OfFKEOF
ENVIRONMENTAL
INFORMATION
Environmental Information Integration
Requirements
Technical Evaluation of Alternatives
Development Plan
Revised January 30, 2007
-------
Table of Contents
Executive Summary 1
1. Requirements for Environmental Information Integration 4
1.1. Methodology 4
1.2. Findings 5
1.2.1. The strategic context for OEI environmental information services 5
1.2.2. What interviewees told us 7
1.2.3. Supply-driven versus demand-driven information management 9
1.3. Recommendation 11
2. Technical Evaluation of Information Integration Alternatives 12
2.1. Services-oriented Architecture (SOA) 12
2.2. Data Warehousing 15
2.3. Enterprise Architecture 16
2.4. IT Governance 17
2.5. Program Management Office (PMO) 19
3. Development Plan for Information Integration 21
3.1. Option 1: Maintain Status Quo 21
3.2. Option 2: Improve Governance 22
3.3. Option 3: Governance with Top-Down Information Needs 23
3.4. Option 4: Governance with Bottom-Up Information Services 24
3.5. Recommendation 24
3.6. Next Steps 25
Appendix A Interviewees 27
Appendix B Information Needs Areas 29
Appendix C Documents Reviewed 34
-------
Executive Summary
EPA's Office oflnformation Analysis and Access (OIAA) within the Office of Environmental
Information (OEI) engaged DecisionPath Consulting to identify the business requirements for integrated
environmental information and make recommendations for how OEI can more effectively manage
information, better serve its customers, and further its mission. To accomplish this effort, we interviewed
39 EPA employees, primarily from OEI, and reviewed a number of documents about EPA's mission,
strategic plan, management challenges, enterprise architecture, and related business and IT topics.
Major Findings
Both the EPA employees with whom we spoke and the documents we reviewed depict OEI as an
organization at a crossroad. Since its inception in 1999, OEI has focused on three major services. The
first service is IT infrastructure and processing for the agency. The second is to implement a common
mechanism for data intake from EPA partners - the National Environmental Exchange Network and its
supporting structures, such as the System of Registries. The third is a supply-driven approach to
providing information to both internal customers and the public. OEI must continue these efforts.
However, in order to make a larger contribution to the agency and thereby ensure future funding, it must
provide additional customer-oriented information services. Both interviewee comments and reviewed
documents indicate that
Neither OEI's internal customers nor its employees have a clear understanding of its future
direction.
OEI has no mechanism by which it engages with its customers to understand their information
requirements and define information services that have value to them.
OIAA collects data mainly to satisfy statutory requirements. It then uses a supply-driven
approach to leverage that data for additional purposes and for additional customers. This supply-
driven approach, which entails making the collected data available via self-service query tools,
satisfies the needs of some customers. However, it does not adequately fulfill specific
requirements for strategic business information.
IT governance has focused on satisfying the requirements of environmental statutes and on 1T-
centric activities such as enterprise architecture. OEI lacks an effective IT governance
mechanism to help it prioritize and fund IT investments based on business value.
OEI Strategic Context
Since its formation in 1970, the Environmental Protection Agency (EPA) has had a decentralized
organization of media-specific program offices and geographic regions. Political appointees head both
the program offices and regions. Each program office has its own appropriations and its own portfolio of
environmental statutes to execute. Historically, each program office and region had its own IT
department that developed and managed its own information systems. In 1999, many but not all IT
functions were centralized in a new organization, the Office of Environmental Information (OEI).
To-date, OEI has focused on providing IT infrastructure and operations, and on implementing the
National Environmental Exchange Network (and supporting structures, such as the System of Registries)
as a common data intake mechanism by which EPA partners submit statutorily-required data to EPA.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 1
-------
These efforts support the collection and processing of data for its primary purpose: satisfying the
requirements of environmental statutes.
Leveraging the collected data for additional purposes and to serve additional customers is an OEI
challenge that falls primarily to OIAA. Examples of these additional purposes and customers include
assessing the state of the environment, measuring environmental progress, and providing environmental
information to the public. OIAA uses a supply-driven approach to information management: the data
collected to satisfy statutory requirements is made available for additional purposes and to additional
users largely via the self-service mechanism of simple query tools.
EPA's 2006-2011 Strategic Plan articulates what the agency will do over the next five years. The
guidance it provides regarding what EPA considers important represents an opportunity for OEI to focus
beyond its current services to provide high-value additional information services to the agency.
IT investments at EPA are governed by the requirements of the various environmental statutes EPA
executes, as well as by external mechanisms such as the Clinger-Cohen Act of 1996, which mandated a
CIO, the Federal Enterprise Architecture, and the Capital Planning and Investment Control (CPIC)
process. OEI has been working for several years to construct and implement an enterprise architecture
(EA) for EPA. As part of the EA, OEI intends to base EPA's application architecture upon Web services
and services-oriented architecture (SOA).
What Interviewees Have Told Us
We interviewed 39 EPA employees, two-thirds of whom were from OEI and the rest from program
offices, regional offices, and administrative offices. From OEI employees, non-OEI employees, and
reviewed documents, consistent perspectives came to light about OEI, how it engages with and serves its
customers, data integration, information management, IT governance, and other topics. Three key points
emerged:
Beyond the IT infrastructure that it provides, OEI's value proposition to the agency is not well
understood by either its customers in the program offices and regions or by OEI employees. In
order to move forward, OEI must clearly articulate how it supports customer information needs
and it must communicate its roadmap for providing future value-added information services to
those customers.
The fundamental purpose of OEI's activities (and, in fact, of all EPA activities) is to support the
agency mission to protect human health and the environment. OEI does not currently have an
effective governance structure that evaluates and recommends potential information service
investments based on their relative contribution to supporting that mission.
OEI has a concept of providing a portfolio of Web services using SOA, but it lacks an effective
process by which to define, prioritize, design, and govern such a portfolio.
Recommendations and Next Steps
OEI has a vital mission: the creation, management, and use of information as a strategic asset at EPA. The
road to an expanded OEI future and achievement of that mission is clear: in addition to continuing its
current services (IT infrastructure and processing, ongoing work to improve data intake from partners,
and information delivery via a supply-driven approach), OEI must engage with its customers to
understand their business requirements for information and then deliver information services that target
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 2
-------
those requirements. DecisionPath characterizes such an approach as demand-driven: business
requirements "pull" the creation of information services targeted to fulfill those requirements.
DecisionPath evaluated four options for whether and how OEI could adopt the demand-driven approach
to information management and more effectively govern its activities in order to deliver additional
business value to the agency. (See section 3 for a discussion of the four options and the advantages and
disadvantages of each.) Because there is no precedent for wholesale adoption of the demand-driven
model, we recommend that OEI take an incremental approach to demand-driven information services by
beginning with a prototype.
Specifically, we recommend that OEI take the following actions:
Demonstrate the demand-driven approach though a carefully-selected prototype with a single
EPA component. The first step is to select an appropriate project (with a receptive EPA
functional organization) for the demand-driven prototype.
Expand the IT governance role of the OEI program management office (PMO) to include
development of a multi-year program plan of IT investments, based on a portfolio of investment
opportunities; identification (or collection) of the investment opportunities that make up the
portfolio, including specification of the potential business value and implementation risks of
each; and monitoring approved and completed projects for achievement of their projected
business value.
Apply the demand-driven approach to public information services to deliver supply-driven
content with demand-driven information presentation. Develop a program plan for a multi-year
effort to extend and enhance the public's access to environmental information.
Section 3.6 outlines a series of next steps for OEI to begin to execute these actions. They are the first step
toward OEI playing a larger role within EPA and assuring future funding by providing additional
information services and incremental business value.
January 30,2007 Environmental Information Integration - Final Deliverable (revised) Page 3
-------
The Statement of Work for this engagement requires DecisionPath to produce three final deliverables:
requirements for environmental information integration, technical evaluation of information integration
alternatives, and a development plan. This consolidated document contains a section for each of the three
deliverables.
The Requirements section describes the methodology we used, the strategic context for OEI services, a
summary of what interviewees told us, and compares OEI's current supply-driven approach to
information management with our recommended demand-driven approach.
The Technical Evaluation section discusses a number of data architectures, data integration approaches,
and IT management techniques, specifically: services-oriented architecture, data warehousing, enterprise
architecture, IT governance, and program management office. It describes how OEI currently applies
these architectures, approaches, and techniques, how it might extend their use, and situations in which
they are and are not appropriate.
The Development Plan section considers four options for how OEI might move forward, including the
advantages and disadvantages of each, and recommends the option DecisionPath believes OEI should
pursue.
1. Requirements for Environmental Information
Integration
1.1. Methodology
DecisionPath consultants interviewed 39 EPA employees about the business requirements for integrated
information, current methods of integrating data, and related topics. These employees were primarily
from OEI: they ranged from technical staff to the Acting CIO. Appendix A contains a list of employees
interviewed.
In addition to these requirements-gathering interviews, we reviewed over 100 EPA documents, including:
2006-20J1 EPA Strategic Plan: Charting Our Course
EPA 's FY2005 Performance and Accountability Report
EPA Draft Report on the Environment
Other pertinent documents
Appendix C provides a complete list of the documents we reviewed.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 4
-------
1.2. Findings
1.2.1. The strategic context for OEI environmental information
services
Environmental Protection Agency (EPA)
EPA was created in December 1970 from pieces of numerous federal agencies, including the Departments
of the Interior; Agriculture; Health, Education, and Welfare (now HHS); the U.S. Atomic Energy
Commission; and others. EPA has always been organized by program (media). Each program office has
always had its own set of environmental statutes to execute and its own appropriations. The 10 EPA
regional offices have also existed from the beginning. Both the assistant administrators in charge of the
program offices and the regional administrators are political appointees.
Office of Environmental Information (OEO
OEI was formed in October 1999 by centralizing various IT functions. Requirements of the Clinger-
Cohen Act of 1996 and other legislation were motivators for this centralization. The formation of OEI
might also have been partially in response to critical reports about EPA IT issues (ineffective information
management, lack of data integration, poor data quality, etc.) by GAO and other oversight entities.
EPA Strategic Plan
2006-2011 EPA Strategic Plan: Charting Our Course describes how EPA will accomplish its mission to
protect human health and the environment. It outlines five goals, each with a number of supporting
objectives. Four of the five goals are media-specific, but the Strategic Plan also contains a number of
cross-goal strategies. The cross-goal strategy "Results and Accountability" provides strategic context to
OEI in the following areas:
Assessing the state of the environment and measuring progress.
Making information more accessible. EPA will focus on four major areas:
Analytical capacity
Governance
Excellence in information service delivery
Innovation in information management
Integrating budget and performance information.
Key Management Challenges
Each year, the EPA Inspector General submits to the Administrator a memorandum outlining the agency's
key management challenges. Two of the 2006 key management challenges deal with data: (1) data gaps,
and (2) data standards and data quality. To quote the memorandum, "If EPA is to manage for results, it
needs to decide what environmental and other indicators will be measured; provide data standards so that
organization responsible for delivering environmental programs are measuring what is important and are
using common definitions; and ensure that data are of sufficient quality for effective decision making."
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 5
-------
Another key management challenge is "EPA's Use of Assistance Agreements to Accomplish its
Mission." EPA spends more than half of budget on assistance agreements, which "are a primary means
EPA uses to carry out its mission." The assistance agreements are with the partners (states, tribes, and
others) to whom EPA has delegated the execution of its programs. With so much of its work done by
these partners, their participation with EPA in data quality is critical.
EPA will spend approximately S600 million per year for systems development and maintenance in both
FY2006 and FY2007. This spending comprises many systems and projects, ranging in size from less than
$100,000 to $25 million. Every year since 1996, OIG has listed "Information Resources Management" or
a variant as a key management challenge. Despite all the IT governance structures and mechanisms
mandated by the Clinger-Cohen Act, effective governance of IT investments to realize business value
remains a challenge for EPA.
Research Orientation
A significant portion of EPA employees are scientists, and EPA is a nexus for environmental research.
This cultural orientation toward scientific research influences the EPA approach to providing information.
Typically, a scientific researcher looking for information:
Is well educated and computer literate.
Obtains his or her information personally (that is, does not have a support staff, except perhaps
junior researchers responsible for their own parts of the project).
Has expert knowledge of the data he or she is interpreting.
Has at least a general understanding of the context surrounding the data.
Wants to see all available data on a topic; does not want the data filtered.
The scientific researcher as an information consumer is well served by a self-service information delivery
model in which all data is available via simple query tools.
Services-oriented Architecture
OEI uses services-oriented architecture (SOA) for the National Environmental Exchange Network, and
the Acting CIO desires to use SOA as the fundamental mechanism for provision of all information
services.
Summary of Strategic Context for OEI Environmental Information Services
The strategic context in which OEI operates both provides direction and constrains what it can do. As
detailed above, this context includes:
EPA culture of stovepiped media-specific program offices, with power and funding decentralized
in the program offices and regions
Cross-goal strategy to assess the state of the environment and measure progress
Cross-goal strategy to make information more accessible
Cross-goal strategy to achieve budget and performance integration
Key management challenge to improve data quality
Key management challenge to work more effectively with partners
Key management challenge to more effectively govern information management
Cultural orientation toward a scientific researcher model for information delivery
Desire to use services-oriented architecture (SOA) as extensively as possible
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 6
-------
1.2.2. What interviewees told us
DecisionPath's initial intention in the requirements-gathering interviews we conducted was to understand
and document the business case for data integration at EPA. Over the course of the engagement, our
emphasis shifted more toward IT governance. The EPA employees we interviewed spoke on a variety of
IT-related topics, including:
The degree of OEI's engagement with its internal customers about their information needs
Data quality, data standards, data stewardship, and metadata management
Data integration - how much there is, and how much is needed
From our interviews, we captured a number of quotes that provide a picture of the current EPA situation
as perceived by the employees we interviewed. The comments below are paraphrased, and are not
identified by specific employee.
Comments about OEI's level of engagement with its customers:
OEI leadership should engage their customers.
OEI is far from the business.
If [OEI] wants to know what decision makers need, they should talk to them, not to data providers.
OEI should participate more actively in [program office's] business.
[EPA manager] questions how well OEI understands what is currently going on at EPA.
Even where data is available and unambiguous, it is usually not wanted in a standalone form.
Politicians and scientists want the context.
Comments about data quality, data standards, data stewardship, and metadata management:
In order to get the information we need, the most critical problem is the quality of the existing data.
The main challenge today is data quality.
Data quality assessment is vital.
EPA does not do data stewardship well.
There is a huge need for metadata.
There are still a lot of data gaps.
Adequate data standards exist but their implementation is inconsistent.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 7
-------
Comments about data integration, or the lack thereof:
The key is to make all the information come together at the right place and time.
EPA can't tell the public as much as it wants to because it lacks the means to add up the data.
EPA uses a lot of manual effort to do integration that should be automated.
The data doesn't have to be integrated, but it does have to be integratable.
Comments about the problems associated with not having integrated data:
For Katrina, the process of moving analytical data from the collection point in the Meld to EPA
headquarters was very slow.
The agency could unite around problems faster with better information sharing.
The EPA has so much data and shares so little of it.
Business Questions and Information Needs Areas
DecisionPath began this engagement anticipating that our requirements-gathering effort would entail
interviewing primarily information consumers - that is, representatives from the program offices and
regions. As it turned out, most of the EPA employees we interviewed were from OEI. Typically, our
requirements-gathering process yields a list of business questions detailing the information the business
users and decision makers want to be able to answer. We then group like business questions into
information needs areas. Because OEI uses a supply-driven approach to information management (see
Section 1.2.3 for an explanation of the supply- and demand-driven approaches), it is not accustomed to
collecting business requirements for information and responding with systems to provide that
information. Nevertheless, our interviews did yield a number of business questions, which we grouped
into five information needs areas. These business questions and information needs areas are an example
of the demand-driven approach to information management. See Appendix B for the information needs
areas and the business questions that comprise them.
Because these business questions came from IT people relating what information they believe the
business users want, they should be validated directly with business users before being acted upon.
Summary of Interviewee Comments
The input we obtained from interviewees and the documents we reviewed coalesced around several key
points:
Beyond the IT infrastructure that it provides, OEI's value proposition to the agency is not well
understood by either its customers in the program offices and regions or by OEI employees. In
order to move forward and be more successful, OEI must clearly articulate how it supports
customer information needs and communicate its roadmap for providing future value-added
information services to those customers.
OEI has a concept of providing a portfolio of Web services using SOA, but it lacks an effective
process by which to define, prioritize, design, and govern such a portfolio.
January 30,2007 Environmental Information Integration - Final Deliverable (revised) Page 8
-------
The fundamental purpose of OEI's activities (and, in fact, of all EPA activities) is to support the
agency mission to protect human health and the environment. OEI does not currently have an
effective governance structure that evaluates and recommends potential information service
investments based on their relative contribution to supporting that mission.
1.2.3. Supply-driven versus demand-driven information management
The Current Supply-driven Approach
OEI's approach to providing information services can be characterized as supply-driven. The data
required by the various environmental statures is collected from EPA partners via the Exchange Network
and stored in program-specific databases and systems. These program-specific databases and systems
fulfill the statutory reporting requirements. Although some of these systems run on OEI's IT
infrastructure, for the most part, they are developed, maintained, and managed by program office and
regional IT organizations rather than OEI.
The Office of Information Analysis and Access (OIAA) within OEI is responsible for enabling the use of
this data for additional purposes, especially analytical and cross-media purposes, and by additional
audiences, such as the public. OIAA's approach to enabling these additional uses is based on the
scientific researcher model described earlier in section 1.2.1. It essentially provides data under the motto
"Here's the data we have: come get what you need."
Events in the larger political environment within which EPA operates can cause frequent and rapid
changes in information needs and priorities; the supply-drive approach to information services has the
advantage of being very flexible. The data is made available to the user in fairly raw form (as opposed to
being structured for a specific analytical purpose), and the user structures it according to his or her needs.
That's the downside of the supply-driven model. The user must
Obtain the raw data via a query.
Integrate it.
Interpret it and add context to it, in order to turn the data into useful information.
The supply-driven model might serve the needs of a researcher performing pure science, but much of
what EPA does consists, instead, of business operations. For example, determining how best to allocate
EPA's finite resources to achieve the maximum environmental benefit is an analytical business process,
not pure science. Many such analytical business processes take place within individual programs, within
program offices, within regions, and at headquarters for the agency as a whole. Business operations,
especially analytical and decision-making operations, are not well served by the supply-driven model.
The Demand-driven Approach
An alternative to the supply-driven approach can be characterized as demand-driven. In this approach,
data is transformed into information and packaged and presented for a specific mission support, policy-
making, decision-making, or some other business purpose. The demand-driven approach to information
management is driven by the business's requirements for specific information.
January 30,2007 Environmental Information Integration - Final Deliverable (revised) Page 9
-------
In EPA's case, the agency mission and strategic plan provide multiple instances of high priority demands
for information. The need to assess the environment and measure progress, the needs of agency decision
makers for specific analyses, the need to improve data quality, and the need to integrate budget and
performance all drive specific demands for information services.
A key principle of the demand-driven approach is "aim high:" to focus on the mission, strategic
objectives, and other high-level definitions of organizational success, and to support their achievement
with specifically targeted information.
Contrasting the Supply-driven and Demand-driven Approaches
The supply-driven approach to information management has the advantages of being simple, low cost,
and flexible. It realizes these advantages by being user self-service-oriented: the user does his or her own
querying, integration, data transformation, addition of business context, and interpretation of the data.
The supply-driven approach is well suited to a class of users that is exemplified by the scientific
researcher. It always will have a role at EPA.
The self-service and ad hoc nature of the supply-driven approach also is its weakness. It is much less well
suited for situations in which
The requirements are well structured and recurring.
Individual users do not possess the background, experience, or business understanding to add
context to the data; therefore, that context must be added for them.
It is critical that multiple users all receive the information. (This often is referred to as "a single
version of the truth.")
The required (historical) data no longer resides in the operational systems.
The solution to the problem requires multi-dimensional analysis.
Data must be integrated from multiple sources.
Because it is "pulled" by business requirements for information, the demand-driven approach is more
targeted to providing information that is useful to the customer (user) and in a way that is useful to and
actionable by him or her. The demand-driven approach to information management requires the IT
function to do more and likely costs more, but has the potential to deliver business value not possible with
the supply-driven approach.
Ventana Research, a leading performance management research and advisory services firm, compares the
two approaches this way: "Ventana Research continues to see a significant focus on the data-to-user
[supply-driven] approach, data warehousing to their information architecture and business requests,
instead of a user-to-data [demand-driven] approach that leverages business intelligence and performance
management. The data-to-user approach along with simply picking a best-of-breed tool for data
integration, data warehousing, metadata management, and business intelligence tools will not necessarily
bring full business value sought by CIO and business management from IT investments."
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 10
-------
1.3. Recommendation
A big advantage of the demand-driven approach is its linkage between business requirements for
information and information services to fulfill that requirement. This linkage provides an opportunity to
deliver business value in the short term. It facilitates a focus on business value and therefore provides a
framework for IT governance.
The demand-driven approach to providing information services is different than and a valuable
complement to the supply-driven approach. It would require OEI proactively to engage with its customers
to determine their information needs and then build information services specifically to meet those needs.
The demand-driven approach would also require OEI personnel to have deeper domain knowledge of the
work of EPA than the supply-driven approach, so that they could provide more useful context with the
information.
The demand-driven approach holds more potential for OEI to add value to EPA than the supply-driven
approach. DecisionPath recommends that OEI adopt the demand-driven model as its primary approach to
information management, while continuing the supply-driven approach for those situations in which it is
appropriate.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 11
-------
2.Technical Evaluation of Information Integration
Alternatives
OEI uses a number of data architectures, data integration approaches, and management techniques to
carry out its mission. This section discusses five such architectures, approaches, and techniques, for the
purpose of understanding exactly what they are, how OEI currently uses them, opportunities to extend
their use, and the situations in which they are and are not appropriate. The discussion topics are
Services-oriented architecture (SOA)
Data warehousing
Enterprise architecture
IT governance
Program Management Office (PMO)
2.1. Services-oriented Architecture (SOA)
Services-oriented Architecture (SOA) is an integration technique that loosely couples software services to
support business processes requirements. SOA implementations typically use a combination of XML-
based technologies to implement Web Services. At its core, SOA is based on the concept of Remote
Procedure Call (RFC). In the late 1990s, XML-RPC was introduced and quickly evolved into Web
Services. The basic idea in all of these variants is that the service provider and service requester agree on
a set of standard communication techniques (an Application Programming Interface or API). For Web
Services, the API is documented in Web Services Definition Language (WSDL). The service requester
needs only to follow the protocol for invoking the remote service, without being concerned about what is
actually happening at the far end of the communication. Web Services use the Simple Object Access
Protocol (SOAP). This infrastructure is further enhanced with a directory service known as User
Universal Description, Discovery, and Integration (UDDI) that provides for service discovery. A service
directory provides API metadata (WSDL) and the location of services that are currently online. In order
to locate various API services, the client program needs only know how to use the service directory.
Taken together, these features of SOA provide for dynamic discovery, dynamic location, and dynamic
invocation of RPC. Web Services add a feature to SOA by using HTTP or HTTPS as their transport
mechanism: those are often the only protocols allowed to pass through an organization's firewall. Figure
1 shows a simple diagram of Web Services.
For data integration, SOA typically is used to implement a federation or [limited] propagation
architecture. SOA typically performs best when small amounts of data are passed back and forth. When
large amounts of data must be moved from point to point, the communication overhead quickly can
outweigh the benefits of a loosely-coupled architecture. This is especially true of XML-based SOA, such
as Web Services, and in general of using XML to encode large volumes of data.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 12
-------
Service
BroAer
Service
Requestor
Service
Provider
Figure 1: How Web Services work.
One of the prevailing architectural visions within OEI is that SOA can and should be used as the basis for
its entire applications architecture. Its proponents envision an architecture that is a collection of Web
services. Services-oriented architecture (SOA) is a "hot topic" receiving a lot of coverage in the technical
media. SOA is a conceptually sound and viable architectural approach; however, its successful
application is not without challenges. A careful review of available literature about successful and
unsuccessful implementations of SOA reveals the following:
Two objectives of SOA are interoperability and service reuse. Service reuse entails using some
of the services developed for application, in application?, some services from application, and
application: in applications, and so on - enabling development of each application to take less
time and money than the application that preceded it.
The degree of service reuse that can be realized depends heavily on identifying the right business
process components to implement as services. Correct identification requires understanding both
the business processes and the information requirements.
"SOA by itself provides no guidance on how to build the right services to meet current business
requirements" Ronald Schmelzer and Jason Bloomberg, Three Roads to the SOA Implementation
Framework.
SOA still is relatively new and unproven. Its promise has not yet been realized in widespread
practice.
SOA success stories are based upon using SOA to solve specific business problems. The use of
the SOA-based Exchange Network as a common data intake mechanism for many EPA partners
to submit data into many EPA program office systems is an example of a SOA application to
solve a specific business problem - how to obtain data from EPA's partners.
Business process modeling often is used to define processes that are composed of services, as
shown in Figure 2. However, enterprise-wide business process modeling efforts are a long-term
undertaking and therefore are not practical to complete before beginning a SOA implementation.
SOA success requires a governance structure that has significant input from the business
user/customer.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 13
-------
"It's becoming evident from the experiences of early adopters that an SOA requires mechanisms
for oversight, policy governance, and change management" Phil Wainewright, Laying the
foundations for SOA.
Not all IT processes are suitable to be architected as a service. Examples of unsuitable processes
include
ETL streams that move large volumes of data (which could be either from source to data
warehouse or from data warehouse to data mart)
ETL streams that entail complex transformation logic
ETL streams that perform significant data cleansing
Federated queries that require data to be combined from many sources
Federated queries with complex logic
Queries that return a large volume of data
From our experience, these processes are much better satisfied by the consolidation approach to
data integration as typified by data warehousing. Business intelligence and data warehousing
generally are not good candidates for SOA, no matter what software tool vendors might claim.
Figure 2: A process for developing SOA-based applications.
It is important to have a clear understanding of the current and possible uses of SOA and the Exchange
Network by EPA. The SOA-based Exchange Network is used for data intake: the data flows from EPA's
partners to EPA. There is limited data flow from EPA to the partners. A partner such as a state already
has in its own system(s) the environmental data it collected and submitted to EPA. If it wants to use this
data, it retrieves the data from its own system(s) rather than "getting it back" from EPA. If an EPA
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 14
-------
partner has a need for environmental information that 1) it didn't send to EPA and doesn't have in its own
system(s), or that 2) must be integrated from various sources, neither the Exchange Network nor SOA are
the right mechanism for that partner to obtain the information it needs from EPA.
2.2. Data Warehousing
Data warehousing is a set of IT techniques to make information available for analysis and decision-
making. It uses the consolidation approach to data integration by pre-building special-purpose data stores
called data warehouses and data marts. According to Bill Inmon, the "father of data warehousing," a data
warehouse is "a subject-oriented, integrated, non-volatile, time-variant collection of data organized to
support management's needs." Although a data warehouse is a database, it differs from a database used
for an On-Line Transaction Processing (OLTP) application in that the data in it is specifically organized
for information distribution instead of for transaction processing. Data warehouses typically have a
relational structure that might not be as normalized as for an OLTP database.
Data marts are special-purpose data stores optimized for information access. They are designed to
facilitate end-user analysis of data. A data mart typically supports a single analytic application used by a
distinct set of end-users.
Data marts can be dependent (that is, sourced from a data warehouse) in a hub-and-spoke data
architecture, or independent in a bus architecture. The hub-and-spoke architecture is associated with Bill
Inmon, and the bus architecture is associated with Ralph Kimball: Mr. Inmon and Dr. Kimball are noted
authorities in the field of data warehousing.
Because data warehouses and data marts use the consolidation approach to data integration, they are well
suited to situations in which:
Large volumes of data must be moved.
Data from many sources must be integrated.
The transformation logic is complex.
The source data requires significant data cleansing.
User needs for information are repeatable and can be predicted in advance.
Users need historical or trend data.
Envirofacts is a data warehouse with loose integration by facility, substance, and geolocation. The TRI
and AQS data marts are independent data marts.
The TRJ and AQS data marts were built using a Kimball-oriented bus architecture. The debate within
OEI about conformed dimensions is a result of this architecture. DecisionPath believes OEI should
migrate toward a hub-and-spoke architecture, with Envirofacts being the hub data warehouse and the TRI,
AQS, and all future data marts being dependent spokes. Such a hub-and-spoke architecture offers more
future flexibility, in these ways:
The debate about conformed dimensions becomes moot: the dimensions in dependent data marts
are copies or subsets of the dimensions in the data warehouse.
Atomic-level history can be stored in the data warehouse rather than in the data marts, which
makes redesign of data marts for additional or changed requirements much easier.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 15
-------
Change to a source system only requires changing one ETL stream (from source to data
warehouse) to reflect the source change, versus requiring changes to the ETL stream from source
to data warehouse (Envirofacts) and the ETL streams from source to each data mart that uses that
source.
Envirofacts initially was developed circa 1993, at a time before good ETL tools were available.
Therefore, most of the ETL to populate Envirofacts is hand-coded. Redevelopment of this hand-coded
ETL using the Informatica ETL tool would make Envirofacts much easier to maintain. The constraint
impeding such a redevelopment effort is lack of funding.
Recap of SOA and Data Warehousing
Both SOA and data warehousing are valid approaches to application architecture and design. The key is
to use each in situations for which it is well suited. The table below describes the situations for which
each is better.
Characteristic or Situation
Method of data integration
Data to be moved
Trigger for data movement
Requirement for latency of data
Frequency of data change
Pattem(s) of usage
User requirements for information
Best if
SOA / Web Services
Federation or propagation
Individual transactions or
messages; small volume, but
perhaps frequent, moves
Event-driven; on demand
Near real-time; low latency
Volatile data
Not well known
Not known in advance; ad hoc
Desired result is a single record
Queries are simple
Need most up-to-date data
Service portfolio is dynamic
Service providers are transient in
location
Data Warehousing
Consolidation
Large volume
Usually scheduled; batch
Some degree of latency is acceptable
Stable data
Well understood; predictable
Predictable and recurring
Data from many sources must be
integrated
Historical data is required
Transformation logic is complex
Significant data cleansing is required
"Single version of the truth" is
important
Multi-dimensional analysis is
required
Aggregated data is needed
Decision Path believes that some of EPA's information needs are best met using a data warehousing
approach and that, therefore, OEI should pursue a combination of both SOA and data warehousing rather
than a pure SOA application architecture.
2.3. Enterprise Architecture
The Clinger-Cohen Act of 1996 assigned federal agency CIOs the responsibility to develop information
technology architectures. The CIO Council began developing the Federal Enterprise Architecture
Framework in 1998 "to promote shared development for common Federal processes, interoperability, and
sharing of information among the Agencies of the Federal Government and other Government entities."
The CIO Council defines enterprise architecture as "a strategic information asset base, which defines the
mission, the information necessary to perform the mission and the technologies necessary to perform the
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 16
-------
mission, and the transitional processes for implementing new technologies in response to the changing
mission needs. An enterprise architecture includes a baseline architecture, target architecture, and a
sequencing plan."
The objectives of the Federal Enterprise Architecture Framework are to
Promote Federal interoperability
Promote Agency resource sharing
Provide potential for Federal and Agency reduced costs
Improve ability to share information
Support Federal and Agency capital IT investment planning
OEI has been working since 2001 to implement an EPA Enterprise Architecture (EA) aligned with the
Federal Enterprise Architecture. The Chief Architect and others continue to work diligently to develop
and communicate EPA's EA.
Some of the program offices, most notably the Office of Pollution Prevention and Toxics (OPPT), are
documenting their business processes for inclusion into the EA. OPPT has documented thirty-two
process flows and one of them, Inventory Update Rule, is one of the first applications (called iEUR) of
ECMS, EPA's implementation of Documentum.
One of the fundamental challenges of the team working on EPA's EA is to make it relevant and actually
influence how IT development funds are spent and how individual projects are designed and built. In
order to be relevant, the EA must be more than an abstract representation of an idealized future state: the
sponsors, users, and developers of systems must be able to tell why and how to apply it to their project(s).
Widespread adoption of the EA within EPA will be very challenging because of EPA's decentralized and
organizationally stovepiped culture. Most of the (primarily OEI) EPA employees that we interviewed
voluntarily voiced support for the EA, and no one explicitly opposed it. However, we do not see
widespread support for the EA in practice. For example, the Chief Architect's September 14, 2006
presentation to the IRM Branch Chiefs Meeting entitled "Evolving EPA's Enterprise Architecture" listed
eleven proposed architecture priorities. However, it also noted that "These 'priorities' highlight how EA
can be used, but have not been selected to be acted on."
EPA's enterprise architecture provides guidance for how systems are to be designed and built. However,
implementing the EA is a massive, long-term effort that will take many years to complete.
2.4. IT Governance
IT governance is a subset of corporate governance that deals with the connection between business focus
and IT management of an organization. Its primary goals are 1) to assure that investments in IT generate
business value, and 2) to mitigate the risks that are associated with IT.
In EPA's context, "business value" can be defined as realization of its mission ("To protect human health
and the environment") and achievement of the goals articulated in its 2006-2011 Strategic Plan.
Compliance with environmental statutes and other legislation and federal policy also might be considered
business value.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 17
-------
In The Information Paradox, John Thorp describes IT governance using the "Four Ares" model. Figure 3
showing the "Four Ares" is based on a diagram in Enterprise Value: Governance of IT Investments, The
Val IT Framework, published by the IT Governance Institute.
The strategic question: Is the
investment:
In line with our vision
Consistent with our
business principles
Contributing to our
strategic objectives
Providing optimal value, at
affordable cost, at an
acceptable level of risk
The architecture question: Is
the investment:
In line with our architecture
Consistent with our
architectural principles
Contributing to the
population of our
architecture
In line with other initiatives
Are
doing the
right
things?
we
getting
the
benefits?
doing
them the
right
way?
Are we
getting
them done
well?
The value question: Do we
have:
A clear and shared
understanding of the
expected benefits
Clear accountability for
realizing the benefits
Relevant metrics
An effective benefits
realization process
The delivery question: Do we
have:
Effective and disciplined
management, delivery,
and change management
processes
Competent and available
technical and business
resources to deliver:
> The required
capabilities
> The organizational
changes required to
leverage the
capabilities
Figure 3: The "Four Ares" model of IT governance.
The IT Governance Institute has developed two complementary frameworks for IT governance: Val IT
and COBIT© (Control Objectives for Information and Related Technology). Val IT and COBIT use
Thorp's "Four Ares" model. COBIT "provides a comprehensive framework for the management and
delivery of high-quality information technology-based services. It sets best practices for the means of
contributing to the process of value creation." Val IT "adds best practices for the end, providing the
means to unambiguously measure, monitor, and optimize the realization of business value from
investment in IT."
The purpose of this discussion of Val IT and COBIT is not to recommend them to OEI, but to
demonstrate that IT governance is:
A mature discipline with well-developed conceptual frameworks
Fundamentally about recognizing that IT is an investment for the purpose of realizing business
value, and managing it that way
The Clinger-Cohen Act of 1996 mandated that certain federal agencies have a CIO, requires federal
agencies to have an Enterprise Architecture, and created the Capital Planning and Investment Control
(CPIC) process. These and other external requirements are a form of IT governance imposed upon EPA,
but complying with the letter of the requirements does not necessarily yield effective governance. The
externally imposed governance structures of the Clinger-Cohen Act attempt to motivate agency IT
behavior through controlling the purse strings, essentially saying, "follow the rules or we won't give you
the money for your IT project."
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 18
-------
In our interviews with them, several individuals indicated that IT governance is weak at EPA. On the
other hand, in the project kick-off meeting, one of the participants said that OEI had built a big IT
governance structure over the past few years. The participant might have been referring to external
mandates imposed on all federal agencies, such as the Clinger-Cohen Act of 1996.
The level of externally imposed governance varies by the size (amount of funding) of the project. The
CPIC process requires the most control for projects greater than $3,000,000, and lesser control over
projects from $250,000 to $3,000,000. One interviewee indicated that there is little governance over or
accountability for projects less than $250,000.
We believe that the governance mechanisms mandated by the Clinger-Cohen Act (enterprise architecture,
CPIC, and so on) do not completely inherently link IT investments with business value. There is potential
for OEI to improve IT governance to achieve such linkage.
In order to realize business value from an IT investment,
The business value to be gained must be specifically identified.
The business process changes by which that business value will be realized must be understood.
How information and/or information technology will enable those process changes must be
specified.
These activities, which normally are part of requirements gathering, require OEI to engage with its
business customers and understand their domains. They cannot be done in isolation from the business.
Returning to the Thorp "Four Ares" model of IT governance, two of the four questions can only be
answered by the business, not by IT, as shown below.
Type of Question
Strategic
Architecture
Delivery
Value
Question
Are we doing the right things?
Are we doing them the right way?
Are we getting them done well?
Are we getting the benefits?
Answered by
Business
IT
IT
Business
2.5. Program Management Office (PMO)
The Project Management Institute (PMI) defines a program as "a group of related projects managed in a
coordinated way to obtain benefits and control not available from managing them individually. Programs
may include elements of related work outside of the scope of the discrete projects in the program." PMI
defines a program management office (PMO) as "the centralized management of a particular program or
programs such that corporate benefit is realized by the sharing of resources, methodologies, tools, and
techniques, and related high-level project management focus."
In simple terms, a PMO manages a group of related projects to identify and realize synergies among
them. These synergies could be reuse of common resources, tools, or objects; coordination of
dependencies between projects; and so on.
OEI has a PMO that primarily focuses on providing a toolkit of software tools (sign-on, portal, ETL, and
so on) for use by the program office IT groups to develop their own applications. The potential benefits
of a PMO with such an emphasis are primarily in the economics of the IT infrastructure (such as
avoidance of multiple tools for the same purpose and negotiating leverage for software licensing).
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 19
-------
There also is an ECMS (EPA's implementation of Documentum) PMO. An ECMS PMO is a logical and
traditional application of a PMO to prioritize and sequence the individual Documentum implementation
projects, realize synergies among these projects, and so on.
The OEI PMO as currently tasked serves a useful purpose. OEI has an opportunity to give its PMO a
larger and more value-added mission: to measure, monitor, and optimize the realization of business value
from the investments EPA makes in OEI. Such a mission would have two major parts: investment
decisions (Are we doing the right things?) and realization of benefits (Are we getting the benefits?).
If OEI gives its PMO this mission, several issues will have to be worked out:
What is the relationship between the IT governance work of the PMO and the Clinger-Cohen Act
compliance, enterprise architecture, and CPIC activities performed by other elements of OEI?
The managers in OEI who currently can initiate "less than $250,000" projects with minimal
governance and accountability now will have such projects governed by the PMO. This cultural
change might not be well accepted unless it is carefully managed.
The larger question is to what extent the OEI PMO can control or influence IT investment decisions and
projects by the program offices and regions. If it cannot, its impact will be significantly reduced.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 20
-------
3. Development Plan for Information Integration
There are a number of ways OEI can combine the supply- and demand-driven approaches to information
management described in section 1 of this document with the technical alternatives - SO A, data
warehousing, EA, IT governance, and PMO - described in section 2 into a program plan for moving
forward. This section covers four such options, recommends one of them, and outlines next steps for OEI
to adopt the recommended option. The four options are:
Option 1: Maintain status quo
Option 2: Improve governance
Option 3: Governance with top-down information needs
Option 4: Governance with bottom-up information services
3.1. Option 1: Maintain Status Quo
The status quo option can be described as:
Continue the present supply-driven approach to providing information services to internal EPA
customers.
Continue to provide environmental information to the public via the EPA Web site.
Incrementally upgrade and extend the capabilities of the Web site.
Advantages
The primary advantage of continuing the status quo is that it requires no change and no incremental effort.
Disadvantages
OEI's future, including the prospect of additional funding, may be limited.
Beyond provision of IT infrastructure, OEI's value proposition to its internal customers is not clear to
them. Therefore, they turn to their internal IT departments for services rather than to OEI.
OEI's ability to serve its external customers, the public, is limited, because it doesn't know their specific
needs so can offer only undifferentiated supply-driven information services to them.
There is limited linkage between OEI's services and EPA's business needs. At some point in the future,
some linkage might be provided by full implementation of the Enterprise Architecture.
Like all federal agencies, EPA faces budgetary constraints and must make difficult spending choices and
trade-offs. Without linkage between OEI's services and EPA's business needs, it is difficult to make IT
investment and technology decisions.
Without a strong business case for its services, it is difficult for OEI to make and defend specific
technology investment choices.
January 30,2007 Environmental Information Integration - Final Deliverable (revised) Page 21
-------
3.2. Option 2: Improve Governance
OEI already has components that work on the EPA Enterprise Architecture (EA) and perform the other
functions required by the Clinger-Cohen Act, such as the CPIC process. These existing governance
mechanisms fall short both in the identification of specific business value for EPA and in tracing a link
between IT investments (whether they be in application systems or in tools and technology) and
realization of that business value. Without linkage to business value, IT investments risk being
technology solutions looking for a problem.
An option to improve OEFs IT governance to achieve a tighter linkage between investment and business
value would be to expand the governance role of its program management office (PMO). Such a role
would entail:
Development of a multi-year program plan of IT investments, based on a portfolio of investment
opportunities
Identification (or collection) of the IT investment opportunities that make up the portfolio,
including specification of the potential business value of each and the implementation risks
involved
Figure 4 graphically illustrates the development and management of the opportunity portfolio, with
prioritized implementation of selected opportunities.
Portfolio
Development
Information Services Portfolio
Cross-Goal Management Processes:
Environmental Performance Analysis
and Measurement, Policy Impact
Analysis, Budget/Performance
Integration
Program/Region Processes:
Internal Operations & Infrastructure
(From Federal Enterprise
Architecture)
Implementation Risk
Portfolio
Management
Must Haves
Easy Wins
High Risk/Reward
Why Do It?
Opportunity E
Implementation
Figure 4: Development and Management of an IT Investment Portfolio.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 22
-------
The objective of this PMO governance role is to ensure that IT investments are prioritized and approved
based upon cost/value trade-offs.
In order to be successful in this expanded role, the PMO should report to the CIO and the OEI Board of
Directors and have sufficient authority to develop investment recommendations pertaining to the
information services portfolio.
3.3. Option 3: Governance with Top-Down Information
Needs
If OEI wants to deliver the maximum business value to the agency, it should combine the improved
governance of Option 2 with a change to the demand-driven model for information management, and
proactively engage with EPA executives in the program offices and regions to determine their information
needs. In this option, OEI would utilize different approaches for its internal and external customers:
Internal customers: Canvas agency executives to determine information needs and set agency-
wide priorities for enhanced information services.
External customers: Develop a program plan for a multi-year effort to extend and enhance the
public's access to environmental information.
Because "the public" is such a large and heterogeneous customer group, it would be quite
difficult for OIAA to engage with it at the level necessary to provide demand-driven information
services. However, OIAA could more proactively engage subsets of the public regarding how
they want EPA's information presented to them. Such an approach could be characterized as
supply-driven content with demand-driven information presentation.
Advantages
Engaging with agency executives regarding their information needs, and responding to them, would give
OEI an opportunity to obtain their support for its value proposition, and potentially make it easier to
obtain funding for its proposed information services.
EPA's external customers, the public, would benefit from a more managed program to provide
environmental information to them in formats tailored to their needs.
Disadvantages
OEI does not have a history of engaging with agency executives regarding their business requirements for
information, so the demand-driven approach would be a significant change for both OEI and agency
executives.
Because of EPA's decentralized culture of program office and regional stovepipes, it might not be
possible to obtain executive consensus regarding information services priorities. Each program office and
region might have its own parochial set of priorities.
Such a top-down approach and building the consensus it requires regarding information services priorities
takes time. There might be a perception that this approach would take too long to realize benefits.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 23
-------
The business engagement aspect of this option would be a very significant change from the way OEI
currently operates, and therefore would require large cultural change within OEI.
3.4. Option 4: Governance with Bottom-Up Information
Services
Option 4 is a more measured undertaking than Option 3. Rather than attempting an agency-wide top-
down approach to OEI providing information services to its internal customers, it starts small by using a
"prototype" to demonstrate the demand-driven model of information management. This prototype would
be limited to a single selected EPA organization (program office, region, or headquarters component). It
should be an example of a "must have" as shown in Figure 4: that is, it should have high performance
impact but low implementation risk. The prototype's purpose would be to demonstrate how the demand-
driven model works and to demonstrate the business value that it can achieve.
The approach to providing information services to OEI's external customers, the public, would be the
same for Option 4 as it is for Option 3: develop a program plan for a multi-year effort to extend and
enhance the public's access to environmental information, then execute that plan.
Advantages
A prototype is a low-cost, low-risk way for OEI to explore the demand-driven approach to information
management with an internal customer.
OIAA could use a successful demand-driven prototype as a marketing tool to demonstrate a new
capability to other internal customers. Additional capabilities and information services are the best way
for OIAA to guarantee future funding and increase its impact upon the agency.
A successful prototype will demonstrate that targeted demand-driven information services:
Can quickly deliver business value while longer-term initiatives such as enterprise architecture
are in progress.
Can coexist with the current supply-driven information services.
3.5. Recommendation
The four options presented are not mutually exclusive. With the exception of Option 1 (maintain status
quo), all utilize the OEI PMO to improve IT governance. Options 3 and 4 both use a program plan to
extend and enhance public access to environmental information. The primary difference between Options 3
and 4 is the extent to which they pursue a demand-driven model for information management: Option 3
takes a top-down comprehensive approach, while Option 4 takes a more limited prototype-based approach.
Option 4, Governance with Bottom-up Information Services, offers the best combination of benefits and
probability of success for OEI. A successful prototype that demonstrates the applicability of the demand-
driven approach and its greater ability to deliver business value will give OEI the opportunity to further
expand use of the demand-driven approach.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 24
-------
3.6. Next Steps
Executing Option 4 has three distinct parts: developing a demand-driven prototype, repurposing the OEI
PMO to play a larger governance role, and creating and then executing a program plan for public
information services to realize "supply-driven content with demand-driven information presentation."
Demand-Driven Prototype
The first step in developing a demand-driven prototype is to identify an EPA program office, region or
headquarters component with a business need for information that the prototype will satisfy, and then to
obtain that organization's agreement to participate. Selection of the right information service to prototype
is critical.
Because the demand-driven approach is new to OEI, it might want to consider expert assistance with the
prototype project, particularly in gathering the business requirements it will satisfy.
Use OEI PMO to Improve Governance
The essence of the additional role for the OEI PMO is to manage a portfolio of investment opportunities
(projects to create new or improved information services) in order to achieve maximum business value.
This additional role will be a significant change both for the PMO and OEI as a whole, and this change
should be managed as a project.
Organizational aspects of this PMO role change include the PMO charter, authority, reporting
relationship, size/resources, and the relationship between the PMO's new governance activities and CPIC,
enterprise architecture, and other existing governance structures and processes.
The additional work of the PMO includes
Collecting and identifying the IT investment opportunities (projects) that will make up the
portfolio. These opportunities include projects already in progress, approved but not yet started,
requested but not yet approved, and unmet information needs not yet recognized as projects.
For each opportunity in the portfolio, understanding and documenting its potential business value,
costs, and business and technical risks.
Working with internal customers and OEI executive management to prioritize the opportunities
based on business value, cost, and risk.
Creating and maintaining the opportunity portfolio, and communicating its contents using
mechanisms such as the graphic in Figure 4.
Creating and maintaining a multi-year program plan to execute the prioritized and sequenced
opportunities in the portfolio.
Providing oversight of individual projects in progress. (The project managers of the individual
projects will retain responsibility for managing them.)
Monitoring completed projects for realization of business value, which often lags behind
completion of the project.
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 25
-------
The PMO will be the mechanism by which OEI explores the demand-driven approach to information
management. Therefore, initially it will be the locus for acquisition of new skills in business
requirements elicitation, portfolio management, and demand-driven information services. When the
prototype is successful and OEI moves to adopt the demand-driven approach more broadly, these skills
will need to diffuse more widely throughout OIAA and eventually throughout OEI.
A program management office chartered as described is a mainstream application of the PMO concept,
but might be larger in scope and scale than OEI's PMO experience. OEI might want to obtain expert
assistance for the PMO until its new role has been completely integrated into the organization.
Program Plan for Public Information Services
The public is a large and heterogeneous group of customers for OEI's information services. Because the
environmental data that EPA and its partners collect primarily is driven by statute rather than customer
requests for information, the content of OEI's public information services must be largely supply-driven.
However, the way environmental information is presented and delivered to the public is less constrained.
OIAA, and more specifically its Information Access Division (IAD), is the public-facing part of OEI via
the EPA Web site. IAD already reaches out to the public to determine its information needs through a
variety of mechanisms: focus groups, satisfaction surveys, and the like. These and additional mechanisms
can be used to categorize users by type or interest (such as teachers, farmers, real estate purchasers, and
so on), so that the EPA Web site presents environmental information in a manner tailored to the specific
needs of each group. The result would be demand-driven information presentation for the public.
Implementing demand-driven information presentation by category of public customer would be managed
best by an overall program plan for public information services.
Summary
These next steps to develop a demand-driven prototype, improve IT governance via an expanded OEI
program management office, and create a program plan for enhanced public information services will
move OEI forward toward playing a larger role within EPA and delivering more value to both internal
and external customers. They require change primarily from OIAA rather than from the Office of
Information Collection (QIC) or the Office of Technology Operations and Planning (OTOP). They also
begin to define a compelling new role and value proposition for OIAA: provision of demand-driven
information services.
January 30. 2007 Environmental Information Integration - Final Deliverable (revised) Page 26
-------
Appendix A Interviewees
DecisionPath has held thirty-three information requirements-gathering meetings with 39 EPA employees:
Person
Count
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
21
32
33
34
22
35
6
36
8
23
37
26
6
6
7
36
38
39
Group
1
1
2
3
3
4
4
4
5
6
7
7
8
9
10
II
12
13
13
13
14
14
14
14
15
16
17
18
19
20
21
22
23
23
24
25
26
27
28
28
29
30
31
31
31
32
32
32
32
33
Interviewee
Mike Petruska
Ben Smith
Nancy Wentworth
Leo Gueriguian
Lisa Ayala
Lionel Brown
Emma McNamara
Dalroy Ward
Todd Holderman
Branch Chiefs Meeting
Cindy Dickinson
Larry Fitzwater
Mike Barrette
Odelia Funke
Gene Durman
Lisa Jenkins
Oscar Morales
Connie Dwyer
Chuck Freeman
Chns Clark
Doreen Sterling
Pat Garvey
Sara Hisel-McCoy
Jonda Byrd
Mike Cullen
Kevin Phelps
Mary McCafTery
Mark Hamilton
Kevin Kirby
Brion Cook
David Hindm
John Sullivan
Pat Garvey (2)
Alex Klassaeg
John Harman
Connie Haaser
Sara Hisel-McCoy (2)
Maryane Tremaine
Lionel Brown (2)
Rick Martin
Dalroy Ward (2)
Jonda Byrd (2)
Craig Hooks
Mary McCaffery (2)
Lionel Brown (3)
Lionel Brown (4)
Emma McNamara (2)
Rick Martin (2)
Linda Travels
Warren Beer
Assistant
Administrator
OEI
OEI
OEI
OARM
OCFO
OEI
OEI
OEI
OPPTS
OEI
OEI
OECA
OEI
OAR
OSWER
OEI
OEI
OEI
OEI
OEI
OEI
OEI
OEI
OEI
OSWER
OEI
OW
OEI
OPPTS
OECA
OEI
OEI
OEI
OEI
OSWER
OEI
Region 7
OEI
OEI
OEI
OEI
OEI
OEI
OEI
OEI
OEI
OEI
OEI
Region 9
Office
OIAA
OIAA
OIAA
99?
???
OIAA
OIAA
OIAA
OPPT
QIC
QIC
OC
OTOP
OPMO
PMO
QIC
QIC
QIC
QIC
QIC
QIC
QIC
OIAA
OTOP
IMO
...
999
QIC
OPPT
OC
OTOP
QIC
QIC
QIC
OEM
QIC
PLMG
OIAA
OIAA
OIAA
OIAA
...
OIAA
OIAA
OIAA
OIAA
MTSD
Division
TRIPD
TRIPD
EAD
999
999
IAD
IAD
IAD
IMD
CSD
CSD
EPTDD
MISD
99?
999
CSD
IESD
IESD
IESD
IESD
IESD
CSD
IAD
999
...
...
???
IESD
IMD
EPTDD
MISD
IESD
IESD
CSD
NPPD?
CSD
...
IAD
IAD
IAD
...
IAD
IAD
IAD
...
...
Branch
...
...
...
999
999
...
ISB
SIB
DSB
DSB
999
...
???
99?
_
...
IETB
1ETB
...
IEPB
...
PPMB
999
...
_
???
ISSB
...
...
...
IEPB
IEPB
DSB
999
...
IRM-B
ISB
PPMB
...
...
...
...
...
1RM
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 27
-------
A breakdown by organization shows that most of the people we have interviewed work in OEI, as shown
below.
Organization
OEI
Program offices
Administrative offices (OARM & OCFO)
Regional offices
Total
Number of Interviewees
26
9
2
2
39
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 28
-------
Appendix B Information Needs Areas
The result of our information requirements-gathering interviews typically is a number of business
questions that users want to be able to answer. We group these questions according to a common theme
and a specific set of users into information needs areas. An information needs area may drive the creation
of a data mart which targets that theme and that set of users.
Our interviews with EPA personnel have yielded five information needs areas:
Emergency planning and response
Environmental management of health risk and prioritization
Environmental policy analysis
Public information
Enforcement and compliance
Because we inferred many of the business questions from the conversations we had with EPA personnel,
and because we primarily have spoken with OEI personnel and therefore have been at least one degree of
separation from the actual users of information, both the business questions and the information needs
areas need to be validated with users. We believe direct communication with users in the program offices
and regions will yield additional information need areas.
Emergency Planning and Response
The vast majority of incidents to which EPA's Office of Emergency Management responds are localized
and straightforward. EPA handles such incidents well using existing information and business processes.
Major incidents, such as Hurricane Katrina, the space shuttle disintegration, and the Exxon Valdez oil
spill, are more problematic. Such incidents occur on a much larger scale, are more geographically
dispersed, have multiple EPA employees and other agencies working on them, and both require and
generate much more data.
9/11 and Katrina have required a paradigm shift from localized emergency response to a coordinated
national approach for major incidents. Because these types of incidents are more challenging, and
because coordination and response to these types of incidents is time-critical (e.g., need for potable
water), EPA needs integrated information to better support emergency response. To more effectively
respond to such incidents, EPA needs:
Better ability to transmit the data collected from multiple field locations and bring it together at
EPA headquarters into a holistic picture of the incident.
Ability to identify, by geolocation, all relevant facilities in the geographic area affected by the
incident. While all of the facility records in FRS contain geolocation (in some cases, populated
by an address-matching algorithm), the accuracy and usefulness of this data varies. For example,
in a large site, the geolocation data in FRS might not correspond with the physical location of the
item within the site.
Better ability to integrate facility data stored in regional systems into FRS for use by headquarters
personnel.
January 30,2007 Environmental Information Integration - Final Deliverable (revised) Page 29
-------
Better ability to incorporate information from external sources (such as Tier 2 prevention data
from states) into EPA's own systems and to use this combined data quickly in the incident
response. The external data most needed is locations of things (that is, facilities) in the incident
area that are known to states and other parties, but are not known to EPA.
EPA has limited ability to combine data from multiple incidents to answer questions from the public,
Congress, and other stakeholders. Because incident data is stored and organized in a site-specific and
incident-specific manner, cross-incident analysis is laborious and often depends on the memory of field
response personnel rather than being data-based. The organization of the data also makes it difficult to do
analysis by category of incident for continuous improvement of prevention and preparedness.
Representative Business Questions:
In how many sites did EPA have to clean up chemical X? Which sites were they?
For a major incident, such as Hurricane Katrina, what potential hazards exist by substance, media,
facility name, and geolocation of facility?
For a major incident, such as Hurricane Katrina, what is the current measure of pollutant/hazard
by media, facility name and geolocation of facility? Is this level hazardous?
1 Based on post-incident measures of pollutant/hazard by media, what actions are needed to remedy
the problem by priority and geographical range?
Based on existing post-incident hazards, what health protection actions (such as warning local
residents not to drink the water) are required by geographical and time range of the problem?
Environmental Management of Health Risk and Prioritization
To support its mission of protecting human health, EPA must be able to correlate exposure to
environmental hazards with the potential health risks of such exposure. A national, cross-media picture
of current and anticipated future environmental hazards is necessary to equip EPA with the information it
needs to evaluate agency tradeoffs and set agency priorities. Integrated information on environmental
hazards is also needed at the geolocation level so that localities can evaluate and prioritize environmental
initiatives based on understanding the environmental hazards in their area and the risks they pose to
people and wildlife. Finally, EPA, the states, and the tribes need to work with the same integrated
environmental information so that they can coordinate their efforts to answer to the public and remedy the
highest priority environmental problems that exist.
Currently, while there is some integrated environmental information, it has the following limitations:
Cannot Relate Levels of Exposure to Health Risk Factors
While partial cross-media information at the local level (zip code) exists in TRI and Envirofacts,
it does not contain all pollutants and does not provide information that would allow stakeholders
to analyze how varying levels of exposure to those pollutants present risks to human health and
wildlife. Therefore, it is difficult to set priorities at an agency and local level. In some cases,
such as lead in children, direct linkages between levels of exposure to the pollutant and associated
health risks may be possible. In other cases, only a loose linkage to health hazards may be
feasible (e.g., documents relating levels of exposure of a substance to increased health problems).
January 30, 2007 Environmental Information Integration - Final Deliverable (revised) Page 30
-------
1 Difficult to Identify All Environmental Hazards by Facility
While program offices collect information at the local level, different systems sometimes identify
the same facilities in different ways, making it difficult to provide a complete picture of the
pollutants that exist at specific locations. This information would allow EPA to better quantify
the total environmental impact of a facility across all substances it emits. This information,
combined with health risk information, would enable agency and local level cross-media
prioritization of facilities based on highest levels of risk.
Health Risk Information Only Available at Summary Level
While health risk information is addressed at a summary level in the Strategic Plan, it is currently
not available by geographical area to guide tactical planning needed to identify and remedy high-
priority environmental hazards.
Conflicting Environmental Information
Information used by the states and tribes to manage local environmental priorities does not
always match information used by the EPA due to timing differences and differences in
information identifiers. This can cause EPA, the states, and the tribes to reach different,
conflicting conclusions regarding environmental hazards and priorities.
While EPA has made solid progress in obtaining information on environmental hazards, integrating this
information across media and providing information on known health risks based on levels of exposure
would provide improved guidance for agency and local-level planning, prioritization, and funding. It
would also improve the agency's ability to relate its priorities and funding to improving outcomes.
Representative Business Questions:
What are the health risks for a given substance by exposure and demographics?
What are the health risks for a given location, by demographics?
What are the health risks for a given practice (e.g., eating shellfish) by source and demographic?
What is the current state of air quality? How has this changed over time?
What is the current state of water quality? How has this changed over time?
What level of water quality problems exist, by contributing source of problem, and by location of
problem?
For a given substance, what are the locations (geolocation) of the sources of the substance?
Environmental Policy Analysis
EPA's 2006-2011 Strategic Plan continues and reinforces a focus on environmental and human health
outcomes. This focus on outcomes requires the ability to evaluate, select, and implement programs to
achieve those outcomes, determine the societal and EPA costs of the programs necessary to do so,
understand the relationship between costs and benefits, and assess the extent to which the outcomes are
being realized. When existing approaches do not achieve the desired outcomes, EPA needs new and/or
additional approaches to close the outcome gap. Such analyses require sophisticated scenario modeling
January 30,2007 Environmental Information Integration - Final Deliverable (revised) Page 31
-------
capability using economic, population, public health, environmental, and other data. EPA currently has
limited capability for such scenario modeling. As a result, the analysis done is laborious, semi-manual,
often uses incomplete data, and does not consider as many variables or as many scenarios as would be
useful. The availability of high quality cross-program environmental data integrated with external
economic, public health, and other data would facilitate analysis and contribute to better outcomes.
Representative Business Questions:
What is the environmental improvement trend based on enforcing an existing EPA regulation?
What is the expected improvement versus the actual improvement?
What environmental problem areas do not appear to be improving over time based on enforcing
existing EPA regulations?
What is the most cost-effective way (among several alternatives) to achieve a targeted
environmental or human health outcome? What are the projected costs and benefits of each
alternative approach?
What is the cost (for industry and society in general, for EPA to monitor and enforce, and so on)
of a specific regulation? Is the regulation achieving its intended environmental and societal
outcomes?
Public Information
Inherent in EPA's mission to protect human health and the environment is the obligation and opportunity
to provide information about the environment, environmental hazards and risks, and environmental
protection and remediation efforts to the public and other stakeholders. While EPA wants to provide
information so that stakeholders are informed about the environment, some of EPA's data is sensitive
(proprietary, of potential use to terrorists, protected by privacy regulations, and so on). Therefore, access
to some types of information must be carefully controlled.
One way EPA provides information to its stakeholders is reactive: responding to FOIA requests,
Congressional inquiries, correspondence from citizens, and so on. Such information requests can be
wide-ranging. EPA needs to be able to respond to such requests with the right information in a timely
and cost-effective way that is consistent with privacy and sensitivity constraints.
Another way EPA provides information is proactively, through self-service programs such as Envirofacts
and Window to My Environment delivered via the EPA Web site and other vehicles.
The reactive and proactive communications are often geographically- or location-oriented. In contrast,
EPA's data is collected and organized primarily for specific regulations and is therefore organized in
systems by media type. Much of it also comes from external entities, such as regulated industries, states,
tribes, and other EPA partners, which makes it challenging for EPA to ensure the data's quality and
completeness. Currently, much time and manual effort is expended integrating data that originates from
multiple EPA and external systems in order to answer information requests. Data quality issues also
require time and effort to research and to resolve. They sometimes result in less complete or correct
information provided to the requestor.
The ability to quickly and easily provide environmental information integrated by geographic location
would enable EPA to provide more information that is useful to the public and other stakeholders.
January 30,2007 Environmental Information Integration - Final Deliverable (revised) Page 32
-------
Representative Business Questions:
What are the funding needs of each non-federal SuperFund NPL (national priority list) site?
(This was a question from Congress to EPA OIG.)
Which SuperFund sites do not have sufficient funding allocated to maintain clean-up progress?
What risks are associated with delaying clean-up completion?
Enforcement and Compliance
EPA employs an integrated approach of assistance, incentives, and civil and criminal enforcement to
maximize compliance with applicable laws and regulations and to reduce threats to public health and the
environment. Largely through its Office of Enforcement and Compliance Assurance (OECA), EPA
performs inspections of sites and facilities, collects and integrates compliance data, develops compliance
monitoring programs to support inspections and self-reporting, supports enforcement activities, prosecutes
intentional misconduct, and oversees the enforcement of EPA's national hazardous waste cleanup programs.
OECA is a relatively new organization within the EPA. It was created by centralizing compliance and
enforcement personnel from the program offices. These centralized personnel use data from the various
databases within the program offices, and they need to integrate this data into a holistic picture. For
example, OECA needs to see all the permits, RMPs, enforcement actions, penalties, and so on for a given
corporation, no matter how many physical locations it has, how many legal entities compose that
corporate entity, in how many regions it has facilities, and how many regulations apply to it.
Although OECA has the IDEA data warehouse that integrates compliance and enforcement data across
programs, IDEA does not satisfy all of its information needs.
Representative Business Questions:
0 What were the total enforcement actions by region, medium, industry, or time?
" Is there a correlation between the number of regulated facilities in a locale and the quality of its
drinking water?
" For a given corporation, what is the compliance and enforcement history across all legal entities,
facilities, and programs?
What trends can be seen in penalties assessed over time by region, industry, or program?
For a given facility, which program offices regulate it?
How many SuperFund sites were in the regulated community before they were designated
SuperFund sites? Which ones are they? (This was a question from OIG.)
What facilities with compliance problems are pan of the same corporation? Which corporations
have the best compliance records?
What is the best deployment of limited inspection, compliance assistance, enforcement, and
prosecution resources across the population of regulated entities and facilities?
January 30,2007 Environmental Information Integration - Final Deliverable (revised) Page 33
-------
Appendix C Documents Reviewed
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Date
Mar-So
Feb-90
Jun-91
Apr-92
Nov-93
Sep-94
Mar-96
Jun-97
Mar-98
Jun-98
Jan-99
May-99
Aug-99
Aug-99
Sep-99
Sep-99
Nov-99
Feb-00
May-00
Jun-00
Jul-00
Aug-00
Oct-00
Oct-00
Nov-00
Nov-00
Nov-00
Jan-01
Feb-01
May-01
Aug-01
Aug-01
Oct-01
Source
GAO
GAO
GAO
GAO
GAO
GAO
EPA
EPA
GAO
EPA
EDSC
EPA
EPA
EPA
GAO
CIO Council
EPA
EPA
EPA
GCN
GAO
EPA
GAO
EPA
EPA
EDSC
EDSC
EDSC
CIO Council
EPA
EPA
EDSC
EDSC
Title
CED-80-18 Stronger Management of EPA's Information Resources Is
Critical To Meeting Program Needs
PEMD-90-3 Hazardous Waste: EPA's Generation and Management
Data Need Further Improvement
T-IMTEC-91-16 Ineffective Information Management Impedes EPA's
Enforcement Mission and Cross-Media Initiatives
IMTEC-92-14 Environmental Enforcement: EPA Needs a Better
Strategy to Manage Its Cross-Media Information
AIMD-94-25 EPA Toxic Substances Program: Long-standing
Information Planning Problems Must Be Addressed
RCED-94-93 Toxic Substances: EPA Needs More Reliable Source
Reduction Data and Progress Measures
AIRS - AQS Modernization Proposal System Management Plan for
AQS
Executive Summary: AQS Conceptual Design Document
Measuring Performance and Demonstrating Results of Information
Technology Investments
OIG OFFICE OF WATER DATA INTEGRATION EFFORTS
FINAL SIC/NAICS DATA STANDARD
Standard Update
Standard Update
Aiming for Excellence - Actions to Encourage Stewardship and
Accelerate Environmental Progress
ENVIRONMENTAL INFORMATION: EPA Is Taking Steps to Improve
Information Management, but Challenges Remain GAO/RCED-99-261
Federal Enterprise Architecture Framework
Standard Update
Standard Update
Standard Update
EPA CIO Levine cut his teeth on states' systems
AIMD-00-215 Information Security: Fundamental Weaknesses Place
EPA Data and Operations at Risk
Standard Update
GAO-01-97T Environmental Information: EPA Needs Better
Information to Manage Risks and Measure Results
Blueprint for a National Environmental Information Exchange Network
BIOLOGICAL TAXONOMY DATA STANDARD BUSINESS RULES
Standard Data Elements For Facility Identification
Standard Data Elements for Latitude/Longitude
Standard Data Element for Date
A Practical Guide to Enterprise Architecture
Standard Update
Standard Update
Standard Data Elements for Biological Taxonomy
Standard Data Elements for Chemical Identification
January 30,2007 Environmental Information Integration - Final Deliverable (revised)
Page 34
-------
No.
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
Date
Dec-01
Feb-02
May-02
Jul-02
Aug-02
Sep-02
Sep-02
Sep-02
Sep-02
Sep-02
Nov-02
Dec-02
Jan-03
Jan-03
Jan-03
May-03
May-03
Jun-03
Jun-03
Sep-03
Nov-03
Nov-03
Nov-03
Feb-04
Mar-04
Apr-04
Oct-04
Nov-04
Nov-04
Jan-05
Jan-05
Jan-05
Mar-05
Apr-05
May-05
May-05
Jun-05
Jun-05
Jun-05
Source
EPA
EPA
EPA
GAO
EPA
EPA
EPA
EPA
EPA
EPA
EPA
EDSC
EDSC
EDSC
EDSC
EPA
EPA
EPA
EPA
EPA
EDSC
EDSC
EPA
EPA
EPA
EPA
EPA
EPA
EPA
EPA
Mimno, Myers,
& Holum
Mimno, Myers,
& Holum
EPA
EPA
EPA
EPA
EPA
NAPA
EPA
Title
EPA's Key Management Challenges
Standard Update
Background Information on the National Priorities List (NPL)
Federal Civil Penalties Inflation Adjustment Act
Standard Update
EPA's Strategic Plan form Homeland Security
EPA's Key Management Challenges
OIG CERCLIS Data Quality
OIG EPA Management of Information Technology Resources Under
The Clinger-Cohen Act
OIG Information Technology: EPA Management of Information
Technology Resources Under The Clinger-Cohen Act
Registry Update
Standard Data Elements for Reporting Water Quality Results for
Chemical and Microbiological Analytes
Standard Data Elements for Contact Information
Standard Data Elements for Enforcement/Compliance
Standard Data Elements for Tribal Identifier
Registry Update
EPA's Key Management Challenges
Draft Report on the Environment 2003
Draft Report on the Environment Technical Document
2003-2008 EPA Strategic Plan
Federal Facility Identification Data Standard
Standard Data Elements for Permitting Information (Final)
Enterprise Architecture at EPA
Registry Update
Health & Ecoinformatics
EPA's Key Management Challenges
OEI Celebrating Five Years of Success Accelerating Our Progress in
the Future
CDX Partner Systems
CDX Technologies
Sample TRI Report
Conformed Dimensions for the EPA Enterprise Architecture:
Implementation for the AQS Data Mart
Conformed Dimensions for the EPA Enterprise Architecture:
Recommended Implementation Strategies
AQS Data Mart Beta Testers Page
EPA's Key Management Challenges
Registry Update
Charge to the Peer Reviewers: Air and Other Relevant Indicators for
the U.S. EPA's 2007 Report on the Environment Technical Document
A Report of the Environmental Information Consortium
An Integrated Facility Identification System: Key to Effective
Management of Environmental Information at the EPA
Annual Performance Plan and Budget Overview
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 35
-------
No.
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
Date
Jun-05
Jun-05
Jun-05
Jun-05
Jun-05
Jun-05
Jun-05
Jun-05
Jul-05
Jul-05
Aug-05
Sep-05
Nov-05
Nov-05
Nov-05
Dec-05
Jan-06
Jan-06
Jan-06
Jan-06
Jan-06
Jan-06
Feb-06
Feb-06
Mar-06
Mar-06
Mar-06
Mar-06
Mar-06
Mar-06
Source
EPA/Lockheed
Martin
Planview
EPA
MetaCarta
EPA
EPA
EPA
EPA
EPA
EPA
EPA
EPA
EPA
Mimno, Myers,
& Holum
EPA
EPA
EPA
Gartner
EDSC
EDSC
IT Governance
Institute
EPA
EPA
EPA
Gartner
IT Governance
Institute
IT Governance
Institute
IT Governance
Institute
EPA
EPA
Title
Franchise Development A Strategy for Reuse of Methods and SOA
Components
It Portfolio Management In The Federal Government
The Toxics Release Inventory (TRI) and Factors to Consider When
Using TRI Data
Improve Environmental Compliance and Analysis
Substance Conformed Dimension Data Model
Substance Conformed Dimension Data Modeling Discussion Meeting
Minutes
Substance Conformed Dimension Business Rules
Substance Conformed Dimension Permissible Values List
OEI's Applied Analyses and Applications Workshop: OEI's Role in
Science and Analysis
Talking Points for Linda Travers: OEI's Applied Analyses and
Applications Workshop
Registry Update
OIG EPA Needs to Improve Oversight of Its Information Technology
Projects
U.S. EPA PERFORMANCE AND ACCOUNTABILITY REPORT
Enterprise Data Integration at EPA
FEA The Data Reference Model
Tools for Effective Program Management: Getting budget, financial and
environmental results information to EPA employees
2005 PRESIDENT'S MANAGEMENT AGENDA RESULTS REPORT
Poor-Quality Data: The Sure Way to Lose Business and Attract
Auditors
BIOLOGICAL TAXONOMY DATA STANDARD
MEASURE DATA STANDARD
COBIT 4.0: Control Objectives For Information And Related
Technology
Functions and Services Provided by CDX
Registry Update
AQS Data Mart Data Model
Enterprise Information Management: Getting Value From Information
Assets
Enterprise Value: Governance Of IT Investments: The Business Case
Enterprise Value: Governance Of IT Investments: The ING Case Study
Enterprise Value: Governance Of IT Investments: The Val IT
Framework
Goal 1 : Clean Air and Global Climate Change RESPONSES TO
STATE & TRIBAL ISSUES
GOAL 2: CLEAN AND SAFE WATER RESPONSES TO STATE &
TRIBAL ISSUES
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 36
-------
No.
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
Date
Mar-06
Mar-06
Mar-06
Apr-06
May-06
Jun-06
Jun-06
Jun-06
Jun-06
Jun-06
Jul-06
Jul-06
Aug-06
Aug-06
Sep-06
Sep-06
Sep-06
Oct-06
tbd
Source
EPA
EPA
EPA
EPA
EPA
EPA
OMB
GAO
OMB
GAO
GAO
EPA
Scottish
Executive
GAO
EPA
EPA
EPA
Lockheed
Martin
EPA
Title
EPA STRATEGIC PLAN 2006-201 1 GOAL 3 - Land Preservation and
Restoration Responses to State and Tribal Issues
GOAL 4 - HEALTHY COMMUNITIES & ECOSYSTEMS Responses to
State and Tribal Issues
GOAL 5 - COMPLIANCE AND ENVIRONMENTAL STEWARDSHIP
RESPONSES TO STATE AND TRIBAL ISSUES
EPA's Key Management Challenges
EPA 2006-201 1 Strategic Plan: Charting our Course (Draft)
AQSP&A Transaction Generator
PEA Consolidated Reference Model Document
GAO-06-669 Clean Air Act: EPA Should Improve the Management of
Its Air Toxics Program
Revision Summary Document for the FEA Consolidated Reference
Model Version 2.0
Information Technology Architecture and Systems Issues: Comments
from EPA
GAO-06-780 Paniculate Matter: EPA Has Started to Address the
National Academies' Recommendations on Estimating Health Benefits,
but More Progress Is Needed
AQS Data Mart Training
Effective Provision of Environmental Information and Advice: A
Scoping Study
GAO-06-1032T Chemical Regulation: Actions are Needed to Improve
the Effectiveness of EPA's Chemical Review Program
OEI/IESD Overview
Evolving EPA's Enterprise Architecture
2006-201 1 EPA Strategic Plan: Charting Our Course
Application Solution Architecture: Concept of Operations - Re-
alignment to Service Oriented Architecture
Implementing the Federal Enterprise Architecture Framework at EPA
January 30, 2007 Environmental Information Integration - Final Deliverable (revised)
Page 37
------- |