#505-00*8 -
V0 ENFORCEMENT DIVISION
I U.S. ENVIRONMENTAL PROTECTION AGENCY
A FEASIBILITY STUDY OF A
POTENTIAL ADP SYSTEM FOR
THE ENFORCEMENT DIVISION
ABSTRACT
PREPARED UNDER
CONTRACT NO: 68-01-3834
NOVEMBER 20, 1978
U.S. Environmental ]*.••.,.*,„
401 v "st-e*t. o.w.* *' *
EPA
220/
1978.2
ARTHUR YOUNG & COMPANY
-------
-------
ARTHUR YOUNG X COMPANY
IO25 CONNECTICUT AVENUE N. W.
WASHINGTON. O. C 2OO36
November 20, 1978
Mr. William R. Milligan —+-1 pt-fct^11
Project Officer
Enforcement Division (EN-338) ^ v ^XT... >,, -----
Environmental Protection Agency '.^viVgton, C3 204*0
401 M Street, S.W. *.---
Washington, D.C. 20760
Reference: Contract No. 68-01-3834
Subject : Delivery of Final Report
Dear Mr. Milligan;
Arthur Young and Company is pleased to submit the Final Report
of a Feasibility Study of a Potential ADP System for the Enforcement
Division, prepared under the above referenced contract.
This Final Report has several components:
An Abstract of the Report
The Final Report of the Feasibility Study including a Manage-
ment Summary
Three Appendices which examine cost and systems impacts of
the recommended system.
The accomplishment of this study was achieved through an intensive
and thorough requirements analysis at the EPA headquarters, region/
and state levels; definition of feasible alternative systems for
the processing of Discharge Monitoring Reports; a cost and benefit
analysis resulting in the selection of a Compliance Analysis System
(CAS) to support Enforcement Division requirements; and a time-
phased implementation plan for installation of the proposed system.
This report represents the completion of our activities for the
engagement.
-------
I
I
I
I
1
I
I
1
1
I
I
ft
II
II
II
II
II
II
II
-------
Mr. William R. Milligan
Environmental Protection Agency
page (2)
November 20,1978
Should you have any questions concerning the report or the
study in general, please do not hesitate to contact Ms. Ellen DuPuy
or me at (202) 785-4747.
Very truly yours,
ARTHUR YOUNG & COMPANY
Gerald Mendenhall
Partner
-------
II
II
II
II
II
II
It
II
II
II
II
II
II
II
II
II
II
II
I
-------
ABSTRACT
-------
II
II
jl
II
II
II
it
II
II
II
II
II
II
I
I
I
I
I
I
-------
ABSTRACT
This Final Report presents recommendations for a Compliance
Analysis System (CAS) based upon a study conducted by Arthur Young &
Company for the Enforcement Division of the Office of Water
Enforcement. The recommended system concept and the analysis leading
to the recommendations are discussed in detail in this report.
The draft Final Report for the study was delivered to the
Enforcement Division in April, 1977. Subsequent to the delivery of
the draft report in April, 1977, Arthur Young & Company was tasked to
further investigate the cost and other systems impacts of CAS.
These additional analyses are documented in appendices to the Final
Report. EPA comments regarding the draft report were also reviewed
and incorporated, as appropriate, in this report.
1.
STUDY OBJECTIVES
In initiating this feasibility study, the objectives of the
Enforcement Division were to:
Increase the effectiveness of DMR violation detection and
enforcement through automated limit checking and maintenance of
violation history files
Develop improved reporting procedures to assist in the monitoring
and management of the enforcement process.
Improve the productivity of the substantial number of EPA
personnel dedicated to the manual processing of DMR's through
selective automation of these procedures.
Our analysis, presented in this report, indicates that these objectives
are attainable and, as a result, we have developed specific
recommendations for a Compliance Analysis System (CAS) to support the
Enforcement Division. Since a system to fully support Enforcement
Division requirements requires significant development effort and
time, we have proposed a long-term effort. However, in response to
near-term requirements, we proposed a basic CAS system which would
achieve significant benefits, while providing a means to easily evolve
to a full CAS capability. We understand that the Enforcement Division
has already taken significant steps in the direction of their near-
term requirements in line with these recommendations.
-------
2. COMPLIANCE ANALYSIS SYSTEM
II
The encouragement of uniform and standard procedures for
detection of violations
The system and procedures associated with Discharge Monitoring
Report violation detection and enforcement will be addressed in this ||
study as the Compliance Analysis System (CAS). Based upon the stated ||
objectives, the design of a complete CAS must meet the following
criteria: _.
. The capability to process and detect violations from DMRs ™
for all permittees
The capability of processing DMR forms in a timely manner ||
it
The efficient use of personnel resources necessary to -.
directly maintain and operate the system II
The ability to provide information to assist in making the
best possible use of enforcement staff resources. II
In the analysis, five levels of automation were identified for
CAS based on the current processing procedures and design criteria. ||
As presented in Exhibit 1-1, these are: ||
System Automation Option 1 - automation of the |.
facility/permit identification and DMR schedule information II
to enable automated tracking of DMRs due to be received. «
This option represents the status at which most Regions
exist today. II
. System Automation Option 2 - automation of the effluent
limitation data to enable automatic identification of DMRs m
that are delinquent, incomplete or outside permit limits p
System Automation Option 3 - automation of the violation _
history file to enable generation of violation reports and •
automatic identification of chronic violators ™
System Automation Option 4 - automation of the exceedent •
criteria to enable automatic identification of the severity j§
of each violation and provide uniform criteria for
determining severity of violations *
System Automation Option 5 - automation of agency response
information to enable identification of outstanding —
violation actions and the generation of a non-compliance •
DMR report which includes actions taken against permittees. ™
3. SUMMARY OF ANALYSIS
I
The analysis of the feasibility and desirability of automating
the processing of DMR forms must consider in detail the technical, £
I
I
-------
EXHIBIT 1-1
52
I*
^
TL
>iftt
3S
; 51
yy n /
O
P
111
II
3 3
11
IS :
-------
operational and cost impacts of these options and various
implementation alternatives. We present in Sections II, IV, V and VI
of the Final Report detailed consideration of these issues in
developing a CAS system. In particular, we have considered:
Cost - the costs of both implementation and operation of
the alternative CAS concepts
Implemenatation Time - the time phasing required to
successfully implement the system
Responsiveness - the ability of the system to meet the needs
of each of the organizations with responsibility for
enforcing the NPDES program.
Because of the immediate need for an improved processing capability,
we have focused attention on both what can be practically achieved in
the short run and what is desirable over the long-term. The basic
decisions in this analysis include:
(1) Processing All DMRs or Only Majors
There presently exist approximately 50,000 permits and
permit applications in the EPA Regions and NPDES states. Of these
permits, only about 8,000 or approximately 16% are classified as
majors. Although the ultimate goal of any DMR processing system
is the timely and uniform enforcement of all DMRs, the emphasis
in the near-term must be the majors. The effort required to
implement the effluent limitation data on majors and minors would
be substantially more than that necessary to implement only
majors, and would involve susbstantially greater risk of system
development failure. As a result in the near term, the
implementation of an effluent limitation file containing only
majors would substantially assist in monitoring the major sources
of pollution.
(2) Central vs. Regional Data Entry Procesing
The issue of Central vs. Regional data entry must be
evaluated on both cost and the management issues of responsibility
for the source data. The EPA Regions and NPDES States presently
issue the permits and .have responsibility for maintaining the
permits on site to support enforcement. It is therefore desirable
to also have the regions input and maintain the data files which
reflect information taken from the permits: facility
descriptions, DMR schedules, and permit limitations.
For DMR data entry and error corrections, the Regions are
the most familiar with the individual permittee, and their
discharge history and problems. Further, because the DMR forms
are legal documents, they must be maintained at the point where
legal action is initiated. The recent trend toward Regional
I
I
I
I
I
I
I
i
i
i
i
i
i
i
i
i
i
i
i
-------
minicomputers and their use in intelligent data entry further
encourages the Regional data entry approach.
(3) Data Entry Methods
The analysis of alternative data entry techniques considered
commercial and in-house (governmental) for the following
techniques: keypunch, keybatch, OCR and Mark Sense. Both the
OCR and Mark Sense approaches were discarded due to cost or
operational disadvantages. Keypunch, both in-house and
commercial, costs about twice as much as keybatch for the same
service. The cost of commercial and governmental keybatch were
very close. However, taking into consideration OMB Circular A-
76 and the requirement to hire keybatch operators for in-house
government operation, we recommend the commercial keybatch method.
In addition to direct cost advantages, commercial keybatch can
best handle the peaks and lows in DMR receipt, provide for editing
of data fields at the time the data is entered, and uses output
media which is easily transported, handled, and stored.
(4} Use of Existing EPA Systems
Due to the immediate need for a DMR processing system and
concern for costs, the analysis considered the potential for use
of existing EPA systems for the near term. Two existing
information systems were considered in our analysis; the Permit
Compliance System (PCS) and STORET.
Some major information categories required by CAS are
presently contained within PCS: facility identification and DMR
schedule information. In addition, PCS supports the DMR forecast
function by tracking DMRs that are due to be submitted. The use
of PCS therefore would enable the basic system to be implemented
much earlier than if these functions had to be programmed from
scratch.
The other alternative considered STORET, which is a special-
purpose data base system designed for supporting water quality
information applications. STORET at present contains effluent
limitation data on two State-enforced NPDES programs and supports
a program which can identify effluent limitation violators.
STORET does not support at present the tracking of DMR submittals
or the action taken against violators. STORET provides a strong
data retrieval and manipulation language.
(5) Degree of Automation
The progressive levels of automation considered provide
increased capability and benefits while incurring more
development time and higher costs. System Option 4
provides a level of automation which can be implemented in the
near term (1 to 1-1/2 years) while providing significant benefits
-------
I
to the Regions. This option would provide automated initial
screening of DMRs, basic screening to identify severity of each •
violation, and reporting of violation histories. It would further g
provide substantial benefits by enabling the redirection of the
considerable manual effort presently required to manually track _
DMR receipts,, screen them for completion, determine whether the •
permittee exceeded his limits and identify the severity of the *
violation. Finally, this degree of automation should enable
processing of basic DMR data within about 10-20 days from time •
of receipt, versus 90 days currently in some instances. J
(6) Conventional File Management vs. Data Base Management •
Systems (DBMS)J
The evaluation of whether to implement data files under _
conventional file management vs. a DBMS must consider: time to •
implement, cost, skills of EPA personnel, and complexity of data *
relationships. Although the use of a DBMS typically reduces
programming time and provides a more stable and flexible system, •
we recommend that a DBMS based system not be attempted in any m
near-term system. PCS, which could serve as a basis for a near-
term CAS system, presently uses conventional files. The •
conversion of these files to DBMS and required program I
modifications would require substantial time.
However, the advantages of a DBMS environment make it I
desirable for any long-term CAS system. The DBMS environment *
would provide for easier file maintenance, a reduction in data
redundancy and more efficient processing. These considerations •
will be particularly important if both major and minor permits m
are implemented in the system.
(7) Coding of Effluent Limitations
I
In order to implement the Compliance Analysis System, _
effluent limitation data must be coded, keyed, and entered onto •
the automated system. The transcription of this data from NPDES ~
permits onto coding sheets for subsequent keying requires 1.3
hours per permit based upon Enforcement Division estimates. •
Because there are approximately 50,000 NPDES permits presently Q
issued, these coding activities would require a significant effort
to complete. Based upon these conservative assumptions, the •
conversion estimates would be approximately 31 person-years and J
a cost of $1,430,000.
Because this has a significant impact on the decision to I
implement CAS, alternatives which have the potential for *
substantial cost savings should be considered. These
opportunities for improvement are: •
Initially key effluent limits reported from the discharges
directly from the Discharge Monitoring Reports and institute m
I
I
-------
4.
sampling procedures for verifying the accuracy of the
reported effluent limitations over time. As permits are
issued, reissued, or modified, effluent limitations could be
coded and entered onto CAS.
Enable states that wish to use CAS to code effluent
limitations themselves for all state issued permits.
Utilize effluent limitation data directly from regions that
already have the data in machine-readable form.
We believe that by implementing the above alternatives, singly
or in combination, the time and cost to EPA of entering effluent
limitation data can be substantially reduced.
RECOMMENDATIONS
Based upon our analysis, we find that automation of the DMR process
is feasible, however, we recommend a Compliance Analysis System to be
implemented in two phases. There is a clear need for immediate support
in processing of DMRs. A hasty attempt to meet this need and meet
longer term CAS requirements could involve substantial risks in
implementation. As a result, we have recommended a basic system to
address the most pressing needs and a longer term effort to evolve
this basic system to meet the full requirements.
(1) Near-Term Approach (Phase1)
Exhibit 1-2, on the following page, represents a conceptual
design which we recommend be implemented in Phase 1. To meet the
immediate need for supporting DMR processing, the Phase 1 CAS
system design would consist of the following components:
The PCS data file should be used as the source of facility
identification and DMR schedule data. The current PCS
programs should be used for maintenance of this data.
An Effluent Limitation File should be developed for only
major permittees using conventional file management
techniques.
The DMR forms should be automatically preprinted and mailed
to the permittees once a year. More frequent issuance would
incur additional cost for processing and mailing, while less
frequent issuance would require additional cost for
generating new forms when permits are modified.
The system should provide for automated screening of major
DMRs which are delinquent, incomplete or in violation.
The implementation should include a national exceedent
criteria routine which can assist in analyzing the
priorities for violation enforcement investigations.
-------
Q.
LU
u
O
u
2
at
to
w
<
u
Ul
1
0.
ii
U <9
II
EXHIBIT 1-2
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
. The system should provide for creation and maintenance of
a Violation History File.
This capability reflects system Option 4, discussed earlier, and
will provide the following benefits:
The automated initial screening of major DMRs which could
redirect an estimated 15 to 25 man-years of effort toward
processing of minors or other enforcement activities
The automated substantative screening of DMRs
The reduction of the time to screen DMRs to 10-20 work days
The automated identification of chronic violators.
The cost to fully implement the recommended Phase 1 system
is estimated to be $409,000 and could be implemented in 12 months.
The estimated annual direct cost of operating this system would
be $351,000. Exhibit 1-3 displays the cost associated with this
system while Exhibit 1-4 depicts the implementation plan and the
required resources needed to implement it. In addition to these
direct resource requirements, we anticipate the requirement for
2-3 management/analyst personnel at EPA headquarters to support
the implementation and operation of CAS.
(2) Long-Term Approach (Phase 2 - Majors and Minors)
Exhibit 1-5, on the following page, presents a conceptual
overview of the Phase 2 system. To meet the long-term needs of
the Enforcement Division, we recommend that the Phase 2 CAS system
design consist of the following components:
The provision for screening of all DMRs including minors.
This will require the coding of all minor effluent limits
and entering them onto the data base.
The implementation of a true data base under the control of
one of the recognized DBMS. This will require the
development of a data structure, and conversion of existing
data files to the data base.
The implementation of the action tracking capability under
the DBMS.
This capability reflects system Option 5, discussed earlier, and
will provide the following benefits over the Phase 1 approach:
Efficient processing of both major and minor DMRs
Automated action tracking
-------
EXHIBIT 1-3
—
3
Z
Z
ta
CA
8
o
z
IZ
OPERA'
9NISS3I
tj
0
cc
CN
o
CM
00
LU
o
<
cc
O
to
<
Q
•
§o
0
00 C
m" so
to
LU
U
cc
s
LU
CC
COMPUTER
ING
^
o
a"
-
< ENTRY
's
H
.•*
o o 51
o S qj
o CN r: i
^3 *^ m ;
LU
U
<^
Z
LU
Z
s
L? ^ ^
> 0 0
2
»
>
a>
eo
o
O
CM
«
to
i
Ui
O
I
2
cc
O
O
cc
NG
<
cc
a
U)
M 2
LU - CC
D H S
a z o
S 3 §
§ u. t
O O 2
= 9.5
8
en
LU
a
to
LU
CO
(O
cc
H
LU
0
cn
LU
U
cc
i
LU
cc
cc
a.
1
u
LU
a"
a
LU
CO
O
LU
Z
LU
i
? o
J
LU
CC
Z
LU
Z
CC
o
C3
u.
O
Vi
V)
8
LU
Q
3
U
2
O
2
W
LU
8
I
11
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
-------
EXHIBIT i-4
•o
0J
^ 5:
^^_ nj
< i
Z "5.
< £
iu
o
<
!3
o.
O
u
o
'i
o
CO
cs
Person-Month
Requirements
CM
CM
S
0
M
O £
< 5
5 s
1-
•»
en
CM
«-
IMPLEMENTATION
MOO
, . . a - 3
q
1
•
f ?
r [:
1 i
9 1
§4
1
1
I1
4
m
< t-
«fl rf _ (O uj j
s ! 1 5 a 5 |
" = a s s i 1
i s 1 1 I s S
Z Z H *» O S
8uj M 3 ae y H P-
3 - 1 t < 2 |
§£=!
CUxa.OxKv3
^ojco^iiitdr^cb
TOTAL 95.5
-------
EXHIBIT I-
Q.
LU
O
LU
o
Ui
I
a.
i I
H
s
• o
Ii
I
Si
g,
Ib
3
ACTION
1 HACKING
MODULE
II.
;s|
««
11
£ ^
«
!
it
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
Significantly enhanced capability for management reporting
and ability to respond to information and analysis requests.
The Phase II CAS system approach will cost an additional $1,475,000
to implement and take 14 months to implement after the Phase 1
design has been implemented. The coding of effluent limits for
minors will require approximately 26 person-years of effort, and
thus will probably not be completed within 14 months. Nonetheless,
the system could be implemented without all minor being processed.
The Phase 2 system will cost $615,000 annually to operate. Exhibit
1-6 presents the cost of the Phase 2 system while Exhibit 1-7
presents the implementation plan and resource requirements.
(3) STORET Interfaces
As a result of our analysis, we recommend that interfaces
be built between CAS and STORET. Rationale for a CAS to STORET
interface include the following:
To allow utilization of STORET's graphics software for
displaying actual DMR effluent data
To facilitate combined analysis of DMR effluent data in CAS
and ambient water quality data (as well as other compliance
data) within STORET
The functions to be peformed by the interface are:
Transmit, on a routine basis, new DMR effluent data from CAS
files as well as newly issued permits, modified permits, and
reissued permits information
Write the data to the STORET data base using STORET's file
access techniques and make this data available for analysis
Remove detailed effluent data from CAS once the transfer
operation is successfully completed, except for use in
compliance history analysis.
The delineation of DMR processing responsibilities between the
two systems would be:
Functions performed by CAS
. Data entry and editing of NPDES permit condition data
and DMR data
DMR preprinting and receipt tracking
DMR screening to enable automatic identification of
DMRs that are incomplete, outside permit conditions, or
out of compliance with exceedent criteria
-------
EXHIBIT 1-6
03
CO —
O IB
w S
— LU
<
31
3.
•3
3
j
<
s
z
z
<
Cfl
1—
M
8
0
LU
/n
S1 <
H
<
ec
LU
0*
C
0
(J W
Z <
« t;
to <
LU C
O
£
S
S
o
t-»
C/5
HI
U
c:
1
L^
e
jj
LU
H
COMPU
•
a
S
CN
fM
a
z
j
<
i_3
CN
6
CN
>
o
CO LL
0 :
n
-------
EXHIBIT 1-7
9
"5
•a
S 5
UJ O
h- CO
w 1
^"*
w ®
en g
5 I
< °
1 I
S i
i 1
^ a
| 0
O =
O v
0)
(0
a.
— *2
1 §
i £
• S"
c .b
O 3
i i"
a. cc
Z
O
l_
IMPLEMEI
S
cc
u
H
cc
UJ
Z
a
z
o
_J
o
u.
MONTHS
M
«N
^
£
e
05
a
-
CO
U)
tr
CO
fs
-
c/j
a.
UJ
to
z
o
<
1-
ai
S
HI
_i
a.
5
CM uT> fO CN OT dDuO^^TtA
^*
n
IB
i
H
i
, , P
Ii
I
w t u>
z < - «
g H § 3 £
s g i | 2 a
S S £ | „ 5 S
i i § S i i . i b is
I t g i i i I i § §
SUJiCQ>-«o2riG
KH?<«Ooa."iu:
w = a:iDJt
>Siru^«"a.Q.OT-;5
wCHaHaMja.;; w
5S^z£§^3Q^J^ zQS>Z*~
uj o >oinQuJiiJZJ>
ceo ujQu
-------
I
5. SUMMARY
- Monitoring of enforcement actions and violation history
Functions performed by STORET •
Archiving of actual DMR effluent data
Generation of graphic representations of DMR effluent •
data and NPDES permit condition data
Combined analysis and presentation of STORET water |
quality data, and DMR effluent data and permit
conditiions data. _
Based upon our investigation of a potential ADP system for the I
Enforcement Division we find that: m
The operating costs for CAS Phase 1 are estimated to be 22% •
under the current DMR operating costs p
The operating costs for CAS Phase 2 are estimated to be 22% •
above the current DMR operating costs I
There are significant economies of scale in DMR processing
indicated in our Region II/Region IV comparison (Appendix •
A) •
An estimated savings of 15-25 people is anticipated with
the implementation of CAS Phase 1.
The above findings bring us to the following conclusions: •
There will be a significant substitution of computer
resource and data entry costs for staff costs in the
transition from the current operating environment to either •
(Phase 1 or 2) CAS environment. That is, even though 15 to •
25 people would be able to be directed to other enforcement
activities, there would not be a corresponding cost savings. •
In fact, an 22% increase in operating costs is estimated if |
CAS Phase 2 is implemented.
Management information for assessing program effectiveness •
and operational efficiency under an automated system would "
be much more available than under total manual systems such
as exist in many regions. I
These conclusions lead us to recognize that the decision to
implement CAS is much more a function of the value of the capability •
than a function of dollars saved. If CAS Phase 1 is implemented |
centrally, which would keep maintenance and operating costs to a
minimum, some savings in the form of re-directed resources may be •
I
I
-------
realized. However, the major decisions is whether the capability
provided by CAS can justify the initial investment and ongoing costs
of the system.
Because of the substantial additional capabilities that CAS will
provide, and the quality of management information available from CAS,
it is recommended that the Compliance Analysis System be implemented.
The recommended time phase approach will require a substantial
investment on the part of the Enforcement Division. The projected
direct cost impact is shown for the next six years (FY 1979 - 1984}
in Exhibit 1-8. In addition to those direct costs, the Division will
require an internal headquarters support staff of 2 to 3 management
analysis to implement and operate CAS beyond that currently required
to support PCS. We believe that the benefits identified in this report
are substantial and justify this investment.
10
-------
EXHIBIT 1-8
ENFORCEMENT SYSTEM COSTS OVER SYSTEM HORIZON
PHASE 1
IMPLEMENTATION
OPERATION
PHASE II
IMPLEMENTATION
OPERATION
TOTAL
FY-79
409,280
-
-
-
409.280
FY -80
-
351,074
558.440
909.514
FY '81
-
175,000
499,444
465,308
1.140.252
FY-82
-
-
416.850
614.739
1.031.589
FY'83
-
-
-
676,213
676,213
FY'84
-
-
-
743,834
743,834
------- |