SYSTEMS ENGINEERING

technique of using knowledge from various branches of engineering and science to introduce technological innovations into the planning and development stages of a system. Systems engineering grew out of the overlapping of various engineering disciplines developed in the 19th and 20th centuries. Its first application as a specific discipline was in the organization of commercial telephone systems in the 1920s and '30s. Many systems-engineering analytic techniques were developed during World War II as methods of increasing the efficiency of military equipment deployment. Postwar growth in the field was spurred by advances in electronic control systems and by the development of computers and information theory. The systems engineer is usually called upon to incorporate new technology into a system that is: (1) man-made; (2) large and complex (where a change in one part, or subsystem, may affect many others); and (3) stochastic (subject to random, unscheduled changes). After identifying the objectives of the current system, the systems engineer adjusts the new technology to maintain that objective. A systems engineer draws upon his expertise in science, technology, and management to solve complex problems of cost and scheduling. One of the tools that an engineer uses to manage complex systems is a flowchart. A graphic depiction of a system, in which geometric figures represent various subsystems and arrows represent their interactions, the flowchart helps point out the interrelationships of processes within the system. A more quantitative technique uses mathematical models that reduce system interactions to mathematical equations. Probability theory, statistical analysis, and computer simulations are also tools of the systems engineer. The classic application of systems engineering is in commercial telephony. Systems-engineering techniques are used to solve problems in such diverse fields as military and space systems and transportation and pollution control. technique of using knowledge from various branches of engineering and science to introduce technological innovations into the planning and development stages of a system. Systems engineering is not so much a branch of engineering as it is a technique for applying knowledge from other branches of engineering and disciplines of science in effective combination to solve a multifaceted engineering problem. It is related to operations research but differs from it in that it is more a planning and design function, frequently involving technical innovation. Probably the most important aspect of systems engineering is its application to the development of new technological possibilities with the specific objective of putting them to use as rapidly as economic and technical considerations permit. In this sense it may be seen as the midwife of technological development. The word systems is frequently used also in other combinations, especially when elements of technological advance are not so important. Systems analysis is an example. Systems theory, or sometimes systems science, is frequently applied to the analysis of physical dynamic systems. An example would be a complex electrical network with one or more feedback loops, in which the effects of a process return to cause changes in the source of the process. In the development of the various engineering disciplines in the 19th and 20th centuries, considerable overlap was inevitable among the different fields; for example, chemical engineering and mechanical engineering were both concerned with heat transfer and fluid flow. Further proliferation of specializations, as in the many branches of electrical and electronic engineering, such as communications theory, cybernetics, and computer theory, led to further overlapping. Systems engineering may be seen as a logical last step in the process. Systems engineers frequently have an electronics or communications background and make extensive use of computers and communications technology. Yet systems engineering is not to be confused with these other fields. Fundamentally a point of view or a method of attack, it should not be identified with any particular substantive area. In its nature and in the nature of the problems it attacks, it is interdisciplinary, a procedure for putting separate techniques and bodies of knowledge together to achieve a prescribed goal in an effective manner. In general, a systems engineering approach is likely to differ from a conventional design approach by exhibiting increased generality in its basic logical framework and increased concern with the fundamental objectives to be achieved. Thus, at each stage the systems engineer is likely to ask both why and how, rather than merely how. In addition to systems engineering, it is important to define systems themselves. The systems with which a systems engineer is concerned are first of all man-made. Second, they are large and complex; their component parts interact so extensively that a change in one part is likely to affect many others. Unless there is such interaction, there is little for the systems engineer to do, at least at the systems level; he can turn immediately to the components themselves. Another important characteristic of systems is that their inputs are normally stochastic; that is, the inputs are essentially random functions of time, although they may exhibit statistical regularities. Thus, one cannot expect to foresee exactly what the system will be exposed to in actual operation, and its performance must be evaluated as a statistical average of the responses to a range of possible inputs. A calculation based on a single precisely defined input function will not do. Systems may also vary depending on the amount of human judgment that enters into their operation. There are, of course, systems such as electrical circuits, automated production equipment, or robots that may operate in a completely determinate fashion. At the other extreme, there are management and control systems, for both business and military purposes, in which machines in a sense do most of the work but with human supervision and decision making at critical points. Clearly these mixed human-machine systems offer the greatest variety both of possibilities and problems for the systems engineer. Aspects of such systems are treated in the article human-factors engineering. Additional reading Two classics on systems engineering are Philip M. Morse and George E. Kimball, Methods of Operations Research, rev. ed. (1951, reissued 1970), the introductory discussion still useful for orientation but the specific examples mostly from World War II; and Operations Research for Management, vol. 1 ed. by Joseph F. McCloskey and Florence N. Trefethen (1954), and vol. 2 ed. by Joseph F. McCloskey and J.M. Coppinger (1956), with a broad range of examples, many of which are still of interest. Later books, with more stress on mathematical techniques, include Charles West Churchman, Russell L. Ackoff, and E. Leonard Arnoff, Introduction to Operations Research (1957, reprinted 1961), a fair survey of the field but more useful for its description of operations research methodology; and Frederick S. Hillier and Gerald J. Lieberman, Introduction to Operations Research, 3rd ed. (1980), with more emphasis on methodology (and probabilistic techniques in particular) and less suited to the general reader. Kenichi Ohmae, The Mind of the Strategist: The Art of Japanese Business (1982), is a good survey of strategy techniques.The earliest significant books directly related to systems engineering are Harry H. Goode and Robert E. Machol, System Engineering: An Introduction to the Design of Large-Scale Systems (1957), and Robert E. Machol, Wilson P. Tanner, and Samuel N. Alexander (eds.), System Engineering Handbook (1965). These cover the philosophy and methodology of systems engineering; however, both books are directed primarily at large military systems and are less useful for other applications. Another book that covers both philosophical and technical aspects of systems engineering is Arthur D. Hall, A Methodology for Systems Engineering (1962, reprinted 1968). Harold Chestnut, Systems Engineering Tools (1965), and Systems Engineering Methods (1967), deal largely with methodology. James Botkin et al., Global Stakes: The Future of High Technology in America (1982, reprinted 1984), explores specific methods of development. More general references include Steven L. Dickerson and Joseph E. Robertshaw, Planning and Design, The Systems Approach (1975); and Dean Karnopp and Ronald Rosenberg, System Dynamics: A Unified Approach (1975).Among shorter works, an article by Hendrick W. Bode, The Systems Approach, in National Research Council, Panel on Applied Science and Technological Progress, Applied Science and Technological Progress: A Report to the Committee on Science and Astronautics, U.S. House of Representatives, pp. 7394 (1967), furnishes a broad discussion of the field. G.M. Jenkins, The Systems Approach, J. Systems Engng., 1:349 (1969), offers a valuable elementary introduction to systems engineering. Charles Hitch, Sub-optimization in Operations Problems, Journal of the Operations Research Society of America, 1(3):8799 (May 1953), is a classic.