Extracts from: Hitchins, D.K.,
Advanced Systems, Thinking and Management,
Artech House, Norwood, MA, 2003
- The file SystemsPrinciples.mov identified by urn:sha1:ZOHWPVLRUJ5GSEH7MYHMTFHZZ3HAURH2 is licensed to the public under the Attribution 3.0 Unported license.
Holism and Emergence
Systems engineering may have existed, at least as a philosophy, for thousands of years. The Fourth Dynasty of the ancient Egyptian culture created enormous pyramids in relatively short times some 4,600 year ago without any significant tools other than ropes. The Great Pyramid of Khufu, the last remaining wonder of the ancient world, is still the largest man-made stone building on earth, and it was constructed in about twenty years.
The ancient Egyptians drew up careful plans, and built the many parts of the pyramid complex in parallel, a practice that today's systems engineers would call concurrent or simultaneous engineering, intended to reduce project duration. The Egyptians had a goal, and they had a system for building so that upwards of twenty thousand men could work simultaneously: leveling, marking out, quarrying, hauling, raising, constructing causeways, pyramid sides, valley temples; walls, chambers, etc.
This outstanding feat illustrates some of the features expected of systems engineering: well-planned, well managed, effective, efficient, systematic, organized. (The same might also be said of the ancient Egyptian civilization at that time which came to be looked upon as the Golden Age.) However, these features are not exclusive to systems engineering; project management might justly lay claim to some or all of them. What is it that is special or unique about systems engineering?
The motives for conceiving modern systems engineering are to be found, at least in part, in past disasters. Arthur D. Hall III cites: the chemical plant leakage in Bhopal (1986); the explosion of the NASA Challenger space shuttle(1986) and the Apollo fire (1967); the sinking of the Titanic (1912); the nuclear explosion in Chernobyl (1986) and the disaster at Three Mile Island power plant (1979). He cites, too, the capture of markets by Japan from the U.S., the decline in US productivity and the failure of the US secondary school system. He identifies the millions of people dying of starvation every year while other nations stockpile surplus food, medical disasters such as heart disease, while governments subsidize grains used to produce high cholesterol meat, milk and eggs; and many more. One implication is clear: systems engineering faces challenges well beyond the sphere of engineering.
The modern philosophy - the why and the how of today's systems engineering - developed partly at NASA in the 1960s and 1970s see the table.
Early Developments in Systems Engineering Philosophy
Developing Philosophy
Exemplifier
Systems engineering…
…requires a clear, singular mission and goal. To put a man on the Moon and return him safely to Earth by the end of the decade:
- There should be a sound concept of operations (CONOPS) from start to finish of the mission
- The mission will be executed in phases, marked by transitions. The first phase will place 3 men in Earth orbit
- There should be an overall system design that addresses the whole mission from start to finish. The full CONOPS should be demonstrably realized in the design
- An overall system design is needed to ensure there are no weak links in the chain and that the concept of operations can be completely realized
- Overall system design can be partitioned into complementary interacting subsystems. Each subsystem should have its own clear mission and concept of operations
- Each subsystem must be clearly defined in terms of fit, form, function and interface to assure overall system integrity upon integration
- Each subsystem may be developed independently and in parallel with the others, provided that fit, form, function and interfaces are maintained. Where any emerging deviations are unavoidable, whole system redesign may be revisited.
- As part of development, each subsystem should be rigorously tested in a representative environment, such as would be presented by the extremes of the mission and by the behaviour of other subsystems.
- Deviations at subsystem level will affect other subsystems and the overall mission system
- Upon integration of the subsystems, the whole system should be subject to tests and trials, real and simulated, that expose it to extremes of environment and to hazards, such as might be experienced during the mission. These would include full mission trials where recovery from defect was possible
- The whole system, including the operators and crew, should be subjected to rigorous tests and trials in representative environments, including hazards and emergencies.
Working on high profile, high-risk unprecedented ("green field") projects such as Apollo gave NASA clarity of view. Every ounce of payload counted. Every interface must match. Every function must contribute. Nor was it sufficient for the designer and developer of any one subsystem to state that his part was good and that it was someone else's problem. The whole mission system might be developed as separate subsystems, but they must all develop correctly to the overall plan or they may all be subject to change.
On the other hand, it became clear that the detail of what was inside the various subsystems was of secondary importance at system design level, provided the subsystem could be depended upon to operate as it should and meet the criteria of compatible fit, form and function.
Although the term may not have been used, theirs was - had to be - an organismic approach, constrained as they were by hard limits in rocket lifting power. The overall payload consisted of men, and several craft that fitted together like Russian dolls, including a moon-lander with sufficient fuel both to soft land and to lift off again into Moon orbit, a return craft with fuel, and a reentry vehicle. Changes to any one part of the payload that affected mass, moment of inertia, interface, form, volume, function, capability, etc., were likely to impact other parts of the payload, which would necessitate re-budgeting, redesign and redevelopment. The whole payload was, therefore analogous to an organism in that the various parts were mutually interactive, interdependent and combined to produce emergent properties, capabilities and behaviors.
It was not possible for one mission system designer to know the intimate details inside each and every subsystem, but then he did not need to know. There was a system design team for each subsystem, whose task was analogous to that of the whole mission system designers. These subsystem designers were vitally interested in detail down to sub-subsystem level. And so on.
So it became evident that the complexity could be encapsulated inside each subsystem, and that each subsystem, no matter how complex it might be on the inside could be represented by its external fit, form, function and interface. These properties were emergent, and so there arose the association between systems engineering and emergence. Although it may not have been stated as such, the emergent properties, capabilities and behaviors of the whole system derived from the emergent properties, capabilities and behaviors of the contained subsystems, their interactions and interchanges - there was no other source.
Philosophy Fundamentals
So the philosophy of systems engineering was, and is, fundamentally holistic, in the sense that any system should be conceived, designed and developed as a whole. The basis for this philosophy is to found in the pathology of systems that have failed in the past, in the poor performance of systems that have been "cobbled together" from available or separately developed, parts, and from the failure of many complex projects to complete at all. Only by considering the whole problem, the whole issue, is it feasible to conceive and create a whole, balanced solution. Holism not only recognizes the importance of 'wholes,' but also recognizes that wholes may be greater than the sum of their rationally separable parts, i.e., that wholes may exhibit emergent properties, capabilities and behaviours...
In some senses, systems engineering was - and is - about turning the notion, 'that some wholes cannot be deconstructed,' on its head, and creating systems from parts, such that the whole exhbits properties, capabilities and behaviors that are not exclusively attributable to any of the parts. Since emergence is not to be found in the parts alone, it follows then that emergence arises from the interactions between the parts of some whole, or from synergies between the many and various functions within the whole...and thereby lies the attractive potential of 'something for nothing,' for some additional capability, or efficiency, or effectiveness...
The philosophy of systems engineering is also organismic, in that the whole system-to-be-conceived/designed/created/operated is viewed as an open system, and as analogous to an organism. The various subsystem parts are organized, yet seen as interactive and mutually interdependent, such that constraints on the whole system necessitate both complementation and compromise within the parts and their interactions. Only in this way, and from this standpoint, is it possible to create optimal systems, that is, systems that satisfy limiting criteria such as value-for-money, cost-effectiveness, all-up mass/volume/form/moment, etc.
Finally, the philosophy of systems engineering is synthetic, in that systems are built from parts that are themselves systems, interconnected in such a way that the whole delivers requisite emergent properties, capabilities and behaviours. Synthesis is the opposite of reduction. Reduction looks into a system; synthesis looks out of a system. Reduction breaks down; synthesis builds up. Analysis, looking into things, yields knowledge; synthesis, looking outwards, gives understanding. Synthesis is more than integration, however. The parts that are brought together interact with, and mutually change, each other's behaviour. Each part contributes to the whole, and each part contributes to making the CONOPS manifest.
At the heart of systems engineering philosophy, however, is the intent to synthesize solution systems with emergent properties: essentially, this is what charaterizes and distinguishes systems engineering from other pursuits, practices and disciplines. Systems engineering is, in part, a legacy of the systems movement, and of systems theory and systems science. Systems researchers recognized that the behaviour of some 'wholes' could not be explained exclusively by the behaviour of their rationally separable parts - sometimes properties, capabilities and behaviours emerged as a direct result of interactions between those parts. Philosophically, it followed that it should be possible, at least in principle, to synthesize systems with prescribed emergent properties...and that possibility, that potential, gave rise to systems engineering.
Approaches to Systems Engineering
There are many different kinds of systems and many approaches to creating/enhancing systems. The following mind-map shows four:
- Unprecedented systems - i.e., those for which there has been no predecessor. These are sometimes referred to as green-field systems, when a system is being introduced where none has existed before
- Operational systems - those which are in operational use, and which require to be improved in terms of performance, opertional availability, effectiveness, etc.
- Volume Supply Systems - those used to manufacture products/artifacts/systems in volume. Volume Supply Systems are unusual in that they continually morph as they adjust to changes in the market, changes in fashion, change in demand, etc. They are also unusual in that they appear not to have a discernible lifetime. Like an animal body, they metabolize, continually replacing parts, skills, personnel, methods, etc., so anticipating and avoiding the onset of senility.
- Evolving Systems - these are systems where successive incarnations change only marginally from their predecessors, so emulating progressive evolution rather than any suggestion of revolution. The objective is to minimize risk.
References
Hall, Arthur D., III, Metasystems Methodology, Oxford, England: Pergamon Press, 1989.
Ackoff, R. L., Creating the Corporate Future, New York, NY: q, 1981.