Research on Production Balance Technology of Robotic Assembly Line Based on Fruit Fly Optimization Algorithm

To address the balancing problem of the robot bilateral assembly line, a mathematical model is constructed with cycle time and energy consumption as optimization objectives, subject to multiple constraints. This model is solved using an Improved Self-adaptive Fruit Fly Optimization Algorithm (ISFOA). In the algorithm design, tent chaotic mapping is introduced during population initialization to ensure uniform distribution of the population. Additionally, an individual information utilization mechanism is incorporated, where there is an 80% probability of generating the next generation based on the global optimum and a 20% probability of generating it through random uniform perturbation of individuals, thereby fully utilizing population information. Simulated annealing-based multiple mutation rules and local perturbation are also integrated into the algorithm. Considering the characteristics of the model, a four-layer coding scheme is designed, and a decoding method is employed to achieve a better balance between local and global optima. Finally, the effectiveness of the proposed ISFOA algorithm for solving the Robotic Two-sided Assembly Line Balancing Problem of Type-2 (RTALBP-2) is verified by solving various cases of different scales and comparing its performance with other algorithms.