A random-keys genetic algorithm for scheduling unrelated parallel batch processing machines with different capacities and arbitrary job sizes

Abstract

A batch processing machine (BPM) can simultaneously process several jobs and has wide applications in various industrial environments. This paper studies the problem of minimizing makespan on unrelated parallel BPMs with non-identical job sizes and arbitrary release times. In the environment of unrelated machines, each machine has a processing speed for each job. The unrelated BPM problem is the most general case of parallel BPM problems and is closer to actual production conditions. The problem under study is NP-hard. We present two lower bounds for the problem. Then a genetic algorithm based on random-keys encoding is proposed to solve the problem. The performance of the proposed algorithm is compared with a commercial solver (ILOG CPLEX) and two meta-heuristics published in the literature: a recent iterated greedy algorithm and a particle swarm optimization algorithm. Computational experiments show that the proposed algorithm produces better solutions compared to the other methods. The quality of the proposed lower bounds is evaluated as well.