Imagini ale paginilor


TUESDAY, MAY 17, 1966



Washington, D.C. The subcommittee met at 9:30 a.m., pursuant to call, in room 4232, New Senate Office Building, Senator Gaylord Nelson, chairman of the special subcommittee, presiding:

Present: Senators Nelson (presiding), Pell, Javits, and Murphy, Senator NELSON. We will open hearings on S. 2662, the Scientific Manpower Utilization Act.

We have two witnesses this morning. This is part of a series of hearings that we have been conducting on this bill. The first were conducted in California.

We have heard from witnesses from the aerospace industry. The two witnesses this morning are Alain C. Enthoven, Assistant Secretary for Systems Analysis, Department of Defense; and Mr. Henry S. Rowen, Assistant Director of the Bureau of the Budget.

This bill is designed to bring to bear on social and economic problems at the State, national and regional level new techniques developed over the last few years in defense and space efforts, and in budget ef. forts. These techniques, usually described as systems approaches or systems analysis, have played a very successful role both in aiding the engineering breakthrough into space and in military hardware development, and also by providing the management tools for informed decisionmaking.

The management application of systems analysis that has been so useful in the Defense Department is known as the planning, programing, and budgeting system (PPBS).

On October 12, 1965, the Budget Bureau issued its well-known bulletin 66–3, in which it informed the other agencies of the Federal Government that the President has directed the introduction of PPBS in the executive branch.

At today's hearings, we are pleased to have with us two gentlemen who have been instrumental in the development of PPBS. The problems faced by this Nation at the State, regional, and national level in the field of education, welfare, conservation, transportation, pollution, urban planning, and land use control, to name only a few of the staggering list, are certainly no less urgent or complex than the problems faced by the Defense Department.

It is therefore crucial, it seems to me, that we find ways to use the very best in advanced techniques to bear on this problem at the State, national, and regional level. It seems to me that the system holds great promise for meeting the social and economic problems of the State, regional, and national level.

These hearings have two purposes: first, to explore systems analysis and systems approaches to find the best way to make them available; and, second, to develop a hearing record that will make clear to local officials and national officials, with administrative rather than technical responsibilities, what the systems approaches are all about.

I will call first on Mr. Enthoven. I would hope that both Mr. Enthoven and Mr. Rowen will feel free any time to raise questions about the other's testimony or to add to it during the course of the testimony.

Mr. Enthoven.



Mr. ENTHOVEN. Thank you, Mr. Chairman.

Mr. Chairman, it is a great pleasure for me to appear before you this morning and to have the oportunity to make a contribution to the use of systems analysis on problems of State and local governments. I believe that this is a most worthwhile objective and that there are clear possibilities for making major contributions to the public welfare through the broader use of systems analysis at all levels of government.

What I have to say will be based on our experience in the Department of Defense. But, I want to emphasize at the outset my conviction that the problems of State and local government and the problems education, natural resources management, pollution of the environment, and public health and welfare are no more complex and no less amenable to systematic, rational analysis than are the problems of defense. I need only mention our current problems in NATO and in defeating aggression in southeast Asia to islustrate the point that we have our share of complex problems. While I would not want to suggest that system analysis has solved these problems, I think that it is fair to say that a systematic and integrated approach to the gathering and presentation of information on the alternatives available to our Government has made the work of our responsible decisionmakers easier and more productive than it might otherwise be.

I understand that you are interested in identifying and making available useful literature on the systems analysis approach. I am very happy to be able to call to your attention and to submit for the record a prepublication copy of “A Modern Design for Defense Deci. sion,” an anthology on Defense management, programing and systems analysis, including speeches and articles on these subjects by Secretary McNamara, former Assistant Secretary of Defense (Comptroller) Charles Hitch, and myself. This volume is being published by the Industrial College of the Armed Forces and the hard-bound printed version will be available late in June.

Senator NELSON. You are submitting a copy of this for use by the subcommittee?

Senator NELSON. That will be put in the subcommittee files.

Mr. ENTHOVEN. There is a great deal that might be said about the systems analysis approach. In this statement, I would like to pick out à few of the aspects that seem to me to be especially relevant and to make these points largely by the use of excerpts from “A Modern Design for Defense Decision."

In my statement, I would like to expand on the following points:

(1) Systems analysis is a reasoned approach to problems of decision, accurately described as “quantitative commonsense."

(2) Systems analysis is an application of scientific method, using that term in its broadest sense.

(3) There are limitations in the application of systems analysis, although these have often been overstated.

(4) In 1961, the Defense planning and budgeting system had to be changed to permit the application of systems analysis.

(5) Systems analysis is a regular working contributor to the annual Defense decisionmaking cycle.

(6) Two necessary conditions for the successful application of systems analysis as a working part of an operating organization are that it be used by decisionmakers, and that it be fed with ideas by a broadly based interdiscipilinary research program.

(7) Systems analysis can be applied to the problems of State and local government, including programs for social welfare.

Senator NELSON. Will you elaborate a little bit on that? In what way do you use the interdisciplinary approach in the development of your systems analysis in the Defense Department? What disciplines are involved in the Defense Department development of a program?

Mr. ENTHOVEN. The main disciplines that we draw on in our weapons systems and strategy studies are the military profession; that is, officers experienced in military operations; physics, engineering, and other physical sciences, economics, political science, and other social sciences. The men from these professions work together in collaboration, oriented toward the solution of the problems at hand. The main focus has to be the problem itself, and not the prerogatives or the use of this or that particular discipline.

It is very important to make sure that the physics used in our weapons system analysis is good physics. So you want to have good physicists. But once the physicists are working on the problem, they have to be willing to transcend the traditional boundaries of their academic background.

Senator Nelson. In deciding on a particular weapon, for example, where it might be used or where it might be based, does the Defense Department use the counsel of political and foreign relations experts for some advice on what the effect might be? In other words, the development of a weapons system may require the use of a certain weapon, an airplane or something else, based on foreign soil; and as long as it is there, it may be most effective and the best weapon we have. But if something developed politically that required the withdrawal of the weapon and you could not use it at all, what effect would that have on the weapons system you have developed?

Are political experts used to raise these questions so that the decision maker or makers in the Defense Department can at least evaluate that problem in making a determination as to the weapons system?

Mr. ENTHOVEN. Yes, sir. We do that, both in our research program and in our staffing of specific decisions that have to be made by the Secretary of Defense and with the help of the State Department. We go through a process very much along the lines of what you described. That is, a group working on the choice of weapons systems might well include a political scientist who would observe that the different basing concepts associated with different weapons would have substantial implications for our relations with foreign countries. He would raise and identify the issues and clarify what the implications are for the decisionmakers.

Senator Nelson. How is the system set up? You do have to have an analysis of the overall weapons system. It has a thousand parts. What are the mechanics for setting up the analysis of the weapons system?

Mr. ENTHOVEN. Let me take an example from the research program, one in which Mr. Rowen was a leading participant back in the early 1950's, when the Rand Corp. embarked on a major study of the basing and operation of the Strategic Air Command.

One of the main problems that study identified was the fact that our basing of strategic bombers on overseas bases close to the U.S.S.R. and outside our warning net left these bombers vulnerable to surprise nuclear attack. The study examined a number of alternatives for the solution of this problem, alternatives for guaranteeing the security of our strategic retaliatory power. The alternatives were described and evaluated primarily from the point of view of our strategic retaliatory capability, but also from the point of view of what implications and requirements they would have for overseas bases and therefore for our dependence on other countries.

Actually, this particular study, which is an example of the best of this kind of research, was done mainly by a group of four men within a larger organization from which they could draw for assistance on particular pieces of the work by aeronautical engineers, physicists, political scientists, and the like.

Senator NELSON. All part of the Rand Corp.? Mr. ENTHOVEN. All part of the Rand Corp.; yes, sir. Senator Nelson. Did they also use whatever they needed from the whole Military Establishment?

Mr. ENTHOVEN. Yes. In fact, we had very good assistance and cooperation from the Air Force; in this case, the Strategic Air Command. Air Force officers were present at the Rand Corp., participating in and contributing to the studies, and providing very necessary and most valuable advice on the practicability of the operations and the realistic aspects of the operational problems.

Senator NELSON. Senator Murphy, we have testifying presently Mr. Enthoven, of the Defense Department, and Mr. Rowen, of the Budget Bureau, with him.

Senator MURPHY. Thank you.
Senator Nelson. You may proceed.


Mr. ENTHOVEN. Systems analysis is nothing more than quantitative or enlightened commonsense aided by modern analytical methods.

What we seek to do in the systems analysis approach to problems is to examine an objective in its broadest sense, including its reasonableness or appropriateness from a national policy point of view, and then develop for the responsible decisionmaker information that will best help him to select the preferred way of achieving it. This process of selection requires that we first identify alternative ways of achieving the objective and then estimate, in quantitative terms, the benefits (effectiveness) to be derived from, and the cost of, each alternative. These aspects of the problem that cannot easily be quantified are explicitly stated. Of course, this would include the foreign policy aspects. In principle, we strive to identify the alternative that yields a specified degree of effectiveness for the least cost or, what is the same thing, the greatest effectiveness for a given cost. In essence, it is a way of dealing with the basic economic problem-how best to use our limited national resources. So much for what systems analysis is. A few words on what it is not.

Systems analysis is not synonymous with the application of computers. There is no essential connection between the two. Certainly the development of the former in no way depends on computers. Some researchers, working within the limits of the systems analysis approach, try to do their analyses by means of large-scale computer simulations. Actually, the computer simulation approach so far has not been particularly fruitful as a method of weapon systems analysis. However, the potential advantages offered by high-speed electronic computers are very great. One of the primary advantages of the computer to the systems analysis function is to permit us to examine a much larger number of alternatives in a shorter period of time than would be otherwise possible. This is especially important in the case of very complex and interrelated systems where hand calculations would limit the time available for the more important work of analysis. I intend to try to exploit more fully the potential of high-speed computers. But I would like to make it clear that I view the computer as a mechanical aid in my work and not as the substance of my work.

Moreover, systems analysis is not mysterious or occult. It is not performed with the help of a mysterious black box. A good system analyst should be able to give a clear nontechnical explanation of his methods and results to the responsible decisionmakers.


I would like now to turn to what I believe are some of the basic characteristics of the systems analysis method. Systems analysis is at once electric and unique. It is not physics, engineering, mathematics

, economics, political science or military operations, and yet it involves elements of all of the above disciplines. But regardless of its makeup, the art of systems analysis

and it is an art-like the art of medicine, must be based on the scientific method, using this term in its broadest sense.

What are the relevant characteristics of scientific method as applied to the problem of choosing strategies and selecting weapon systems, or, for that matter, to the analysis of any problem of public policy involving allocation of the Nation's scarce resources? I would like

« ÎnapoiContinuă »