Dr. Rita Colwell
Chairman Smith, Ranking Member Johnson, members of the Research Subcommittee, I appreciate the opportunity to be here today to discuss the National Science Foundation's management and oversight of large facilities. My colleagues, Dr. Boesz and Dr. Jones, will also provide comments.
Throughout its 50-year history, NSF has enjoyed an extraordinarily successful track record in providing state-of-the-art facilities for science and engineering research and education. NSF management and oversight have enabled not only the establishment of unique national assets, but have also ensured that they serve the science and engineering communities and the discovery process as intended.
NSF has enabled construction of unique and scientifically productive telescopes and managed their operation by funding non-governmental institutions created for that purpose. For decades, the University National Oceanographic Laboratory System of research ships has made outstanding contributions to multidisciplinary research and education worldwide and at the most remote regions of the Earth. The Deep Sea Drilling Project and its successors have operated with a truly exemplary record of scientific discovery and contract management.
For more than 40 years NSF has managed the Nation's research program in Antarctica. Since the 1970s it has been widely recognized for its impressive management of the overall U.S. Antarctic Program that now encompasses planning and budgeting for logistics, facilities, and research support. More recently, NSF has enabled major advances in building and managing large-scale computational infrastructure supporting virtually all fields of science and engineering. Last month NSF awarded $53 million to a consortium of universities and institutes for a distributed terascale computing system.
NSF's approach to facilities management differs from most R&D agencies. NSF does not directly construct or operate the facilities it supports. Typically, the Foundation makes awards to external entities, primarily universities, consortia of universities or non-profit organizations, to undertake construction, management and operation of facilities. NSF retains responsibility for overseeing the development, management and successful performance of the projects. NSF decision-making not only involves establishing new facilities but also reconstituting or phasing out support for facility projects that have reached the end of their useful lives.
These partnerships have proven remarkably effective in terms of overall cost and performance. The reasons include the talented program officers at NSF, the great creativity of the Nation's academic enterprise and the "enlightened self-interest" of a community wishing to use the facilities for research. An essential added benefit of NSF's model is the opportunity to train young scientists and engineers by allowing them direct involvement in planning, construction and operation of major facilities and large-scale instrumentation.
Currently, NSF invests over $1 billion annually in facilities and other infrastructure projects. Over time, the portfolio of facilities has grown and diversified. For example, our portfolio now includes distributed projects that challenge traditional management and oversight approaches. With emerging multidisciplinary science and engineering opportunities, large facility projects are becoming increasingly complex and more challenging technical and management issues are on the horizon.
Growth and diversification require that NSF remain attentive to the ever-changing issues and challenges inherent in planning, construction, operation, management and oversight of large facility projects. NSF's current review, management, and oversight processes are designed to "build right those facilities that are the right ones to build." Today I will describe some additional steps we are taking to improve coordination, collaboration, and learning among NSF staff and external partners in order to facilitate a process of continuous improvement, based upon the learning of best practices.
Priority Setting: "Is it the right facility to build?"
Many of NSF's large facility projects are supported through the Major Research Equipment, or MRE, account - an agency-wide capital asset account that funds the construction or acquisition of projects with costs ranging from tens to several hundreds of millions of dollars. Large facility projects are also funded through the Research & Related Activities (R&RA), and/or the Education and Human Resources (EHR) accounts.
Currently, NSF is funding the following MRE projects:
- Atacama Large Millimeter Array R&D (ALMA)
- High-performance Instrumented Airborne Platform for Environmental Research (HIAPER)
- Large Hadron Collider (LHC)
- George E. Brown Network for Earthquake Engineering Simulation (NEES)
- Polar Support Aircraft Upgrades
- South Pole Station Modernization (SPSM)
- Terascale Computing Systems
The following projects have been included in prior budget requests and remain scientific priorities:
- ALMA Phase II ($9 million for maintenance of the established infrastructure for ALMA was included in the Research and Related Activities account in our FY 02 request, pending a review of overall facilities management issues.)
- EarthScope I (US Array and San Andreas Fault Observatory at Depth)
- National Ecological Observatory Network (NEON)
Finally, the following projects have been approved by the National Science Board and are priorities for consideration in future budget requests:
- EarthScope II (Plate Boundary Observatory)
- Ice Cube Neutrino Detector
- Ocean Observatories
- Rare Symmetry Violating Processes
I would now like to explain how these and other candidate MRE projects are reviewed, prioritized and eventually selected for inclusion in NSF's budget request. (This process is more fully described in Guidelines for Planning and Managing the Major Research Equipment Account, which provide the basis for budgeting for MRE facilities, and can be found at http://www.nsf.gov/home/about/mre01.html.)
First I would like to describe the criteria NSF uses to select large facility projects for inclusion in a budget request and to determine which projects it will eventually fund. These criteria are listed below. The first two are the general NSF merit review criteria used in the merit review process. The next six are additional criteria that NSF and the NSB use to review and approve projects for inclusion in a budget request. The relative importance of these criteria is not fixed but depends upon the nature of the facility project and the changing context that envelops it.
- Intellectual Merit
- Broader Impacts
- Importance to science and engineering
- Cost-benefit and risk analysis
- Readiness to implement and manage
- Appropriateness for NSF
- Balance across fields and disciplines
- Synergy with other large facilities supported by NSF, other agencies, and other nations
The chart below depicts NSF's process for the review and approval of large facility projects considered for funding through the MRE account. Several steps are involved as a project moves from an idea to a line in the NSF annual budget.
The first step in the process is the early identification of an MRE Pipeline Project as such. These potential projects are conceived of in the science and engineering community, often as a result of emerging opportunities, and are often many years in development following initial conceptualization. Currently, NSF has about 20 projects in the discussion phase. These projects span the range from "gleam in the eye" to projects that are in the conceptual development stage.
At the conceptual development stage, an NSF Program Manager works with the scientific and engineering communities to develop a shared understanding of the project's concept, resource requirements, and schedule and performance issues. As the project moves toward a more formal definition, the NSF Program Manager consults with members of the scientific and engineering communities to help refine project design and definition and to formulate scientific/technical and management plans. It is important to note that, although NSF may be providing some funding for early planning and definition, it has not made any commitment to support the establishment of the facility.
Developed projects are then proposed by an NSF Originating Organization(s) for consideration by the MRE Panel. These projects may be based on a proposal already submitted and evaluated using NSF's merit review process. When an Assistant Director believes a project is ready to be considered for inclusion in an NSF budget request, it may be brought to the MRE Panel for consideration. The project is considered by NSF's MRE Panel, which is chaired by the Chief Operating Officer and includes the Assistant Directors, Program Office Heads and the Chief Financial Officer. It is during these discussions that issues such as appropriateness for NSF, balance across fields and disciplines and readiness to implement and manage are first addressed.
The MRE Panel considers the projects on the basis of the review criteria specified earlier and makes recommendations to the Director. Using the review criteria, the Director selects candidates for the NSB to consider during one of its five meetings throughout the year. Once a project has been approved by the NSB, the Director, at her discretion, may recommend the project for inclusion in a future budget request to OMB. The project is generally considered a viable candidate until it is funded. However, significant changes to project definition, scope, and funding must be reviewed and approved by NSF and the NSB.
Once funded, changes in project cost are governed by a Delegation of Authority from the NSB. Under this authority, the Director may subsequently amend the award to commit an additional 20 percent or $10 million, whichever is less. Any funding that exceeds this requires additional NSB review and approval. Our policies also require that the Director be notified in writing, by the cognizant Assistant Director, prior to the expenditure of additional funding. NSB-approved awards are now coded in our electronic proposal and award system, and we have implemented an automated alert -- for program managers and grants officers -- identifying the awards and the NSB delegation conditions.
OMB gives NSF budgetary guidance that reflects the Administration's priorities. The Director submits a proposed budget to OMB within the guidance - a budget that reflects priorities established in conjunction with the NSB. OMB and NSF then enter into a negotiation phase, which results in the priorities outlined in the final funding request. Following the appropriations process, the Director allocates funds to the relevant projects. If necessary, a program solicitation or RFP is prepared and, following receipt and merit review of the proposals, one or more awards are made.
This is the overall process by which large facility projects are selected and funded. It is important to note that during this process candidate facilities are not given specific priority ranks. In any given year, the MRE projects that appear in the President's budget request will depend on NSF's current assessment of opportunities and needs and the parameters provided by OMB. The highest priority projects appear in the budget request.
This process is very open and involves a great deal of interaction with the external scientific community and other stakeholders. This selection process has served NSF and the nation well in providing the most competitive state-of-the-art science and engineering facilities -ones that are certain to contribute significantly to the nation's research and education endeavors.
Facilities Management And Oversight: "Is it built right?"
Given the increasing complexity and scope of its facilities, NSF has recognized the need to mitigate risks by ensuring management and oversight benefit from contemporary best practices. Improving coordination, collaboration and shared learning among NSF staff and external partners facilitates identification and incorporation of best practices. To this end, and to comply with instructions in A Blueprint for New Beginnings: A Responsible Budget for America's Priorities (February 2001), NSF developed the Large Facility Projects Management & Oversight Plan.
The Plan outlines NSF's goals and strategies for integrating its current procedures and processes into a next-generation system for selecting, managing and overseeing large facility projects. In answering the President's request, the Plan focuses on the management and oversight of large infrastructure and facility projects.
In its Plan, NSF addresses improvements in four areas:
- Organizational and staff capabilities enhancement to improve coordination, collaboration and shared learning among NSF staff and external partners,
- Comprehensive guidelines and procedures for all aspects of facilities planning, management and oversight,
- Review and approval of large facility projects, and
- Coordinated and proactive oversight of all facility projects to ensure success.
The draft Plan has been reviewed by OMB, NSF's Assistant Directors, the Office of Inspector General, and the NSB. On September 10th, the finalized plan will be submitted to OMB.
The Plan preserves clear lines of authority, responsibility and communication from the NSF Director to the NSF Program Manager to the awardee Project Director. In every large facility project, the NSF Program Manager exercises primary responsibility for all aspects of project management. Working closely with the NSF Program Manager, the awardee designates one person - with strong management experience - as the Project Director, who has overall control and responsibility for the project in the awardee organization. NSF vests responsibility for monitoring the business operations of large facility projects in its Chief Financial Officer (CFO). NSF personnel reporting to the CFO ensure that all policies, guidelines and procedures are followed and that the awardee is in compliance with the business, operations and financial requirements.
To enable the efficient evolution of NSF's large facility projects from their pre-formulation through operations, NSF will establish a new position for a Deputy, Large Facilities Projects. The LFP Deputy will report directly to the CFO. Through a nationwide search, NSF will recruit an individual with extensive experience in the building, management and oversight of large science and engineering facilities.
The LFP Deputy will be responsible for:
- Providing expert assistance to NSF's science and
engineering staff on non-scientific aspects of
project planning, budgeting, implementation, and
- Developing, implementing and managing, with NSF-wide
input and concurrence, management and oversight
policies, guidelines and procedures; and,
- Ensuring shared learning of best practices by fostering coordination and collaboration throughout NSF to facilitate application of lessons learned from each project.
The LFP Deputy will be supported by permanent NSF staff with experience in project management, planning and budgeting, cost analysis and oversight. The personnel will be a centralized resource assembled to assist NSF Program Managers with their management and oversight responsibilities and to develop and conduct post-award oversight of business operations, financial and internal control systems, and cost and schedule performance. The LFP Deputy and his or her staff will provide a watchful eye over all of the MRE projects and many of NSF's other facility projects to ensure that potential challenges are identified early and handled quickly and effectively.
I would like to invite members of the committee or your staff to attend the best practices workshop for managing large facility projects that we have scheduled for September 21st. The workshop will focus on the development of comprehensive guidelines for NSF's management and oversight of large facility projects. Representatives from NSF and Federal agencies involved in the management of large research facilities, including Department of Energy and NASA, and managers of existing NSF and Federal facilities will share their experiences in construction and operation of the facilities for which they are responsible.
As I have stated, we submit the NSF's Large Facility Projects Management and Oversight Plan to OMB on September 10. This Plan includes an aggressive schedule for developing and implementing each of the major components. We have worked closely with the National Science Board in the development of the plan, and will continue that close collaboration in its implementation. We are pleased with the dialogue we have had with this Committee, and with the OMB.
Mr. Chairman, in concluding, NSF is keenly focused on maintaining excellence in selecting, managing and overseeing large facility projects for science and engineering. Science and engineering facilities are fundamental and a much needed resource for our country. They expand the frontier of knowledge and are highly sophisticated in construction and operation. We are confident our current processes and the measures we have formulated clearly demonstrate NSF's leadership in planning, executing and overseeing a growing number of complex, large projects for science and engineering for our Nation.