The System Complexity Metric (SCM) Predicts System Costs and Failure Rates

A complex system has many parts and interactions and so is difficult to understand.
Systems with higher complexity generally have higher costs and failure rates. A System
Complexity Metric (SCM) is defined to be the sum of the number of nodes, N, in the system
block diagram plus the number of one-way interactions, I, between the nodes. SCM = N + I.
SCMs are easily determined by direct inspection of high level block diagrams of life support
systems. System cost was found to be directly proportional to SCM. The system MTBF (Mean
Time Before Failure) is the inverse of the system failure rate. MTBF = 1/f. The system MTBF
was found to be proportional to SCM^(-2.2) for estimated preflight MTBFs. As is typical for
systems that are not extensively tested and redesigned to eliminate unexpected failure modes,
the life support flight failure rates were about ten times higher than the preflight estimates
and the MTBFs one-tenth the preflight estimates. The system MTBF was found to be
proportional to SCM^(-2.6) for observed flight MTBFs.