An improved artificial bee colony algorithm with MaxTF heuristic rule for two-sided assembly line balancing problem

Two-sided assembly line is usually used for the assembly of large products such as cars, buses, and trucks. With the development of technical progress, the assembly line needs to be reconfigured and the cycle time of the line should be optimized to satisfy the new assembly process. Two-sided assembly line balancing with the objective of minimizing the cycle time is called TALBP-2. This paper proposes an improved artificial bee colony (IABC) algorithm with the MaxTF heuristic rule. In the heuristic initialization process, the MaxTF rule defines a new task's priority weight. On the basis of priority weight, the assignment of tasks is reasonable and the quality of an initial solution is high. In the IABC algorithm, two neighborhood strategies are embedded to balance the exploitation and exploration abilities of the algorithm. The employed bees and onlooker bees produce neighboring solutions in different promising regions to accelerate the convergence rate. Furthermore, a well-designed random strategy of scout bees is developed to escape local optima. The experimental results demonstrate that the proposed MaxTF rule performs better than other heuristic rules, as it can find the best solution for all the 10 test cases. A comparison of the IABC algorithm and other algorithms proves the effectiveness of the proposed IABC algorithm. The results also denote that the IABC algorithm is efficient and stable in minimizing the cycle time for the TALBP-2, and it can find 20 new best solutions among 25 large-sized problem cases.