Investigation of large scale multi-echelon service systems

There exist many instances of large scale systems composed of operational units each having a number of standalone operating machines experiencing periodic breakdowns. Within this category of systems, there are some which experience relatively severe operating conditions and a high level of utilization, and so have a significant requirement for maintenance. Studies are required to facilitate the development of managerial policies for the design and operation of such systems. Past work in the operations management and operations research literature has used many restrictive assumptions in order to make optimization studies of these systems feasible. Unfortunately, this has led to results which are often limited in their applicability. In this paper, simulation analysis is used to investigate the impact of various policies in one such system with a view of testing certain observed practices and intuitively appealing heuristic decision rules, in order to provide some guidance for actual decisions. In particular, the interdependent policies for operating machine replacement, transportation, distribution and maintenance of a common type of system are investigated for a range of system capacity, demand and failure parameters. The analysis leads to various conclusions, some agreeing with intuition and others not, that provide guidance for system design and operation.