Two-dimensional packing algorithm for autoclave molding scheduling of aeronautical composite materials production. (August 2020)

• Record Type: Journal Article
• Authors: Xie, Naiming; Zheng, Shaoxiang; Wu, Qiao
• Is Part Of: Computers & industrial engineering. Volume 146(2020)
• Publication Date: 2020-08
• Subjects: Composite material -- Scheduling -- Autoclave molding -- Two-dimensional packing algorithm -- Heuristic algorithm; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2020.106599
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: A novel modeling of the problem considering the two-dimensional constraints. VSBP-TC under the varied bin packing approaches to deal with the problem. Different scales numerical cases showed that it is a more applicable version. Real constraints to propose a closer look to reality and a more applicable model. Abstract: Due that the usage of composite material is increasing rapidly in aircrafts. To improve production efficiency is undoubtedly a useful strategy for solving the contradiction of production and supply of composite materials. This paper aims to study the autoclave molding scheduling problem so as to break through the bottleneck of composite material production. Considering tasks processed in an autoclave is in a batch rather than one by one, the constraint about space matching of tasks and inner platform of the autoclave was transformed as a two-dimensional rectangle bin packing problem. And then a mixed integer programming model was established for solving autoclave scheduling problems. In which, the objective function was defined as minimizing the makespan of all batches. A hybrid algorithm of the heuristic rule and local exact optimizing strategy was further designed to solve the model which imposed the geometric constraints accordingly. Finally, different scales of computational instances were adopted to test the effectiveness and efficiency of the proposed algorithm. Results show that the proposed algorithm is more effective than meta-heuristics and more efficient than exact methods, especially when the task scale is relatively large. Therefore, it is a feasible and effective schedule for composite material production scheduling.