Systems
Development
Center
INITIAL REPORT ON THE
FEASIBILITY OF AN
EPA SYSTEMS
DEVELOPMENT CENTER
EXECUTIVE SUMMARY
EPA
220/
1989.21
SEPTEMBER 29,1989
-------
-------
FEASIBILITY STUDY
FOR
SYSTEM DEVELOPMENT CENTER (SDC) IMPLEMENTATION
EXECUTIVE SUMMARY
CONTRACT | 68-01-7444
DELIVERY ORDER # 7444-006
PREPARED FOR:
OFFICE OF INFORMATION RESOURCES MANAGEMENT
ENVIRONMENTAL PROTECTION AGENCY
401 M STREET, SW
WASHINGTON, DC 20460
SUBMITTED BY:
COMPEX CORPORATION
5500 CHEROKEE AVENUE
SUITE 500
ALEXANDRIA, VA 22312
SEPTEMBER 29, 1989
HEADQUARTERS LIBRARY
BMRONMBfttl PROTECTION MOW
WASHINGTON. D.C. 20460
-------
-------
EXECUTIVE SUMMARY
This document presents the feasibility of establishing the EPA
Systems Development Center (SDC), a centralized facility to manage,
support, and coordinate the system life cycle (SLC) in the
development of higher quality, cost-effective, efficient and
maintainable computer systems. The establishment of the SDC is in
accordance with the Systems Modernization Initiative (SMI), which
directs the EPA to build and renew systems for more accessibility
and usability by a growing population of information users.
The Feasibility Study
A feasibility study was conducted by implementing a
prototype development center and three pilot projects of differing
development types within the SDC structure. The pilot project team
was modelled as envisioned in the operational SDC environment:
staffed with the EPA Office of Information Resources Management
(OIRM) desk officers, contract developers and EPA program office
information users. An illustration of this model is presented in
Figure 1. A structured development methodology was implemented to
test a rigorous, standardized, information-oriented approach to
the system life cycle. James Martin's Information Engineering
Methodology (IEM) and Texas Instrument's (TI) Information
Engineering Facility (IEF) were selected as the methodology and
associated Computer-Aided Software Engineering (CASE) tool
2
-------
Mode I name: SDC
Subset name: RLL
count
RCTOH ' • -M
•pfi^iiffrf
.j
HL
>_>
i.V
11*
|
G
S
JDELIVER
Sept. 29. 1389
en
EKD
P.-ige: 1
-------
to be tested in the pilot studies because of their orientation
toward defining the business structure and user requirements
through informational (data) flow, the full implementation of the
system life cycle, and the well-defined framework for establishing
enforceable development standards.
Benefits to EPA
This Feasibility Study documents the benefits of the SDC
prototype and the IEM implementation, outstanding technological
issues, and SDC implementation considerations as experienced in the
pilot projects. These results will help guide management in their
evaluation of SDC effectiveness in building better, more usable
systems and help determine acceptance and further implementation
of the SDC.
Accomplishments of the Pilots
Using the structured methodology techniques, the project team
accomplished the development of two microcomputer information
systems and the high-level analysis of an existing, well-used water
quality system (Figure 2).
The STORET water quality system was defined and described; its
user profile and information needs were identified; and a logical
model of the water quality enterprise was developed. This
-------
FIGURE 2
PROJECTS
CLIENTS
are supported by
DESK
OFFICERS
STORET
PRAMS
MWIMP
-------
constituted the initial assessment of the strategic planning stage
within the methodology and provided management with an information
architecture from which systems can be built. The Paperwork
Reduction Act Management System (PRAMS), an existing minicomputer
system supporting information collection request management for the
Information Policy Branch (IPB), was constrained by the existing
hardware technology. The team redesigned this system and enhanced
it based on the IPB needs defined in the detailed business area
analysis of the methodology. The system is in the testing stage
of development at this time. The Medical Waste Information
Management Project (MWIMP) is a new EPA program in compliance with
both the Medical Waste Tracking Act (MWTA) of 1988 and the
resulting EPA program to monitor this process for two years in five
states. A new microcomputer system is being developed to
accommodate this need. The project is completed through the design
stage and is expected to be implemented in December 1989.
Summary of Findings
Results generated from the pilot projects identified several
benefits in accordance with SDC goals.
One benefit has been the consistent demonstration that
Information Engineering answers the objectives for the EPA
information needs management. The models developed in the pilot
studies provided a sound base for the understanding of the business
-------
areas represented. The strategic planning stage of Information
Engineering, analyzing information needs from a high-level
management view and independent from current computer systems, may
provide EPA a new perspective with which to define information
systems more applicably across functional enterprises and less
focused within a program unit.
The IEF CASE tool provides rigid enforcement of consistent,
uniform methods and comprehensive automated consistency checks for
high-level quality assurance. Developers found that the step-by-
step progression of the early SLC stages, in conjunction with early
prototyping, enforced the front-end analysis often omitted or
reduced due to time constraints. Early problem identification,
increased user understanding, and the automated environment made
the development environment more effective.
There is confidence among the program office information users
that the systems being developed will better meet their business
needs and be more usable. This was attributed to the standardized
diagrams and techniques, providing understandable graphics and
increased detail for the program office information users in
describing the prospective system; increased communications with
the SDC project team; and the review and approval process (Figure
3) enforced within the methodology and CASE tool.
-------
00
I
SUBMITS
CLIENT
OVERSEES
IS SUPPORTED BV
1
APPROVES
REGION
PROGRAM
OFFICE
Model name: SDC
name: flLL
PROJECT
REQUEST
w
1
Sept. 29. 1983
12:-18
EHD
Page: 1
-------
Experience Gained
Experience gained from the pilot studies and findings
documented in this report are important to the SDC's implementation
planning. Areas of consideration include the development of
customized methodology training particular to the program office
staff; the applicability and use of the methodology for differing
project types (i.e., system size and longevity); and the need for
continued high-level technical support throughout the learning
process. The project team structure, as effected in the SDC
prototype, was determined to be a stabilizing and supporting
influence for the program office users. The SDC (Figure 4) is an
effective alternative to the current development team structure.
Recommendations to EPA
The recommendations, as determined from the implementation of
the SDC prototype and associated pilot studies, are as follows.
The SDC should continue to pursue the use of the information
Engineering Methodology, as it provides an information model
applicable to the EPA mission and business areas; it implements and
enforces the full system life cycle tasks; and it provides a strong
framework for the SDC to standardize the EPA development
environment.
-------
FIGURE 4
THE SOC FUNCTIONS
SDC
i
I
/"Product N /'TechnicalN
I Development j I Support I
r
Training
Standards "\ fSDC
I Data
V^Administratio
/^Plan ^ /^Analyze N /^Design ^N /^Construct *N
10
-------
The SDC should continue to search for the "best fit" set of
automated CASE tools for their needs. Although IEF is very strong
in maintaining and enforcing the methodology structure and
consistency rules, it may not be adaptable enough to handle
integrating the needed prototyping and lower-end (coding and
testing) tools needed to support EPA's current systems environment.
Alternative CASE tools that support Information Engineering should
be piloted and compared with IEF. In conjunction with the CASE
tool, automated project management alternatives should be evaluated
as an external workstation component. A combination of tools
should be identified and implemented within the workstation. At
the same time, advances in new technology, continued CASE tool
integration, and alternative target languages, as developed within
the current IEM CASE tools, should be monitored and evaluated for
use in the SDC as they become available.
The initial findings of the pilot projects show that the
implementation of the structured methodology at UPPER (planning,
analysis and design stages) CASE levels has the potential for
building a better, more usable system. The development projects
should continue to be monitored for quality assessment based on
testing, maintainability with change, and productivity resulting
from the front-end effort.
11
------- |