Performance evaluation for concurrent processing in cyclic systems

A new technique for modelling and performance evaluation of a loosely synchronized system composed of a set of cyclic processes interacting via shared and reusable resources is presented. Analytical models relating system dynamics to the behaviour of component processes are considered. Their application within a framework of self-synchronization is discussed. The approach is applicable to cyclic concurrently interacting processes, the evolution of which depends on the complex interactions of various discrete events such as the arrival or departure of a job or the initiation and completion of a task. The analysis is illustrated using the example of a robotized flexible assembly system, where one of the main concerns is optimization of a performance index related to the system steady-state behaviour.