A heuristic approach for a scheduling problem in additive manufacturing under technological constraints. (April 2021)

• Record Type: Journal Article
• Authors: Aloui, Aymen; Hadj-Hamou, Khaled
• Is Part Of: Computers & industrial engineering. Volume 154(2021)
• Publication Date: 2021-04
• Subjects: Additive manufacturing -- Production time -- Scheduling -- Planning -- Placement; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2021.107115
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: We investigate the problem of production scheduling using Additive Manufacturing. We propose two models to estimate the production time for two technologies. We propose MILP model for the placement and scheduling problem. We develop a heuristic to solve large instances of the problem. The numerical experiments indicate that the heuristic proposed is efficient. Abstract: In the context of the future industry, companies have urged to innovate the manufactured products. Today, additive manufacturing makes it possible to respond to the needs of the market in terms of customized production. The recent advances in additive manufacturing technologies have caused a considerable increase in the number of products manufactured by additive processes in industries. In order to satisfy customers' demands and make the investment in additive machines profitable, it is necessary to deal with the production organization in additive manufacturing. This research focuses on the scheduling and nesting problem of production with technological constraints. The objective is to minimize the total delay of the parts to be produced and to maximize the use rate of the additive manufacturing machines. Two models are proposed for powder-based laser technologies and multi jet fusion technology, to estimate the production time in additive manufacturing based on real data. The nesting and scheduling problem is modelled by mixed linear programming. A small example is used to validate the proposed model using the Cplex solver. Due to the NP-hardness of the problem studied, this research develops a heuristic approach to solve large-sized instances. Computational experiments conducted on small and medium size instances indicate that the proposed heuristic is capable to give better solutions within a reasonable time. To evaluate the heuristic performances on large instances, a comparison of the heuristic results is performed with the lower bounds obtained by relaxing the model. The numerical results show that the solutions found by our heuristic are near to the lower bounds proposed.