ADB’s Successor—The NIH Business System
If you purchase a pipette, prepare travel papers, pay bills, move a piece of equipment, or seek the advice of an outside expert, you use a facility called the Administrative Data Base (ADB). The ADB handles approximately 150,000 transactions a day for inventory, procurement, property, travel, human resources and central accounting. "You won’t see the ADB credited on papers published in Cell or Nature, but it has supported the daily activities that are associated with NIH’s progress during its growth years. It has served NIH well," said Colleen Barros, executive officer at the National Institute on Aging. Barros was chosen by NIH Deputy Director for Management, Tony Itteilag, as head of the project to update NIH’s administrative computing system.
"The business community at NIH knows that the ADB has been its lifeblood but realizes that a next-generation version is overdue," said Barros. "With electronic commerce and the World Wide Web in our everyday lives, the situation is in need of serious attention. Personal computers, the ‘mouse’ that sits on our desk, the ‘dot-com’ revolution, and buying airline tickets online—none of those existed when the ADB was designed, and it would take significant resources to upgrade to today’s standards." The effort to implement a new administrative computing system—the NIH Business System (NBS)—to replace the ADB is expected to take four years and cost an estimated $50 million. "We all realized that a ‘fix’ would be very expensive and very challenging," noted Barros. But there will be offsets between the new and old system that will help mitigate total costs.
The Analytical Phase Is Almost Over
In September 1999, Phase I of the project began—writing a "business case" for the substantial investment in NBS. This has involved a study of system requirements by almost 200 people from all parts of the NIH administrative and business communities under the guidance of Tony Itteilag, who chairs the NBS Steering Committee. (The steering committee’s members include executive officers, the scientific community and OD functional managers.) The purpose of this complex effort has been to identify requirements for the new system, as well as to recommend the best avenues for meeting those requirements. Because the transition will affect the daily lives of many hundreds of NIHers and because the ADB is a critical resource, the implementation team is being careful to consider the risks of moving from an old to a new system. The careful planning and implementation for this project will minimize the risk that can occur with the introduction of changes into a stable environment.
During the Phase I study, two choices quickly emerged—to develop a proprietary, customized, or "homegrown" application such as the ADB, or to buy a commercial product known as an enterprise resource planning (ERP) package. CIT investigated the feasibility of evolving the ADB into a next-generation system and presented a proposal for the proprietary system. Major vendors responded with proposals and demonstrations of their ERP packages, which were carefully evaluated by the NBS workgroups and the steering committee.
The timely completion of this analysis attests to the broad participation and hard work of many people throughout the NIH community—the business, scientific and IT communities. "All involved are to be congratulated for their commitment and seriousness of purpose. They worked extremely hard, while simultaneously doing their regular jobs," said Barros.
The first phase is now almost concluded. The project steering committee has reviewed the Phase I "business case" and made a recommendation to Tony Itteilag. Over the summer, the NBS "implementation plan" will be drafted, including plans for acclimating NIH staff to the new system and communicating the changes that will affect peoples’ daily lives, as well as for staffing, organization, governance, deployment and evaluation. Actual implementation of the plan will begin in the fall, using as much of the business community staff and help as possible. The project goal is to get the system up, do the testing and the piloting, and have it deployed as a stable system by 2004.
ADB Staff Saw the Problem First
"The ADB has been an absolutely critical piece of NIH infrastructure—a really first-rate product and service. For its time, it was cutting-edge," Barros explained. Built in the 1970’s, the ADB was based on the COBOL computer language and IMS database software, and now includes 1.5 million lines of code. In spite of the new features added over 22 years of existence, the ADB has become technologically obsolete, cumbersome, and expensive to maintain even as information technology has exploded. Moreover, the engineers who built the ADB are retiring, and just four or five people now support the ADB.
Barros credits the people who invented the ADB with being the first to address its shortcomings. Commercial products have become more robust and integrated than the ADB. When it became obvious that the technological rate of change was faster than the ABD system could keep up with, Emmett Ward and his staff were the first to propose a change. They pressed for the consideration of alternatives—they looked for what would serve the community best, just as they had done in the 1970’s.
Benefits to the NIH Business Community
Just as the public is doing more shopping and banking online, NIH is benefiting from the boom in e-commerce. NIH scientists want goods and services in the fastest, cheapest way possible and need an administrative system with built-in e-commerce capabilities, automated data analysis, tracking, financial reconciliation, and reporting functions that the ADB cannot deliver without major rebuilding.
Whatever the choice for NBS, the process of finding a replacement for the ADB has provided an opportunity to learn how the private sector manages business functions. "This project has been an incredible catalyst, getting people to talk and think in new ways. It has sparked a lot of debate, and I expect we will have a lot more," said Barros. The NIH business community is now aware of a number of "best practices"—better business processes that have arisen in private business. Many of those best practices are now embedded in the requirements for the new NBS system.
NIH is one of the first in the government to put a concentrated effort into evaluating commercial ERP products. Success will depend on many factors, but most of them are related to change in business practices rather than the technology of ERP.
With thanks to Richard McManus for permission to use portions of Interface 214 (June 23, 2000)
"Campus’ Major Unseen Construction Project—NIH Business System,"
from the NIH Record, April 18, 2000.