Two-agent single-machine scheduling to minimize the batch delivery cost. (February 2016)

• Record Type: Journal Article
• Authors: Yin, Yunqiang; Wang, Yan; Cheng, T.C.E.; Wang, Du-Juan; Wu, Chin-Chia
• Is Part Of: Computers & industrial engineering. Volume 92(2016)
• Page Start: 16
• Page End: 30
• Publication Date: 2016-02
• Subjects: Scheduling -- Batch delivery -- Two agents -- Fully polynomial-time approximation scheme; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2015.12.003
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: Integrated production and batch delivery scheduling with two agents is considered. Constrained optimization problems on various criteria of the two agents are addressed. Computational complexity status and solution procedures are developed for the problems. Numerical studies are conducted to evaluate the performance of the algorithms. Abstract: We consider integrated production and batch delivery scheduling in a make-to-order production system involving two competing agents, each of which having its own job set competes to process its jobs on a shared single machine. To save the delivery cost, the jobs of the same agent can be processed and delivered together batches. The completion time of each job in the same batch coincides with the batch completion time. A batch setup time is incurred before the processing of the first job in each batch. Each of the agents wants to minimize an objective function depending on the completion times of its own jobs. The goal is to determine a schedule for all the jobs of the two agents that minimizes the objective function of one agent, while keeping the objective function value of the other agent below or at a given value. For each of the problems under consideration, we either provide a polynomial-time algorithm to solve it or show that it is NP -hard. In addition, for each of the NP -hard problems, we present a mixed integer linear programming (MILP) formulation and provide a pseudo-polynomial dynamic programming algorithm, establishing that it is NP -hard in the ordinary sense only, and show that it admits an efficient fully polynomial-time approximation scheme, if viable. Finally, we compare the performance of the pseudo-polynomial dynamic programming algorithms against the corresponding MILP formulations with randomly generated instances.