RAND Corporation

Larger RAND Logo

Introduction

As World War II was ending, a number of individuals, both inside and outside the U.S. government, saw the need for retaining the services of scientists for government and military activities after the war’s end. They would assist in military planning, with due attention to research and development.

Accordingly, Project RAND was established in December 1945 under contract to the Douglas Aircraft Company. The first RAND report was published in May 1946. It dealt with the potential design, performance, and use of man-made satellites. In February 1948, the Chief of Staff of the Air Force approved the evolution of RAND into a non-profit corporation, independent of the Douglas Company. On May 14, 1948, the RAND Corporation was incorporated as an independent non-profit organization, and on November 1, 1948 the Project RAND contract was formally transferred from the Douglas Company to the RAND Corporation.

The Articles of Incorporation set forth RAND’s purpose: “To further and promote scientific, educational, and charitable purposes, all for the public welfare and security of the United States of America.” It accomplishes this purpose by performing both classified and unclassified research in programs treating defense, international, and domestic issues. In 2019 staff numbered nearly 2000 researchers and support persons, with about 22% of the researchers being operations researchers, mathematicians, physical scientists, engineers, and statisticians.

For much of its history, RAND’s research departments were discipline based (e.g., mathematics, economics, physics, etc.). However, research is now carried out by units that address social and economic policy issues both in the United States and overseas; by federally funded research and development centers (FFRDCs) that focus on national security; by professors and graduate fellows at the Pardee RAND Graduate School; and by RAND Europe and RAND Australia, independently chartered  affiliates.  In 2019 it had six North American locations: Santa Monica, CA;  Arlington, VA; Pittsburgh, PA; Boston, MA; San Francisco, CA;  and New Orleans, LA (the Gulf States Policy Institute).

In this article, the focus is on RAND’s contributions to the theory and practice of operations research. However, RAND has also made major theoretical and practical contributions in other areas, including engineering, physics, political science, health, and the social and behavioral sciences. A broader and more comprehensive history of RAND’s early years is contained in Jardini (1996).

The First Ten Years (1948–1957)

The first decade saw RAND accomplishments ranging from the beginning of the development of systems analysis, which evolved from the earlier more specific and more narrowly focused military operations analysis of World War II, to the creation of new methodological concepts and techniques to deal with problems involving many variables and multiple objectives.

Systems analysis may be defined briefly as the systematic examination and comparison of alternative future courses of action in terms of their expected costs, benefits, and risks. The main purpose of systems analysis is to provide information to decisionmakers that will sharpen their intuition and judgment and provide the basis for more informed choices. From the beginning it was evident that to be successful, systems analysis would require the conception and development of a wide range of methodological tools and techniques. One of the most important sources of these tools and techniques was the emerging discipline of operations research.

In the early 1950s Edwin Paxson led the project that produced a report entitled Strategic Bombing Systems Analysis (Paxson 1950), which is generally regarded as the first major application of the concept of systems analysis, as well as the source of the name for the new methodology. Among other things, the report advocated the use of decoys to help mask bombers from enemy defenders. This study was a catalyst that stimulated the development and rise of a number of analytical methods and techniques. Some of the more important examples are:

  • Game theory — Mathematics and game theory were prominent subjects in the early research agendas of Project RAND. Lloyd Shapley, J.C.C. McKinsey, Melvin Dresher, Martin Shubik, Rufus Isaacs, and Richard Bellman were among the numerous early RAND contributors to this area, while John Williams and Herman Kahn played an important role in popularizing some of the simpler aspects of game theory (Williams 1954). John von Neumann, who is often cited as the father of game theory, and Oskar Morgenstern, who linked game theory to economic behavior, were active RAND consultants, as were many others with connections to major universities.
  • Enhanced computer capabilities — The Paxson project required computer capabilities beyond those available at that time. This stimulated developments that led to the building of the JOHNNIAC digital computer, which became fully operational in the first half of 1953.  Based on a design by John von Neumann, it was one of the six “Princeton class” stored programming machines.  Due in part to the unreliability of the original JOHNNIAC memory for lengthy computations (Gruenberger 1968), RAND contracted with a subsidiary of Paramount Pictures for the world’s first commercially available magnetic-core store in early 1953 (Ware 2008).  Later that year, an experimental core memory, designed by Jay Forrester, was installed in MIT’s Whirlwind computer (Redmond & Thomas 1975).  The JOHNNIAC core memory was installed in early 1955 (Ware 2008), the year in which William Orchard-Hayes implemented a Linear Programming (see “The Second Ten Years” below) package on the JOHNNIAC that was used for 10 years (Orchard-Hayes 1990).  After using the JOHNNIAC to implement the first distributed on-line time-shared computer system (1960), RAND built the JOHNNIAC Open Shop System (JOSS), the first online interactive computer system for individual users.
  • Dynamic Programming — The Paxson project also demanded the examination, through dynamic programming, of key strategic bomber components (e.g. decoys) in the context of an overall enhanced strategic capability. This, along with the demands of other projects in the early 1950s, provided a significant part of the motivation for the development of the mathematical theory of dynamic programming. Richard Bellman, together with a few collaborators, almost exclusively pioneered the development of this theory. The first RAND report on dynamic programming was published in 1953. Bellman’s well-known book (Dynamic Programming) followed in 1956, and his book with Stuart Dreyfus (Applied Dynamic Programming) was published in 1962.

A second large systems analysis study of this period was a study led by Albert Wohlstetter on the selection and use of strategic air bases. It developed basing and operational options for improving the survivability of SAC forces, and helped shift the focus of strategic thinking in the United States toward deterrence based on a secure second-strike force.

Another major effort beginning in the 1950s that led to the development of operations research tools was research on logistics policy issues. RAND’s involvement with Air Force logistics stressed the demand for spare parts and the need for logistics policies that could cope with demand uncertainty. Major players in this effort were Stephen Enke, Murray Geisler, James Peterson, Chauncey Bell, Charles Zwick, and Robert Paulson. The key analytical issue here was the examination of alternative policy issues under conditions of strategic uncertainty. Early research used “expected value” analysis. Later, RAND researchers developed and used more sophisticated methods, such as:

  • The use of sensitivity analysis to determine what areas of uncertainty really matter in final outcomes
  •  Iteration of the analysis across several relevant future scenarios to seek problem solutions that are robust for several of the possible (uncertain) scenarios. (Herman Kahn is often cited as the father of scenario planning.)
  • Given the outcomes of the above, design R&D activities that will (1) reduce key areas of uncertainty, (2) provide hedges against key uncertainties, (3) preserve options for several possible courses of action, any one of which might be used when the future environment becomes less uncertain.

RAND’s first decade also witnessed the development of a number of methods and techniques that were useful across a range of RAND projects and elsewhere. Some important examples are:

  • Problem Solving with Monte Carlo Techniques — Although not invented at RAND, the powerful mathematical technique known as the Monte Carlo method received much of its early development at RAND in the course of research on a variety of Air Force and atomic weapon problems. RAND researchers pioneered the use of the method as a component of digital system simulation. RAND’s main contributions to Monte Carlo lie in the early development of two tools: generating random numbers, and the systematic development of variance-reduction techniques.
  • Cost Analysis— David Novick, as head of RAND’s Cost Analysis Department, developed the fundamental building blocks of cost analysis during the 1950s and 1960s (Novick 1965). Gene Fisher (1971) documented this work in his seminal book Cost Considerations in Systems Analysis.
  • A Million Random Digits with 100,000 Normal Deviates — The tables of random numbers in this report (RAND 1955) became a standard reference in engineering and econometrics textbooks, and were widely used in gaming and simulations that employ Monte Carlo trials. It is one of RAND’s most widely used books.
  • Approximations for Digital Computers — This book by Cecil Hastings, Jr. (1955) contained function approximations for numerical analysts to use in  digital computations of all sorts.
  • Systems Research Laboratory — This laboratory was set up under the leadership of John Kennedy to help examine how groups of human beings and machines work under stress. The work ultimately led to the formation of the System Development Corporation (SDC) in 1955. SDC was spun off from RAND and became a for-profit corporation in 1965.

The Second Ten Years (1958–1967)

This period in RAND’s history witnessed the beginning of the evolution of systems analysis into policy analysis. It also witnessed a branching out from national security research into research on domestic policy issues.

 One of the most important dimensions of change as systems analysis evolved into policy analysis was the context of the problem being analyzed. Contexts became broader and richer over time. What was taken as given (exogenous to the analysis) before became a variable (endogenous to the analysis) later. For example, in the typical systems analysis of the 1950s and early 1960s, many considerations (such as political, sociological, psychological, organizational, and distributional effects) were not taken into account very well; often not at all.  Thus, as systems analysis evolved into policy analysis, the boundaries of the problem space expanded. This had important implications for changes in concepts and methods of analysis. For example, with respect to models, the demands of the expanded boundaries of the problem space could not be met by merely trying to make models used in policy analysis bigger and more complex. Of equal importance, was the development of sophisticated strategies for the development and use of models.

While the evolution of systems analysis into policy analysis did not progress very far during this period, there are several areas of RAND research that were conducted in broader contexts than were typical of the 1950s. These included Ed Barlow’s Strategic Offense Forces Study (SOFS), Bernard Brodie’s (1959) work on the development of a strategy for deterrence in the new age of abundant nuclear weapons and ballistic missiles, Herman Kahn’s (1960) analysis of civil defense in the event of a nuclear war, and Charles Hitch and Roland McKean’s (1960) book Economics of Defense in the Nuclear Age, which espoused the view that the economic use of scarce resources should be a critical aspect of defense planning. This view led to a concentration on cost-benefit analysis,  which was adopted by Secretary of Defense Robert McNamara and led to RAND’s involvement in the development of the defense Planning, Programming, and Budgeting System (PPBS). RAND designed the PPBS to help the Air Force manage the enormous costs of modern military technologies and weapons systems.

 In addition to policy strides like those discussed above, the second ten years witnessed further development of methodological tools for quantitative analysis — primarily operations research tools. Major advances were made at RAND in the areas of mathematical programming, queueing theory, computer simulation, stochastic processes, and operational gaming.

Linear Programming (LP) was probably RAND’s most important and most extensive contribution to the theory and practice of operations research, as well as to economic decision making. Between 1947 and 1952, George Dantzig and others who worked in the Pentagon on the Air Force’s Project SCOOP developed the simplex method and other basic features of LP. Dantzig moved to RAND in 1952. During the following decade, RAND was the world’s center of LP developments. In addition to methodological developments by Dantzig and other RAND employees and consultants (e.g., the development of the dual simplex algorithm), there was seminal work on classic problems like production planning and the traveling salesman problem. In addition, most of the pioneering computer programming of LP algorithms (e.g., the first code for the revised simplex method) was carried out by William Orchard-Hays, Leola Cutler and others at RAND.

Much of the work of this period is captured in Dantzig’s (1963) book, Linear Programming and Extensions. Seminal work in other areas of mathematical programming also took place at RAND during the 1950s and 1960s. Ralph Gomory developed the first integer programming algorithms; Philip Wolfe, George Dantzig, and Harry Markowitz initiated work on quadratic programming; and George Dantzig and Albert Madansky initiated work on stochastic programming. In 1989, Markowitz was awarded the Von Neumann Prize in Operations Research Theory for his work in the areas of the SIMSCRIPT programming language (see below), portfolio theory, and sparse matrix techniques. In 1957 he introduced the pivot selection rule for inverting a matrix in linear programs (Markowitz 1957), which continues to be widely used to solve very large systems of simultaneous equations whose coefficients are mostly zero.  That year, he and Alan Manne  published perhaps the first description of the essence of the branch & bound method for integer programming in their 1957 paper "On the Solution of Discrete Programming Problems."

 Five other examples of RAND work in the “OR tools” area during this period are worthy of note:

  • Simulation — In the early 1960s, after doing complex simulation modeling “the hard way,” Harry Markowitz and Herb Karr developed SIMSCRIPT, a programming language for implementing discrete event simulation models (Markowitz et al. 1963). This work led in l968 to SIMSCRIPT II, which introduced ideas that eventually inspired the modern object-oriented programming paradigm.
  • Artificial intelligence (AI) — The man–machine partnerships explored in the Systems Research Laboratory gained new impetus as Allen Newell, Herb Simon, and Cliff Shaw began to construct a general problem solving language that employed symbolic (non-numerical) processes to simulate human thinking on a computer. One of their initial efforts to carry out a “theory of thinking” involved programming computers to play chess. On a broader scale, this research resulted in several information processing languages (e.g., IPL-V), which were used in some of the early AI computer work.
  • Flows in networks — In 1962, Lester Ford, Jr. and Delbert R. Fulkerson, RAND mathematicians, published the first unified treatment of methods for dealing with a variety of problems that have formulation in terms of single commodity flows in capacity-constrained networks. Their book, Flows in Networks (Ford & Fulkerson 1962), introduced concepts (e.g., “max-flow/min-cut”) and algorithms (e.g., “out-of-kilter”) that have been used to treat network problems ever since.
  • Branching processes — The notion of a branching process concerns individuals from some population that can reproduce and die (become extinct), subject to some probabilistic laws of chance. The theory of branching processes is the mathematical formulation of the development of that population subject to those laws of chance. RAND mathematician Ted Harris was preeminent in this field, and wrote about it in The Theory of Branching Processes (Harris 1964). The theory has been applied to problems associated with such diverse issues as neutron diffusion, cosmic rays, gene attributes, and biological populations.
  • Multi-echelon Inventory theory —the METRIC model was documented by Craig Sherbrooke (1966). This was a pioneering development in dealing with inventory systems having hierarchies of stockage locations. It enabled the Air Force to strengthen the forecasting of demand for spare parts, and thereby achieve higher performance at much lower cost.

The year 1964 saw the publication of the first of several RAND books by mathematician Edward Quade, who played a major role in developing and disseminating the methodology of systems analysis and (later) policy analysis. Analysis for Military Decisions (Quade 1964) documents an intensive five-day course that RAND offered to military officers and civilian decision makers in 1955 and 1959.

During its second decade (and beyond), RAND did some studies related to the war in Vietnam.   Although more than 500 RAND reports relating to the war were published (Elliott 2010), relatively little of RAND’s work relating to the war appears to have involved operations research.  Notable exceptions are:

  • The work by RAND staff assigned to the joint US/Vietnamese Combat Development Test Command  on border security (Sturdevant 1964) and other operational issues (Elliott 2010).
  • A statistical analysis by Anthony Russo (1967) of the impact of the U.S. Crop Spraying program.
  • An implementation of the Ford/Fulkerson algorithm (Wollmer & Ondrasek 1969) that was used to formulate interdiction strategy (Ondraseck 2011)).
  • Models of the influence of close air support and other contributions upon the outcome of ground engagements of regimental size or smaller (Lind et al. 1971). 

Elliott (2010) notes that “RAND was and still is similar to a university—a community of thinkers with disparate views, ranging from the supportive to the critical of U.S. government policies in Vietnam. There were people who worked to make weapons or technology more lethal. But there were also those who took issue with the destructive use of American power in Vietnam, as well as those who advocated U.S. withdrawal from the war.“

The Third Ten Years (1968–1977)

This period in RAND’s history saw an acceleration of many of the trends begun in the previous ten years. One of these trends involved the development of improved procedures for the use of expert judgment as an aid to military decision making. The Delphi procedures grew out of this effort (Dalkey, et al., 1972). These procedures incorporate anonymous response, iteration and controlled feedback, and statistical group response to elicit and refine group judgments where exact knowledge is unavailable. Other trends involved the continued evolution of systems analysis into policy analysis, and an increasing emphasis on analyzing major domestic research issues.

On July 14, 1966,  RAND’s Board of Trustees decided to terminate its exclusive focus  on military research and diversify into social welfare research.   Systems analysis became the basis for policy analysis across such disparate areas as housing, poverty, health care, education, and the efficient  operation of municipal services, such as police protection and firefighting.  Ed Quade laid out the fundamentals underlying policy analysis in his first non-defense book, Analysis for Public Decisions (Quade 1975).  Among the important developments within RAND was the establishment of the New York City-RAND Institute (NYCRI), and policy analysis studies for the government of the Netherlands.

 The New York City-RAND Institute — In 1968, RAND began a long-term relationship with the City of New York to tackle problems in welfare, health services, housing, fire protection, law enforcement, and water resources. The NYCRI was formally established in 1969. The research staff evaluated job training programs, suggested solutions to shortages of nurses in municipal hospitals, helped change rent control, altered fire department deployment policies, reallocated police manpower, and helped improve Jamaica Bay’s water quality.

The most successful of the NYCRI’s projects was the one devoted to improving the operations and deployment of the Fire Department of New York. In 1968, the major problem facing the Department was the rising alarm rate. Its increasing workload was not significantly relieved by adding more men and equipment; nor were traditional methods of fire company allocation, dispatching, and relocation working. The Institute’s studies altered the way the Department managed and deployed its men and equipment and operated its dispatching system. An integral part of the research involved creation of a wide variety of computer models to analyze and evaluate deployment, which led to the formulation of new policies. Warren Walker and Peter Kolesar were awarded ORSA’s 1974 Lanchester Prize for a paper that described how mathematical programming methods were applied to the problem of relocating available fire companies to firehouses vacated temporarily by companies fighting fires. The entire body of work from this project is documented in Fire Department Deployment Analysis (Walker et al. 1979).

Policy Analysis Studies for the Dutch Government — Reflecting an increasing interest in doing policy analysis studies in international contexts, RAND started working for the Dutch government in the 1970s.

One important study was concerned with protecting an estuary from floods. In April 1975, RAND began a joint research venture with the Dutch government to compare the consequences of three alternative approaches for protecting the Oosterschelde, the largest Dutch estuary, from flooding. Seven categories of consequences were considered for each alternative: financial costs, ecology, fishing, shipping, recreation, national economy, and regional effects. Within each category, several types of consequences were considered. In June 1976, the Dutch Parliament adopted one of the alternatives based in large part on the results of the RAND study: to build a 10 km, multi-billion dollar storm surge barrier with large movable gates across the mouth of the estuary. The study required the development of sophisticated computer models of estuaries and coastal seas.

A second study was focused on improving water management in the Netherlands. Begun in April 1977, the Policy Analysis for the Water Management of the Netherlands (PAWN) project was conducted jointly by RAND, the Dutch Government, and the Delft Hydraulics Laboratory. It analyzed the entire Dutch water management system, and provided a basis for a new national water management policy for the country. It developed a methodology for assessing the multiple consequences of possible policies (e.g., their effects on agriculture, hydrology, irrigation, shipping, industry, drinking water  companies, power plants, salt intrusion, groundwater supplies, and the environment), and applied the methodology to generate  alternative policies and to assess and compare their consequences (Goeller et al. 1983). An integrated system of 50 models was used to evaluate policies that included mixes of building new facilities and changing operating rules to improve water supply, as well as adjusting prices and regulations to reduce demands. The project won a Franz Edelman Award for Management Science Achievement in 1984.

In 1970, RAND established one of the original eight public policy graduate schools in the United States, the Pardee RAND Graduate School (PRGS). PRGS is the world’s largest doctoral program in the field. PRGS doctoral fellows take advanced courses in such fields as economics, statistics, political science, and the social sciences. They also work part-time as members of RAND’s interdisciplinary research teams, which is how they earn their fellowships. Fellows obtain research training in RAND’s classrooms, and get to apply it to real problems with RAND mentors and clients.

The Fourth Ten Years (1978–1988)

During this period, RAND’s research program increased substantially in size and diversity. Many of the trends of the past continued — for example, the increase in efforts devoted to domestic policy research and the tendency to conduct research in broader contexts. Several new trends began to emerge — for example, an increase in emphasis on research done in international contexts other than the (then) USSR. The development of analytical concepts, methods, and techniques also continued. Some of the more important of these were:

  • RAND Strategy Assessment System (RSAS) — Because of perceived limitations in methods of strategic analysis, in 1982 RAND began to develop methods for strategic analysis that combined classical gaming, systems analysis methods and techniques, artificial intelligence, and advanced computer technology. The RSAS provided a structure and tools for analyzing strategic decisions at the national command level as well as decisions at the operational level. It also provided great flexibility in choosing which roles are to be played by people and which by machines.
  • Dyna-METRIC — The Dyna-METRIC logistics support model provided a major new tool for relating the availability of spare parts to wartime aircraft sortie generation capability. The model (Hillestad 1982) combines elements of queueing theory, inventory theory, and simulation. It is still an integral part of the Air Force logistics and readiness management system.
  • CLOUT (Coupling Logistics to Operations to Meet Uncertainties and the Threat) is a RAND-developed set of initiatives for improving the ability of the Air Force logistics system to cope with the uncertainties and disruptions of a conventional war overseas. The CLOUT initiatives are intended to offset the substantial variability expected in the demand for spare parts, maintenance, and other support activities, as well as the consequences of damage to theater air bases. CLOUT was the basis for the DRIVE model (Distribution and Repair in Variable Environments), developed by John Abell et al. (1992), that became the kernel of a management system, called EXPRESS, that is still used today in Air Force depots.
  •  The Enlisted Force Management System (EFMS)—The EFMS project is notable for the scope and complexity of the decision support system that it developed, and for demonstrating how the tools of operations research could be married with emerging information technologies to provide real-time decision support throughout an organization. Warren Walker led a large RAND team that worked with Air Force counterparts beginning in the early 1980s. Together, they produced an organizational decision support system (ODSS) to help make decisions about the grade structure of the enlisted force, enlisted promotion policies, and the recruitment, assignment, training, compensation, separation, and retirement of Air Force enlisted personnel. Since 1990, the EFMS has been the primary analytical tool used to support major policy decisions affecting the enlisted force. The success of the system motivated the publication of the book Building Organizational Decision Support Systems (Carter, et al. 1992).

Recent Developments

After the first 40 years, most of the main trends outlined above continued to play themselves out — for example, domestic research represents about half of RAND’s $350 million annual research budget, and methodological enhancements driven by the practical needs of the problem-oriented research continue to have high priority. Methodologically, some of the most important advances have involved new approaches for dealing with uncertainty in making decisions. Chief among these approaches are Assumption-Based Planning (Dewar et al. 1993), Exploratory Modeling (Bankes 1993), and adaptive policies (Walker et al. 2001). All three of these approaches are combined in a methodology for long-term policy analysis called Robust Decision Making  (Lempert et al. 2003), and are the building blocks for new  methods and tools that have led to the establishment of a new  international society (the Society for Decision Making Under Deep Uncertainty), and the publication of the book Decision Making  Under Deep Uncertainty: From Theory to Practice (Marchau, et al. (eds.) 2019).

 Over the past two decades, RAND has been becoming a more global institution. The RAND research staff now includes citizens of more than 50 nations, and RAND now performs research for many countries besides the United States. In 1992, RAND established an affiliate called RAND Europe. It conducts policy studies to inform public- and private-sector decision making  throughout Europe. Major research efforts have included studies of the safety of Schiphol Airport, ways of improving river dikes in the Netherlands while preserving the environment, a systematic examination of alternative strategies for reducing the negative effects of road freight transport in the Netherlands, and a cost-effectiveness analysis of strategies for improving shipping safety in the North Sea. Other efforts include pioneering work on widening the application of discrete choice analysis, guiding appraisals of transport infrastructures, and providing cost analysis for the UK Ministry of Defence. RAND Europe now operates from its headquarters in Cambridge, England and a representation office in Brussels. With a diverse range of research areas ranging from defense to healthcare, it has become a key provider of evidence-based policy research across the European Union.

More recently, RAND established  RAND Australia in Canberra.  For a short period, there was a RAND-Qatar Policy Institute (RQPI), which served clients throughout the Middle East, North Africa, and South Asia, performing studies on such issues as education reform, health services delivery, governmental reorganization, and environmental health.  

RAND analysts continue to be recognized for their contributions to the profession of operations research and the management sciences. Warren Walker received the INFORMS President’s Award in 1997 for his “contributions to the welfare of society through quantitative analysis of governmental policy problems.”   In 1996, Moore et al. were awarded the Richard H. Barchi Prize for the best paper given at the Military Operations Research Society (MORS) Symposium. Brooks et al. won that prize in 1998. 

Compiled by Warren Walker; Edits by Mark Eisner

Oral Histories

RAND History Project Interviews, 1985 -1990.  National Air and Space Museum, Archives Division, MRC 322, Washington, DC, 20560 (link

Archives

RAND Corporation Archives Available for Scholarly Research (link)  

Links and References

Abell J., Miller, L. W., Neumann, C. E., &  Payne, J.E. (1992). DRIVE (Distribution and Repair in Variable Environments):  Enhancing the responsiveness of depot repair, R-3888-AF, Santa Monica, CA: RAND. (link)

Augenstein B. (1993) A Brief History of RAND's Mathematics Department and Some of Its Accomplishments. DRU-218-RC , Santa Monica CA: RAND (link

Bankes S. C. (1993). “Exploratory Modeling for Policy Analysis”. Operations Research, Vol. 4, No. 3, pp. 435–449.

Bellman E. R. (1957). Dynamic Programming. Princeton, NJ: Princeton University Press.

Bellman E. R., & Dreyfus, S. E. (1962). Applied Dynamic Programming. Princeton, NJ: Princeton University Press.

Brodie B. (1959). Strategy in the Missile Age. Princeton, NJ: Princeton University Press.

Brooks A., Bennett, B., & Bankes, S. (1998). “An Application of Exploratory Analysis: The weapon mix problem”. MORS Journal, Vol. 4, No. 1, pp. 67–80.

Carter G. M., Murray, M. P., Walker, R. G., & Walker, W. E. (1992). Building Organizational Decision Support Systems. San Diego, CA: Academic Press.

Collins M. J. (2002) Cold War Laboratory : RAND, the Air Force, and the American State, 1945-1950 Washington DC: Smithsonian Institution Press. 

Dalkey N. C., Rourke, D. L., Lewis, R. J., & Snyder, D. (1972). Studies in the Quality of Life: Delphi and decision-making. Lexington, MA: D.C. Heath.

Dantzig G. B. (1963). Linear Programming and Extensions. Princeton, NJ: Princeton University Press.  Available from RAND as R-366-PR (link)

Dewar J. A., Builder, C. H., Hix, W. M., & Levin, M. H. (1993). Assumption-Based Planning: A planning tool for very uncertain times, MR-114-A. Santa Monica, CA: RAND. (link)

Digby J. (1989) Operations Research and Systems Analysis at RAND, 1948-1967, N- 2936-RC.  Santa Monica CA: RAND (link)

Elliott M. (2010) RAND in Southeast Asia: A History of the Vietnam War Era. Document CP-564-RC Santa Monica, CA: RAND. (link)

Fisher G. H. (1971). Cost Considerations in Systems Analysis. New York: American Elsevier Publishing Co.  Available from RAND as R-490-ASD (link)

Fisher G. H., W. E. Walker, & M. Rich (2013), RAND Corporation, in Encyclopedia of Operations Research and Management Science, 3rd edition,  New York NY: Springer.

Ford L. R., Jr., & Fulkerson, D. R. (1962). Flows in Networks. Princeton, NJ: Princeton University Press.   Typed manuscript available as RAND report R-375-PR.  (link)

Goeller B. F., et al. (1983). Policy Analysis of Water Management for the Netherlands (Vol. 1), Summary Report, R-2500/1–NETH, Santa Monica CA: RAND (link)

Gruenberger F. J. (1968) The History of the Johnniac.  RM-5654-PR. Santa Monica CA: RAND.  (link)

Harris T. E. (1964). The Theory of Branching Processes. Englewood Cliffs, NJ: Prentice-Hall.

Hastings C. Jr. (1955). Approximations for Digital Computers.  Princeton, NJ: Princeton University Press.

Hillestad R.J. (1982). Dynamic Multi-Echelon Technique for Recoverable Item Control. R-2785-AF. Santa Monica, CA: RAND. (link)

Hitch C. J., & McKean, R. (1960). The Economics of Defense in the Nuclear Age. Cambridge, MA: Harvard University Press.

 Jardini D. R. (1996). Out of the Blue Yonder: The RAND Corporation’s  diversification into social welfare research, 1946–1968.  Doctoral dissertation, College of Humanities and Social Sciences,  Carnegie Mellon University,Pittsburgh, PA.

Jardini D. R. (2013) Thinking Through the Cold War: RAND, National Security, and Domestic Policy, 1945-1975.

Kahn H. (1960). On Thermonuclear War. Princeton, NJ: Princeton University Press.

Lempert R. J., Popper, S. W., & Bankes, S. C. (2003). Shaping the Next One Hundred Years: New methods for quantitative, long-term policy analysis, MR-1626-RPC. Santa Monica, CA: RAND. (link)

Lind J. R., Harris K. & Spring S.G.  (1971) FAST –VAL : A Study of Close Air Support.  R-811-PR, November 1971, Santa Monica CA:  RAND.  (link)

Marchau V. A. W. J., Walker, W. E.,  Bloemen, P. J. T. M., & Popper, S. W.  (Eds.). (2019). Decision Making Under Deep Uncertainty: From Theory to Practice. New York: Springer.

Markowitz, H. M. (1957). “The Elimination Form of the Inverse and its Application to Linear Programming”, Management Science, Vol. 3, No. 3, pp. 255-269.

Markowitz H. M. and A. S. Manne(1957) "On the Solution of Discrete Programming Problems", Econometrica, vol. 25. no. 2, pp. 84-110.

Markowitz H. M., Hausner, B., & Karr, H. W. (1963). SIMSCRIPT: A simulation programming language. Englewood Cliffs, NJ: Prentice-Hall.

Miller L. W., G. Fisher, W. Walker & C. Wolf (1988), "Operations Research And Policy Analysis At RAND, 1968-1988," OR/MS Today v.15, Dec 1988, reprinted as  RAND Note N-2937, April 1989. (link)

Moore S. C., Kakalik, J. S., Benjamin, D. R., & Stanton, R. E.(1996).  Choosing Force Structures: Modeling interactions among wartime requirements,  peacetime basing options, and manpower and personnel policies, MR-550-AF.  Santa Monica, CA: RAND.   Summary and mathematical appendix reprinted as RAND document P-7973.  (link)

Newell A. (Ed.). (1961). Information Processing Language V - Manual. Englewood Cliffs, NJ: Prentice-Hall.

Novick D. (Ed.). (1965). Program Budgeting. Cambridge, MA: Harvard University Press.

Ondrasek M. J. (2011) email exchange with M. J. Eisner 10/21/2011

Orchard-Hayes W (1990) History of the Development of LP Solvers. Interfaces 20:4, July-August 1990 pp. 61-73 (link)

Paxson E. (1950). Strategic Bombing Systems Analysis. Santa Monica, CA: RAND.

Quade E. S. (Ed.). (1964). Analysis for Military Decisions. Chicago: Rand McNally.

Quade E. S. (1975). Analysis for Public Decisions. New York:  Elsevier Science Publishing Co., Inc.

RAND (1955). A Million Random Digits with 100,000 Normal Deviates, MR-1418-RC. Santa Monica, CA: RAND (Reissued in 2001 with a new Foreword by M. Rich. link).

RAND A Brief History of RAND. Web page accessed May 22, 2019. (link)

Redmond K. C. & Smith T. M. (1967)  Project Whirlwind: A Case History in Contemporary Technology.  (Reproduced manuscript, 1975) Cambridge MA: MITRE Corporation (link)

Russo A. J. (1967) A Statistical Analysis of the U.S. Crop Spraying Program in South Vietnam.  RM-5450-1-ISA/ARPA, October 1967, Saigon Vietnam:RAND  (link)

Shaw J. C. (1964). JOSS: A designer’s view of an experimental on-line computing system, P-2922. Santa Monica, CA:RAND.  (link)

Sherbrooke C. C. (1966). METRIC: A multi-echelon technique for  recoverable item control, RM-5078-PR. Santa Monica, CA: RAND. (link)

Specht R. D. (1960) "Rand - A Personal View of Its History," Operations Research vol 8 no 6 (Nov-Dec 1960) pp. 825-83.   Reprinted as RAND note P-1601, October 1968. (link

Sturdevant C. V. (1964) The Border Security Problem in South Vietnam.  RM-3967-ARPA, Saigon Vietnam: RAND (link)  

Walker W. E., Chaiken J. M., & Ignall E. J. (Eds.). (1979). Fire Department Deployment Analysis: A public policy analysis case study. New York: Elsevier North Holland.

Walker W. E., Rahman S. A., & Cave J. (2001). “Adaptive Policies, Policy Analysis, and Policymaking”, European Journal of Operational Research,  Vol. 128, No. 2, pp. 282-289.

Ware, W. H.  (2008)  RAND and the Information Evolution.  Document CP-537-RC. Santa Monica CA: RAND.  (link)

Williams J. D. (1954). The Compleat Strategyst: Being a primer on the theory of games of strategy. New York: McGraw-Hill.  Reissued in 2007 as RAND Document CB-113-1 (link)

Wohlstetter A. J., Hoffman F. S. Lutz, R. J., & Rowen H. S. (1954). Selection and Use of Strategic Air Bases, R-266. Santa Monica, CA: RAND.   (link)

Wollmer R. D. & Ondrasek M. J. (1969) A Model for Targeting Strikes in an LOC Network.  RM-5940-PR Santa Monica CA:RAND  (link)

Associated Historic Individuals

Arrow, Kenneth J.
Aumann, Robert J.
Balinski, Michel
Bellman, Richard E.
Blackwell, David
Conway, Richard W.
Dantzig, George B.
Dyer, James S.
Fishman, George S.
Flood, Merrill M.
Fulkerson, D. Ray
Gale, David
Gaver, Donald P.
Geisler, Murray
Geoffrion, Arthur M.
Hitch, Charles J.
Hu, Te Chiang
Hurwicz, Leonid
Isaacs, Rufus
Karlin, Samuel
Manne, Alan S.
Markowitz, Harry
Maxwell, William
Mood, Alexander
Morse, Philip M.
Nash, Jr., John F.
Porteus, Evan L.
Rice, Donald B.
Rothblum , Uriel G.
Samuelson, Paul A.
Scarf, Herbert E.
Shapley, Lloyd S.
Shubik, Martin
Simon, Herbert A.
Tien, James M.
von Neumann, John
Wagner, Harvey M.
Walker, Warren E.
Walsh, John E.
Wolfe, Philip