Herb Simon on maintenance

Herb Simon on maintenance
Art by Wu Guanzhong

Stewart Brand asks a great question: Did Herbert Simon explore the maintenance of systems?

I haven’t found Simon discuss maintenance explicitly. But two related themes are worth highlighting:

(1) Simon discusses maintenance through the topic of homeostasis. “For example, Maintenance of a system becomes much simpler if the internal temperature is constant.” Keeping a limited set of properties steady state allows resilience to variability and uncertainty in the surrounding environment. In the long run, this enables adaptability.

To Simon, homeostasis reduces system complexity, albeit at a cost of new complexities in the homeostatic mechanisms themselves.

(2) It was actually a question of Maintenance Management that first piqued Simon’s curiosity (at age 19) and became the research problem to fuel his years of work. The question: “How should funds be allocated between playground maintenance vs. programmed activities?”

To Simon, answers from economics 101 were insufficient for 2 reasons: (a) “no one actually thought of the decision in this way, so the hypothesis failed descriptively” (b) it was not obvious how to value one option vs the other, “so the hypothesis failed prescriptively.” (See Ask a Simple Question below.)

Simon concluded that the question of evaluating maintenance was actually a question of how people reason, and that this reasoning through allocation problems is more motivated by identity and loyalty than we might realize.


Maintenance was the problem that spawned Simon’s research: “How should funds be allocated between playground maintenance and programmed activities?”

Simon concluded that “the ready hypothesis from economics”– ie “Divide the funds so that the next dollar spent for maintenance will produce the same return as the next dollar spent for leaders’ salaries” was deficient for 2 reasons: (1) No one thought of the decision in this way, so the hypothesis failed descriptively. (2) It was not obvious how to way the value of one against the value of the other, so the hypothesis failed prescriptively.

“For Simon, the specific problem about behavior on allocations to playgrounds in Milwaukee became a much more general problem: How do human beings reason when the conditions for rationality postulated by neoclassical economics are not met?” His answer: it actually came down to organizational identity. To study allocation problems, the better question asks about how people identify and feel loyalty to their respective organizations.

Principals that inform the design of a system: Homeostasis, Membranes, Specialization, Temporal Specialization

  • Homeostasis– resilience vs. external factors
  • “However complex the external environment, the internal environment becomes much simpler. A familiar example of homeostasis is temperature control. As the rates of various chemical reactions are differentially sensitive to temperature, maintenance of a system becomes much simpler if the internal temperature is constant.”
  • “Another mechanism of homeostasis: maintaining inventories of the inputs to internal processes […] inventories are important to the degree that there is uncertainty about the time required to obtain required substances from the environment, and to the degree that the substances can be storied efficiently”
  • “We see that both variability of the environment and its unpredictability impose a need for homeostasis, and we might conjecture, therefore, that chaotic environments would call for especially powerful homeostatic mechanisms. here again, feedback comes to the rescue, for appropriately contrived feedback devices can sometimes maintain a system within a specific small subregion of the strange attractor of a chaotic system; whereas without feedback, the system would be doomed to range over the whole area of the strange attractor, with consequent wide variability of its environment”
  • “In general we can think of homeostasis as a method of reducing system complexity at the cost of some new complexities in the form of the homeostatic mechanisms themselves”