A branch and price approach to the two-agent integrated production and distribution scheduling. (October 2019)

• Record Type: Journal Article
• Authors: Gharaei, Ali; Jolai, Fariborz
• Is Part Of: Computers & industrial engineering. Volume 136(2019)
• Page Start: 504
• Page End: 515
• Publication Date: 2019-10
• Subjects: Supply chain management -- Branch and price -- Scheduling and distribution -- Multi-agent -- Column generation; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2019.07.050
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: Present a two-agent view to the parallel multi-site supply chain scheduling problem. Develop a mixed integer programming formulation to this problem. Consider multiple customers, routing decisions and batch delivery for this problem. Introduce a hybrid branch and price framework as exact method. Introduce a Bees algorithm to generate initial solutions. Abstract: The integration and coordination of decisions is one of the new approaches in the supply chain environment. The proper orders scheduling and distribution strategy of final products are two key factors in this integration. Here, an integrated production scheduling and distribution problem with routing decisions is discussed in a multi-site supply chain. This problem has been investigated from a multi-agent perspective in which customers' sets, as agents, want to optimize a specific objective on his own set of orders. A mixed integer linear programming (MILP) formulation is developed for this problem. Due to the complexity, it is not logical to solve the problem in a straightforward way such as commercial solvers. Hence, a branch and price framework is introduced in which a Bees algorithm (BA) is used to construct initial columns. Various computational experiments are conducted to evaluate the efficiency of the proposed approach. The results show that the proposed algorithm is superior to the standalone branch and price algorithm and MILP solver in terms of computation times and gap.