A column generation based distributed scheduling algorithm for multi-mode resource constrained project scheduling problem. (November 2018)

• Record Type: Journal Article
• Authors: Changchun, Liu; Xi, Xiang; Canrong, Zhang; Qiang, Wang; Li, Zheng
• Is Part Of: Computers & industrial engineering. Volume 125(2018)
• Page Start: 258
• Page End: 278
• Publication Date: 2018-11
• Subjects: Multi-mode resource constrained project scheduling problem -- Distributed scheduling -- Information-sharing -- Column generation; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2018.08.036
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: Develop a column generation based distributed scheduling algorithm. A multi-mode resource constrained project scheduling problem considering multi-product is studied. Computational experiments are conducted to verify the effectiveness of the algorithm. A general framework of distributed algorithm is proposed to solve a generalized problem. Abstract: This study develops a column generation-based distributed scheduling algorithm for multi-mode resource constrained project scheduling problem. The proposed distributed algorithm shares less information among independent decision makers compared with the traditional integrated approach. In this problem, many independent processors, which can produce different types of products, coordinate with a resource manager (or third-party logistic), who provides different types of vehicles to deliver products to the customers. The problem is decomposed into two parts: production planning for individual processors and vehicle scheduling for the resource manager. A total of 1200 instances are randomly generated to verify the effectiveness of the proposed distributed algorithm. Results of computational experiments verify that the proposed distributed algorithm has good solution quality and calculation efficiency compared with the integrated approach, i.e., CPLEX, hybrid Benders Decomposition, and Lagrangian Relaxation. Lastly, a general framework for the distributed algorithm is proposed to solve a generalized problem.