Determination of the Fleet Size of AGVs with AGV Pools Using a Genetic Algorithm and Artificial Intelligence

The utilization of low-cost AGVs in the industry is increasing every day, but efficiency these systems low due to lack a central management system. Low-cost AGVs’ main characteristic navigation via magnetic sensors, which they follow tape on ground with low-level automation disadvantages are mainly only one circuit assignment and system intelligence. Therefore, this study, AGV pools were employed determine required number. This study begins by calculating number for each combination assigned parking station time window approach. Mathematical-solution-based mixed integer programming was developed find optimum solution. Computational difficulties handled development genetic-algorithm-based approach solutions complex cases. If production requirements change, parameters can be changed adapt there need AGVs. It demonstrated that pool combinations did not lead any loss available shown proposed provides fleet size requires five fewer AGVs, 29% reduction effects parameter changes also investigated artificial neural networks (ANNs) estimate case requirement changes. necessary effect change without compromising computational cost time, especially systems. Thus, an network (ANN), response surface method (RSM), multiple linear regression (MLR) techniques used examine In present case, ANN obtained solution at good rate reduced costs, correction errors compared GA, 0.4% 7% 24% (MLR). results show workshops predict parameters.