LAUSR

The flexible job-shop environments have become increasingly significant because of rapid improvements on shop floors such as production technologies, manufacturing processes and systems. Several real manufacturing and service companies have had to use alternative machines or processes for each operation and the availability of alternative process plans for each job in order to achieve good performance on the shop floor where conflicting objectives are common, e.g. the overall completion time for all jobs and the workload of the most loaded machine. In this paper, we propose a Pareto approach based on the hybridization of an estimation of distribution algorithm and the Mallows distribution in order to build better sequences for flexible job-shop scheduling problems with process plan flexibility and to solve conflicting objectives. This hybrid approach exploits the Pareto-front information used as an input parameter in the Mallows distribution. Various instances and numerical experiments are presented to illustrate that shop floor performance can be noticeably improved using the proposed approach. In addition, statistical tests are executed to validate this novel research.