Abstract: In view of the fact that the logistics interaction point of the facility in the industrial design does not overlap with the midpoint of the aisle sideline, an aisle layout problem that considers the left and right mirror images of the facility is proposed. The integer programming model of the problem is established, and an improved discrete butterfly optimization algorithm suitable for MFCAP is proposed. Based on the standard butterfly optimization algorithm, the algorithm discretizes the encoding method and related operations, improves the search speed of the algorithm through adaptive mode switching probability, and uses methods such as elite enhanced evolution and reverse diffusion catastrophe to improve the search accuracy of the algorithm. In order to verify the correctness of the proposed model, branch and bound algorithm and improved discrete butterfly optimization algorithm are used to accurately solve small-scale MFCAP examples. In order to verify the effectiveness of the proposed algorithm, the improved discrete butterfly optimization algorithm is compared with other heuristic algorithms in medium and large-scale calculation examples. The results show that the proposed improved discrete butterfly optimization algorithm has high quality and efficiency when applied to MFCAP, and it is an effective method to solve the MFCAP problem.

Key words: corridor allocation problem (CAP), butterfly optimization algorithm, integer programming model, mirrorable facility (MF)