Modeling and Optimization of Two-Sided Disassembly Line Balancing Problem Considering Interactions Among Precedence Free Tasks

In disassembly process, task time interactions between two tasks may occur when they do not have any precedence relationship, and this will increase the disassembly time and affect productivity. Therefore, this paper provides an attempt to study a multi-objective two-sided disassembly line balancing problem considering interactions among precedence free tasks, and establishes a mathematical model from efficiency and environment concerns. Due to NP-hard nature of the presented problem, a novel hybrid discrete artificial bee colony algorithm is proposed based on the problem knowledge of the problem to obtain satisfactory solutions within a short computational time. A new encoding scheme based on one-dimensional integer array is designed to represent feasible solutions. A random initialization method based on Logistic chaotic mapping is used to produce an initial population with diversity. The INSERT, SWAP and MUTATION operators developed according to the characteristics of the problem are employed to construct the set of neighborhoods for the VNS method to exploit new solutions. Instead of randomly exploring new solutions in the whole solution space, a MUTATION operator performed on the current best solution is designed to avoid blindness in the exploration progress. The effectiveness and superiority of the proposed algorithm are verified by a series of instances. The analysis results indicate that the proposed algorithm is suitable to solve the presented problem as well as its variants and shows superior performance. Moreover, different disassembly solutions are compared to illustrate the important of balancing the tasks among the disassembly line.