A Novel Simulated Annealing-Based Hyper-Heuristic Algorithm for Stochastic Parallel Disassembly Line Balancing in Smart Remanufacturing

Abstract

Remanufacturing prolongs the life cycle and increases the residual value of various end-of-life (EoL) products. As an inevitable process in remanufacturing, disassembly plays an essential role in retrieving the high-value and useable components of EoL products. To disassemble massive quantities and multi-types of EoL products, disassembly lines are introduced to improve the cost-effectiveness and efficiency of the disassembly processes. In this context, disassembly line balancing problem (DLBP) becomes a critical challenge that determines the overall performance of disassembly lines. Currently, the DLBP is mostly studied in straight disassembly lines using single-objective optimization methods, which cannot represent the actual disassembly environment. Therefore, in this paper, we extend the mathematical model of the basic DLBP to stochastic parallel complete disassembly line balancing problem (DLBP-SP). A novel simulated annealing-based hyper-heuristic algorithm (HH) is proposed for multi-objective optimization of the DLBP-SP, considering the number of workstations, working load index, and profits. The feasibility, superiority, stability, and robustness of the proposed HH algorithm are validated through computational experiments, including a set of comparison experiments and a case study of gearboxes disassembly. To the best of our knowledge, this research is the first to introduce gearboxes as a case study in DLBP which enriches the research on disassembly of industrial equipment.

An artificial bee colony based-hyper heuristic algorithm with local search for the assembly line balancing problems

“…Due to the advantages of hyper-heuristics, many studies utilised the hyper-heuristic algorithms to tackle a wide range of optimisation problems such as university course timetabling (Bai et al ., 2007; Soria-Alcaraz et al ., 2016), knapsack problems (Gölcük and Ozsoydan, 2021), exam timetabling problems (Hao et al ., 2020), vehicle crash-worthiness problem (Li et al ., 2017), quadratic assignment problem (Dokeroglu and Cosar, 2016) and scheduling problems (Koulinas et al ., 2014; Hart and Sim, 2016; Lin et al ., 2017; Deliktaş, 2021). Furthermore, a few studies have explored the application of hyper-heuristic algorithms in addressing different assembly line variations, such as aircraft final ALB (Bao et al ., 2023), two-sided ALB (Rong et al ., 2023), stochastic parallel disassembly line balancing (Hu et al ., 2023), parallel ALB (Seçme and Özbakır, 2019; Özbakır and Seçme, 2022), mixed-model ALB (Ebrahimi et al ., 2023; Cano-Belmán et al ., 2010) and robotic parallel with type-II (Çil et al ., 2017). The existing literature lacks sufficient studies that investigate the use of hyper-heuristic algorithms for solving the simple assembly line problem with type-I.…”

A Two-Stage Screw Detection Framework for Automatic Disassembly Using a Reflection Feature Regression Model

“…Disassembly is the first and inevitable operation in the remanufacturing process due to stricter environmental regulations and growing demand for effective product remanufacturing [ 8 , 9 , 10 ]. In current applications, manual disassembly is still the main strategy, which consumes a large amount of energy and exposes operators to dangerous materials.…”