An efficient approach to stochastic jobshop schedulling: Algorithms and empirical investigations

Jobshop scheduling is one of the important classes of problems planning. The jobshop represents a common manufacturing environment, in which a set of machines is used to process a stream of jobs arriving randomly. The stochastic arrival pattern of the jobs, their processing sequences, and processing times render the scheduling of jobs a complex and difficult task. Furthermore, the scheduling task involves multiple conflicting objectives such as the minimization of the flow time, lateness and the number of late jobs among others. In this research,we develop a framework for the efficient scheduling of jobs considering these multiple objectives. The motivation for this research arises from the good performance of the interactive man-machine scheduling systems in achieving significant levels of attainment in the set of objectives, simultaneously. We develop scheduling strategies to the man-machine approach in this research. The algorithms developed in this research have been extensively tested and evaluated against the traditional scheduling methods using simulation studies. The results show that the proposed approach is efficient and viable for solving practical scheduling problems in the presence of multiple objectives.