Job Rejection Strategies in Genetic Algorithm for Permutation Flowshop Scheduling Problem with Rejection

The permutation flowshop scheduling problem with rejection is studied, with the objective of the minimization of the sum of total rejection cost of all rejected jobs and total tardiness of all accepted jobs. This problem is a joint decision with job rejection and production scheduling, and this paper presented three job rejection strategies under the framework of genetic algorithm. The first one is a serial strategy to reject jobs after scheduling decision. The second one is a parallel strategy to make the two decisions simultaneously. Improved parallel strategy is then presented by employing the serial strategy to optimize the solution obtained by the parallel strategy. Computational experiments are carried out based on a large number of randomly generated instances. The results illustrate that the serial strategy has the best effectiveness but the worst efficiency, and the improved parallel strategy has the close solution quality as the serial strategy but with much shorter computational time.