Definition: The ability and practice of examining the whole rather than focusing on isolated problems (P. Senge) . The act of taking into account the interactions and relationships of a system with its containing environment (Y. Bar Yam, New England Complex Systems Institute).
Keywords: holism, holistic, interactions, multidimensionality, multiple perspectives, relationships, synthesis, synthetic, system thinking, systems thinking
MITRE SE Roles & Expectations: MITRE systems engineers are expected to: (a) understand the linkages and interactions among the elements of their system or enterprise and its connecting entities; (b) align goals and purposes across the enterprise; and (c) ask probing questions and trace the implications of potential answers across the enterprise.
The recently renewed interest in systems thinking in government circles and the engineering community has been fueled, in part, by the movement to apply systems science and complexity theory to problems of large-scale, heterogeneous, information technology-based systems.
Systems thinking is a framework for solving problems based on the premise that a component part of an entity can best be understood in the context of its relationships with other components of the entity, rather than in isolation. The way to fully understand why a problem occurs and persists is to understand the "part" in relation to the "whole." A focus of systems thinking is on understanding the linkages and interactions among the elements that compose the entirety. Systems thinking is often used in problems in which cause and effect are not closely related in space or time, as well as problems in which the relationships among components are non-linear (see also the Systems Engineering Strategies for Uncertainty and Complexity article in this Guide).
Systems thinking requires knowledge and understanding—both analysis and synthesis—represented in the same view. The ability to combine analytic and synthetic perspectives in a single view enables alignment of purposes, which is so important to successful engineering of enterprise capabilities. It allows the systems engineer to ask purposeful questions and trace the implications of potential answers across the enterprise. Would a change in performance at the subsystem level result in a change at the enterprise level? If so, how, and is it important? How would a new enterprise-level need be met?
The following concepts are important in applying systems thinking:
- Analysis: the ability to decompose an entity into deterministic components, explain each component separately, and aggregate the component behaviors to explain the whole. If the entity is a system, then analysis answers the question, "How does the system work?" Analysis results in knowledge of an entity; it reveals internal structure. For example, to know how an automobile works, you analyze it—that is, you take it apart and determine what each part does. This is essential to important activities like repairing automobiles or diagnosing and repairing problems of other, more complicated systems.
- Synthesis: the ability to identify the whole of which a system is a part, explain the behavior or properties of the whole, and disaggregate the whole to identify the role or function of the system in the whole. Synthesis answers the "Why is it what it is?" question. Synthesis is the mode of thought that results in the understanding of an entity (i.e., an appreciation of the role or function an entity plays in the larger system of which it is a part). As an example, the answer to the question of why the American automobile was originally designed for six passengers is because the average family size at the time was 5.6. Every MITRE systems engineer who has defined a system performance specification against mission or operational requirements has used synthetic thinking.
To analyze an unmanned aerial vehicle (UAV), we would logically decompose it into subsystems that perform certain roles or functions (e.g., the host platform, sensors that perform ground target detection and identification, communication systems for receiving commands from controlling entities, and transmitting onboard sensor data to control systems, etc.) down to a level of granularity sufficient to answer the question at hand.
To understand the UAV system synthetically, the first step is to identify the larger system of which the UAV is a part, (e.g., a situation awareness [SA] system of systems [SoS]). The second step is to describe the containing system (e.g., the SA SoS delivers high quality location and identification information on militarily significant objects of interest in a surveillance volume, with an emphasis on ground, sea surface, and low-altitude air vehicles). The third step is to disaggregate the whole to identify the role or function in the larger system of which the UAV is a part (e.g., the organization employing the SA SoS has a mission that focuses on the detection, location, and identification of ground and sea surface vehicles. The UAV in question is equipped with sensors and processing tailored to ground vehicle detection.). Taken together, this synthetic view explains why the organization has ground vehicle detection UAVs in its SA SoS and provides a basis for asking and answering "what if?" questions about the UAV, like: "What if the organization's mission shifted away from ground vehicle detection or moved more towards it?"
Combining the analytical and synthetic perspectives in a single view allows the systems engineer to ask questions and draw implications of potential answers across the enterprise. If the organization's mission shifted to ultra-light airborne vehicle detection, how would SA be accomplished? Could the existing UAVs be re-engineered or refitted with new sensors to detect and identify the new target types? Would a change in performance at the UAV system level result in a change at the SA SoS or mission level? If so, how, and is it important?
Government Interest and Use
The need to apply systems thinking continues to be pervasive across MITRE. It is expected of MITRE by our sponsors. Reference to it is made in our sponsoring agreements:
- From the FAA Sponsoring Agreement: "CAASD is uniquely qualified...to solve problems that are too broad and too complex to become the focus of a competitive procurement..."
- From the DoD Sponsoring Agreement: "While serving the immediate needs of the many individual programs it supports, the C3I FFRDC aligns its work program to assist in achieving integrated DoD-wide enterprise capabilities..."
- From the IRS Sponsoring Agreement: "The FFRDC shall simultaneously direct its efforts to the support of individual programs and projects for tax modernization, and to assuring that these individual programs and projects operate effectively with each other and efficiently support the business objectives of the government..."
Systems Thinking Best Practices
Systems engineering and systems thinking have always been about asking good questions and forming conclusions and recommendations based on the answers. The following are examples of systems thinking questions to consider asking when performing your MITRE systems engineering activities:
- What is my enterprise? What elements of it do I control? What elements do I influence? What are the elements of my environment that I do not control or influence but which influence me? [2, pp. 2-3 – 2-4]
- Can a balance be achieved between optimizing at the system level and enabling the broader enterprise? If the balance comes at the expense of the smaller system, can that be offset or mitigated? How?
- Is interdependence of performance measures (variables) in a system or enterprise hidden by slack? Is the inability to make progress in one measure, except at the expense of others, an indication that the slack among them has been used up? Can a redesign of the system or enterprise remove interdependence or provide additional slack? [2, pp. 4-9 – 4-10]
- How can analytic and synthetic perspectives be combined in a single view to enable alignment of purposes across the enterprise? Would a change in performance at the subsystem level result in a change at the enterprise level? If so, how, and is it important? How would a new enterprise level requirement be met and how would it influence its constituent systems?
- Can the solution space of a seemingly intractable problem be expanded by viewing it in its containing whole? How? [2, pp. 4-3 – 4-4]
References & Resources
- Senge, P. et al., 1994, The Fifth Discipline Fieldbook, New York, NY, Doubleday.
- Rebovich, G., 2005. Systems Thinking for the Enterprise: New and Emerging Perspectives, The MITRE Corporation.
Additional References & Resources
Ackoff, R., November 1993, "From Mechanistic to Social Systemic Thinking," System Thinking in Action Conference.
Axelrod, R. and M. D. Cohen, 2000, Harnessing Complexity: Organizational Implications of a Scientific Frontier, New York, NY, Basic Books.
Gharajedaghi, J., 1999, Systems Thinking: Managing Chaos and Complexity. Boston, MA, Butterworth Heinemann.
"Program Formulation & Project Planning," MITRE Project Leadership Handbook, viewed January 16, 2010.
Rebovich, G., 2006, "Systems Thinking for the Enterprise: A Thought Piece," International Conference on Complex Systems, Bedford, MA, The MITRE Corporation.
Rebovich, G., 2006, "Systems Thinking for the Enterprise: New and Emerging Perspectives," Proceedings of 2006 IEEE International Conference on Systems of Systems
"Systems Integration," MITRE Project Leadership Handbook, viewed January 16, 2010.