Modularity Research

A system is defined as a combination of interacting elements organised to achieve a stated purpose. Since the purpose of the system can be met in a number of ways, a designer can choose one of a number of possible combinations of elements in order to achieve the stated purpose.

Modularity (the degree to which a system's components may be separated and recombined) is a therefore principal notion for all engineered systems, so much so that the need for modularity is almost taken for granted. In the context of system development, Baldwin and Clark (2006) suggest three natural purposes for modularization: to make complexity manageable, to enable parallel work, and to accommodate uncertainty.

There is, however, a limit to the utility of modularization in any context, and modularization is a delicate balancing act – either too few modules or too many modules can have negative consequences on the constructability of a particular design, and possibly on the performance of system. This dichotomy raises the following questions:

  • Is there an optimal level of modularity?
    • How much modularity is enough?
    • How much modularity is too much and why?
    • Is there a measure or an equation or a process that can be "optimized" to achieve "optimal modularity"?
    • If there is an optimal modularity, how much better is an optimally modular system than a non-optimal one, and by what measure?
    • What information is essential for optimal modular systems design?
    • Do the answers to these questions rely on context? That is, is there a general rule for optimality, or does it depend on the nature (such as type or size) of the system?
  • Does optimal modularity depend on the purpose of the system–that is, does it depend on the life-cycle stage?
    • What is an optimal method of modularity for each life-cycle stage?
    • Is the design payoff from any life-cycle perspective sensitive to optimal modularity? That is, can optimality be considered for each life cycle stage, or must it be considered across all stages?

Baldwin, Carliss Y., and Kim B. Clark. 2006. "Modularity in the Design of Complex Engineering Systems." In Complex Engineered Systems, edited by Dan Braha, Ali A. Minai and Yaneer Bar-Yam, 175-205. Springer Berlin Heidelberg.