by OIG  in
various  IRM-related audits.   Overall,  we believe that implementing
the recommendations will result in a stronger IRM program.  When,
on occasion, we disagree with the specifics of how to implement a
particular recommendation,  our strategy is to implement solutions
that are responsive to the recommendation's intent.

     We have in place specific action plans responding to
recommendations from the following audits:
                              XI-7

-------
                                                      APPENDIX XI

     o  EPA's Management of Computer Sciences Corporations
        Contract Activities - of the 98 recommendations in this
        audit, 32 were related and assigned to OIRM for action.

        All actions assigned to OIRM were completed by the end of
        September 1993.

     o . pPA Needs to Strengthen the Acquisition Process for ADP
        Support Services Contracts - This audit recommended that
        14 specific actions be taken.  All actions were completed
        by the end of September 1993.

     o  EPA Needs to Strengthen General Controls Over System
       . software - This audit recommended that 20 specific
        actions be taken.  All actions were completed by
        September 30, 1993.

     o  EPA Must Fully Address Longstanding Information Resources
        Management Problems - This audit recommended that 28
        specific actions be taken.  All actions will be completed
        by the end of October 1995.

     Addressing the IRM concerns raised in the recent past is
considered a very serious matter by our senior leadership team.
We are engaged in addressing these concerns on a daily basis,
and intend to make permanent changes to improve those areas of
weakness, while retaining the solid capabilities in IRM that have
served the Agency well over the years.
                              XI-8

-------
                                                       APPENDIX XII


              GLOSSARY OP ACRONYMS AND ABBREVIATIONS


AA        Assistant Administrator

AOP       Automated Data Processing

AIRS      Aerometric Information Retrieval System

BBS       Bulletin Board System

CASE      Computer Aided Software Engineering

CERCLIS   Comprehensive Environmental Response, Compensation and
          Liability Information System

CFO       Chief Financial Officer

CIO       Chief Information Officer

CO        Contract Officer

DAA       Deputy Assistant Administrator

DI        Data Integration

DOPO    .  Delivery Order Project Officer

DSO       Designated Senior Official

EDI       Electronic Data Interchange

EPA       Environmental Protection Agency

FINDS     Facility Index System

FTE       Full Time Equivalency

GAO       General Accounting Office

GIS       Geographic Information Systems

GPRA      Government Performance and Results Act of 1993

ICMS      Integrated Contract Management System

IDEA      Integrated Data for Enforcement Analysis

IFMS  .    Integrated Financial Management System

                              XII-1

-------
                                                       APPENDIX XII





IMDA      Information Management/Data Administration



IRM       Information Resources Management



IRMD      Information Resources Management Division-Cincinnati



IRMPG     IRM Planning Group



LAN       Local Area Network



NDPD      National Data Processing Division



NPR       National Performance Review



OAM       Office of Acquisition Management



OARM      Office of Administration and Resources Management



OE        Office of Enforcement



OIG       Office of Inspector General



OIRM      Office of Information Resources Management



OMB       Office of Management and Budget



OPPE      Office of Policy, Planning and Evaluation



OPPTS     Office of Prevention, Pesticides, and Toxic Substances



ORME      Office of Regulatory Management and Evaluation



OSWER     Office.of Solid Waste and Emergency Response



PCS       Permit Compliance System



PCSC      Personal Computer Site Coordinator



PO        Project Officer



PWSS      Public Water Service Supply



RAD       Rapid Application Design



RCRIS     Resource Conservation and Recovery Information System



RI/FS     Remedial Investigation/Feasibility Study



RTP       Research Triangle Park



                              XII-2

-------
                                                        APPENDIX XII
SDLC      System Development Life Cycle
SEDM      State/EPA Data Management program
SIRMO     Senior Information Resource Management Official
STARS     Strategic Targeted Activities for Results System
TQM       Total Quality Management
TRIS      Toxic Release Inventory System
                              XII-3

-------
                                                      APPENDIX XIII
                       REPORT DISTRIBUTION

Office of Inspector General

     Inspector General   (2410)

    . Deputy Inspector General   (2410)

EPA Headquarters

     Administrator  (1101)

     Deputy Administrator  (1102)

     Assistant Administrator for Administration and Resources
     Management  (3101)

     Assistant Administrator for Policy, Planning, and
     Evaluation  (2111)

     Assistant Administrator for Enforcement   (2211)

     Office of General Counsel   (2310)

     Assistant Administrator for Water  (4101)

     Assistant Administrator for Solid Waste and'Emergency
     Response  (5101)

     Assistant Administrator for Air and Radiation  (6101)

     Assistant Administrator for Prevention, Pesticides and
     Toxic Substances   (7101)

     Assistant Administrator for Research and Development  (8101)

     Associate Administrator for Regional Operations &
     State/Local Relations  (1501.)

     Director, Office  of Information Resources Management  (3401)

     Agency Followup Official   (3304)
          Attn:  Director, Resource Management Division

     Agency Followup Official   (3101)

     Audit Followup Coordinator  (3102)
          Attn:  Program  & Coordination Office

                             XIII-1

-------
                                                      APPENDIX XIII



     Director, Congressional Liaison Division   (1302)

     Director, Public Liaison Division   (1702)

Regional Offices

     •Regional Administrator, Region 1

     Regional Administrator, Region 2

     Regional Administrator, Region 3

     Regional Administrator, Region 4

     Regional Administrator, Region 5

     Regional Administrator, Region 6

     Regional Administrator, Region 7

     Regional Administrator, Region 8

     Regional Administrator, Region 9

     Regional Administrator, Region 10

Research Triangle Park. North Carolina

     Director, Office of Administration and Resources
     Management (MD-2 0)

     Director, National Data Processing Division/OARM   (MD-34)
                              XIII-2

-------