Job shop control under influence of chaos phenomena

Efficient short-term production control depends on the ability to forecast future behavior of a job shop when it is loaded according to a particular schedule. A simulation of the job shop process based on a well-suited job shop model can be used to evaluate that schedule according to such goals as low in-process inventory, high utilization of workers and facilities, and low tardiness so that some optimization by modifying the schedule can be performed. A dedicated class of augmented Petri nets is introduced as a notation for microscopic deterministic job shop models which are well suited for short-term production control, including job shop scheduling. It is shown how these Petri nets can be automatically transformed into CPM nets using occurrence structures. These CPM nets represent the job shop schedule. In contrast to Gantt charts they are not restricted to the information about the temporal ordering of events of the shop floor, but also contain the full causal structure of the job shop process as well as assertions about the critical path. An attempt is made to determine whether there is a fundamental limit of deterministic simulation of manufacturing facilities emerging out of chaos phenomena due to incomplete specification of deterministic models