Special Report: O.R. & the National Science Foundation

Opportunities for research funding, current trends and new initiatives.

By Michael C. Fu and Russell R. Barton

NSF and OR

The National Science Foundation (NSF) is a $7-billion independent federal agency “charged with advancing all fields of fundamental science and engineering research and related education” (NSF Web site). Unlike the National Institutes of Health (NIH) or National Institute of Standards and Technology (NIST), NSF is not part of any of the cabinet departments. The current director, Dr. Subra Suresh, is an engineer and U.S. National Academy of Engineering member. He joined NSF in October 2010, about the same time as we did, and even as NSF director (and in his previous position as dean of MIT’s School of Engineering), he has a very active research lab.

In some fields, NSF is the primary source of research support, and in many science and engineering fields, including OR/MS, grants from NSF are viewed as the most prestigious funding mechanism. A key reason is the panel review process. Every research proposal undergoes a rigorous peer review by at least three (and usually four) research experts independently, followed by a panel discussion among the reviewers to provide recommendations and rankings of the submitted proposals. The process is highly competitive: Due to budget constraints, only about 10 percent to 15 percent of submitted proposals are funded.

In the August 2009 issue of OR/MS Today [1], program directors (PDs) Cerry Klein and Bob Smith provided an overview of NSF and its O.R.-relevant programs. Much of what was written then still holds today, so the emphasis of this article will be on providing an update on the programs, funding opportunities and recently supported projects, and on describing some current trends. Specifically, the remainder of the article will describe the different types and typical profiles of OR/MS research awards, its three core programs which support OR/MS research, other opportunities within NSF for research support and new initiatives that are being pursued that are directly relevant to the OR/MS community.

Core Program Funding

The three core programs for which the authors are responsible are Operations Research (OR), Service Enterprise Systems (SES) and Manufacturing Enterprise Systems (MES), all of which are found in the Civil, Mechanical and Manufacturing Innovation (CMMI) Division of the Engineering Directorate. In fiscal year 2011, the Engineering Directorate had an overall funding rate of 15 percent, with the CMMI success rate at 14 percent. Core programs are targeted at projects where the lead researcher is from an academic institution, generally research-oriented. There are the two main types of grants:

  • CAREER awards, for junior faculty, and
  • unsolicited proposals, which have two deadlines annually.

CAREER awards have durations of five years and in CMMI are fixed at $400,000; the submission deadline is in mid-July. Each of the three core programs usually awards from one to three CAREER awards per year. Unsolicited research awards are typically for a three-year duration in the range of $200,000-$400,000, with higher amounts for projects with multiple principle investigators (PIs). The funding for unsolicited projects generally covers one graduate student and some partial support for the primary investigator (PI) and any co-PIs, usually in terms of summer support, as well as travel. The current annual submission windows for unsolicited proposals are Sept. 1 to Oct. 1 and Jan. 15 to Feb. 15.

Related to CAREER awards, “Each year NSF selects nominees for the Presidential Early Career Awards for Scientists and Engineers (PECASE) from among the most meritorious recent CAREER awardees. Selection for this award is based on two important criteria: 1) innovative research at the frontiers of science and technology that is relevant to the mission of the sponsoring organization or agency, and 2) community service demonstrated through scientific leadership, education or community outreach. These awards foster innovative developments in science and technology, increase awareness of careers in science and engineering, give recognition to the scientific missions of the participating agencies, enhance connections between fundamental research and national goals and highlight the importance of science and technology for the nation’s future.

Individuals cannot apply for PECASE. These awards are initiated by the participating federal agencies. At NSF, up to 20 nominees for this award are selected each year from among the PECASE-eligible CAREER awardees who are most likely to become the leaders of academic research and education in the 21st century. The White House Office of Science and Technology Policy makes the final selection and announcement of the awardees.”

In 2009, one of the awardees was from the OR Program (Jose Blanchet, Columbia University). Note that the PECASE program cuts across a large number of federal agencies, including NIH, NASA and most of the Cabinet level departments.

In addition to the three core programs described below, other core programs in CMMI that could potentially fund OR/MS-related research include Engineering Design and Innovation, Civil Infrastructure Systems, Infrastructure Systems and Extreme Events, Control Systems and Dynamical Systems. Outside of CMMI, but still in Engineering (ENG), are programs in the Electrical, Communications and Cyber Systems (ECCS) Division such as Energy, Power and Adaptive Systems. Outside of the Engineering Directorate are opportunities in the three directorates: Computer & Information Science & Engineering (CISE), Mathematical & Physical Sciences (MPS) and Social, Behavioral & Economic Sciences (SBE). Some specific core programs include

  • Applied Mathematics and Combinatorics in the Division of Mathematical Sciences (DMS) of MPS,
  • Algorithmic Foundations in the Computing and Communication Foundations (CCF) of CISE,
  • Robust Intelligence and Information Integration and Informatics in the Information and Intelligent Systems (IIS) Division of CISE, and
  • Decision, Risk and Management Science in the Social and Economic Sciences (SES) Division of SBE.

The Operations Research program supports fundamental research leading to the creation of innovative mathematical models, analysis and algorithms for optimal or near optimal decision-making in the design and operation of manufacturing, service and other complex systems. The research focus of the OR program is on methodology and theory, broadly in the traditional areas of discrete and continuous optimization and stochastic modeling & analysis, primarily motivated by problems in engineering, including those arising from the transportation sector, services such as healthcare systems and energy, manufacturing, supply chain management, civil infrastructure and many others. Algorithmic development and rigorous mathematical analysis are key aspects of the research. The models, theory and algorithms that are developed in the research programs serve as critical foundations for the design and operation of civil, mechanical, industrial and manufacturing engineering systems in a safe and efficient manner, thus enabling increased productivity and economic growth. A particular recent interest is dynamic (multi-stage) decision-making under uncertainty. New research thrusts complementing this focus include simulation optimization, approximate dynamic programming, stochastic programming and self-optimizing systems that can observe, learn and adapt to changing environments. Recent workshops reflect these priorities.

The OR program has a normal base budget of between $5 million and $6 million. Over the past six years, the OR program has received between 100 and 140 proposals per year and has funded between 14 and 30 “normal” (i.e., completely panel reviewed) per year, with the low occurring in 2006 and the high occurring in the year with “stimulus” (American Recovery and Reinvestment Act of FY 2009) funds. In the past few years, the program directors (PDs) have strived to fund a larger number of projects with smaller budgets. The amount actually funded has been based on a number of factors, including how highly the panel rated the proposal, how much graduate student support is requested and how much funding the PI(s) already have (both from NSF and from other sources). Table 1 lists many of the research projects funded over the past couple of years.

NSF and OR

Table 1: Examples of recently funded projects in the OR Program.

The Service Enterprise Systems program supports research on strategic decision-making, design, planning and operation of commercial, nonprofit and institutional service enterprises. While any service enterprise might be considered, there is particular focus on healthcare and other similar public service institutions. More than half of the proposal (and half of the awards) fund research in healthcare systems engineering. The SES program emphasizes research topics leading to more effective systems modeling and analysis as a means to improved planning, resource allocation and policy development. It should have an impact on the analytical and computational techniques relevant to extended enterprise operations – that is, it should involve novel economic, mathematical and probability modeling that can be used to help managers make better decisions.

Service Enterprise Systems and Manufacturing Enterprise Systems (described next) are distinguished from the OR program by their application focus. For SES, focus on a specific service enterprise situation should suggest potential model and algorithm innovations. For SES and MES, reviewers have greater concern that the models reflect the operation of real systems, so it is important to demonstrate model fidelity. For the same reason, the proposed research should offer the prospect of implementable solutions or implementable policy: for SES and MES, theory without demonstrated relevance to practice does not fare well in the review process. Table 2 shows highlights for the most commonly funded SES areas.

  • Healthcare: Research in this area creates mathematical models, both stochastic and deterministic, to provide insight and decision support for healthcare delivery and healthcare policy. These include transplant policy, hospital scheduling, vaccine design, radiation treatment, and health screening methods and policies.
  • Supply Chain and Logistics: Research in this area creates new models used for design and execution of energy and other service delivery supply chains.
  • Finance, Energy and Other Markets: Research in this area creates mathematical models to predict the behavior of markets, and includes techniques such as game theory, economics and stochastic processes.
NSF and OR

Table 2: Examples of recently funded projects in the SES Program.

The SES program’s base budget is approximately $5 million and in recent years has funded 14-18 proposals each year out of 115-130 submissions. This year, funding levels (and the number of proposals submitted) will likely be higher because of two special healthcare initiatives. Up to an additional $1.5 million in funds will be available for research proposals meeting the requirements of the Advancing Health Services through Systems Modeling Research, NSF 12-515. Based on the outcome of a 2009 workshop funded by NSF and the Agency for Healthcare Research and Quality, this announcement seeks to fund research collaborations between industrial and systems engineers and health services researchers. A second initiative is the Smart Health and Wellbeing solicitation (NSF 12-512). Three directorates offer it jointly: Engineering, Computer and Information Sciences and Engineering, Social, Behavioral and Economic Sciences. This solicitation will provide approximately $15 million in funding for research to address fundamental technical and scientific issues to move healthcare from reactive and hospital-centered to preventive, proactive, evidence-based, person-centered and focused on wellbeing. Proposals are likely to be collaborations of researchers in OR/MS, computer science, social science, engineering and medicine.

The Manufacturing Enterprise Systems program supports research on design, planning and control of operations in manufacturing enterprises. Research is supported that impacts the analytical and computational techniques relevant to extended enterprise operations including planning, monitoring, control and scheduling of manufacturing and distribution operations, and development of methods for optimization of manufacturing enterprises in the presence of a high degree of uncertainty and risk. As for SES, the research should offer the prospect of implementable solutions. Table 3 shows highlights for the most commonly funded areas of MES.

  • Supply Chain Management: Research in this area creates mathematical models, both stochastic and deterministic, to provide insight and decision support for supply chain issues in the manufacturing enterprise. These include inventory behavior and ordering decisions, transportation of supplies and products, product distribution and logistics modeling and decision support.
  • Manufacturing Process Modeling and Quality Monitoring: Research in this area models the behavior of manufacturing systems to allow the study and development of better scheduling and assembly methods. Related topics include the monitoring of quality of manufactured products and adjustment of processes to maintain high quality.
  • Reliability and Maintenance: Research in this area creates probability models for equipment and products that enable cost-effective maintenance and repair activities. Some research uses environmental and sensor information as input to probability models used in (maintenance and repair) decision support.
NSF and OR

Table 3: Examples of recently funded projects in the MES Program.

The MES program’s base budget is approximately $4 million and in recent years has funded 16 to 27 proposals each year out of 95 to120 submissions.

Russell R. Barton (rbarton@nsf.gov) is the Program Director for Service Enterprise Systems and Manufacturing Enterprise Systems, and Michael C. Fu (mfu@nsf.gov) is the Program Director for Operations Research at the National Science Foundation. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Editor’s note:
This is the first of a two-part series on the NSF and its relevant programs and funding opportunities for operations research. In the second part of the series, set for the June issue of OR/MS Today, authors Russell Barton and Michael Fu will continue to focus on trends and new initiatives, as well as how to write an effective proposal, while answering the frequently asked question: Where does my proposal fit?

Reference

1. Klein, C.M. and Smith, R.L., August 2009, “O.R. & the National Science Foundation,” OR/MS Today, Vol.36, No.4, pp.32-37.