no-idle time scheduling of open shops: modeling and meta-heuristic solution methods

In some industries as foundries, it is not technically feasible to interrupt a processor between jobs. this restriction gives rise to a scheduling problem called noidle scheduling. this paper deals with scheduling of noidle open shops to minimize maximum completion time of jobs, called makespan. the problem is first mathematically formulated by three different mixed integer linear programming models. since open shop scheduling problems are nphard, only small instances can be solved to optimality using these models. thus, to solve large instances, two metaheuristics based on simulated annealing and genetic algorithms are developed. a complete numerical experiment is conducted and the developed models and algorithms are compared. the results show that genetic algorithm outperforms simulated annealing.