No abstract

In traditional scheduling, job processing times are assumed to be fixed. However, this assumption may not be applicable in many realistic industrial processes. Using the job processing time of real industrial processes instead of a fixed value converts the deterministic model to a stochastic one. This study provides three approaches to solving the problem of stochastic scheduling: stochastic linguistic, stochastic scenarios, and stochastic probabilistic. A combinatorial algorithm, dispatching rules and community evaluation chromosomes (DRCEC) is developed to generate an optimal sequence to minimize the tardiness performance measure in the scheduling problem. Thirty-five datasets of scheduling problems are generated and tested with the model. The DRCEC is compared to the Genetic Algorithm (GA) in terms of total tardiness, the tendency of convergence, execution time, and accuracy. The DRCEC has been discovered to outperform the GA. The computational results show that the DRCEC approach gives the optimal response in 63 per cent of cases and the near-optimal solution in the remaining 37 per cent of cases. Finally, a manufacturing company case study demonstrates DRCEC's acceptable performance.

Scheduling different jobs in an appropriate sequence are very important in manufacturing industries due to the influence of conflicting criteria. It becomes difficult to sequence the jobs as the job number increases due to numerous computations involved. In this article, six jobs are considered to be treated on a machine one by one. Seven different priority sequencing rules provide seven different sequencing options for the jobs which are assessed using a set of nine criteria. Preference selection index (PSI) approach, a multi-criterion decision-making (MCDM) technique is proposed to rank them from best to worst. The PSI approach, unlike other MCDM methods, does not require to find the relative significance of the criteria, which reduces work of finding weights of criteria hence is a very easy and effective tool for decision making. A benchmark problem from the previous literature is considered and solved using the PSI approach and the obtained results are found to be correct.

No abstract

No abstract