A pruning pattern list approach to the permutation flowshop scheduling problem

This paper investigates an approach to the permutation flowshop scheduling problem based on Tabu Search with an additional memory structure called a ‘pruning pattern list’. The pruning pattern list approach allows a better use of the critical block information. A solution of the flowshop scheduling problem is represented by a permutation of job numbers. A pruning pattern is generated from a solution by replacing job numbers inside a critical block with ‘wild cards’ so that solutions that ‘match’ the pattern would be excluded from the search. A set of pruning patterns, which is called a ‘pruning pattern list’, is used to navigate the search by avoiding solutions that would match any pattern on the list. Computational experiments using benchmark problems demonstrate the effectiveness of the proposed approach.