MINIMIZATION OF TOTAL TARDINESS ON UNRELATED PARALLEL MACHINES VIA SIMULATED ANNEALING

In the classical Parallel Machine Scheduling problem, there are n jobs and m machines. Each job needs to be executed on one of the machines during a fixed processing time. So the aim is to find the Schedule that optimizes a certain performance measure. This paper addresses, using simulated annealing method, the unrelated parallel machine scheduling problem to minimize total tardiness and as a secondary goal to minimize makespan. The performance of the proposed simulated annealing procedure tested on a set of problems, consisting five problems, each has 3 machines and 28 jobs. Computational results show that the proposed simulated annealing procedure is promising to tackle the unrelated parallel machine scheduling problem without any setup and precedence relations.