Comparative study of search methods and heuristic functions for scheduling of flexible manufacturing systems

Abstract

The scheduling method that uses Petri net modeling and heuristic search[9] is extended to simultaneously optimize multiple objectives which exist in flexible manufacturing systems(FMSs). This method formulates the problem with timed-place Petri net model and applies a search algorithm to the reachability graph of the Petri net model in order to generate an optimal or near-optimal schedule. A difficulty when applying this method to relatively large scheduling problems, is the memory and time requirements. To overcome this, the search algorithm can be modified. The methods to modify the search algorithm are as follows: (i) limit the number of nodes in the list OPEN, method1. (ii) forget the history of the past search at regular intervals, i.e. compare the newly expanded node with the limited number of nodes in the list CLOSED, method2. (iii) search with the list OPEN and an evaluation function, method3. (iv) search only with the list OPEN and no evaluation function, method4. The search algorithm employs an evaluation function including multiple objectives. New heuristic functions are developed for efficient search. Experimental results are presented to show the effectiveness of the proposed method.