Improving Facility Layout Using Genetic Algorithm and Simulated Annealing

PT XYZ is a manufacturing company in the automotive field. Based on the current layout conditions, the metal division is not optimal, as it was not initially designed with proper layout planning, particularly considering the proximity between workstations. As a result, the material flow is inefficient. This can be observed from the placement of several workstations that are still far apart, leading to long material transfer distances during the production process. Given this situation, it is necessary to improve the production floor layout by first calculating the initial material handling costs, and then determining a new production facility layout with a more optimal material transfer distance. Overall, the layout derived from applying the Genetic Algorithm (GA) based on Simulated Annealing (SA) as the initial design achieved an MHES cost of Rp 83,304,888.21. This reflects a cost difference of Rp 30,484,999.71 when compared to the initial layout, indicating a 23.40% reduction in MHES costs. In contrast, the difference in MHES costs between the Genetic Algorithm (GA) and the initial layout is Rp 27,172,147.68, representing a 20.86% decrease. Additionally, the difference between the Genetic Algorithm (GA) costs and those from the Simulated Annealing (SA) results is Rp 46,934,016.13, leading to a 36.03% cost reduction. Therefore, the optimal choice for the layout design in this study is the Genetic Algorithm (GA) based on the results of Simulated Annealing (SA) due to its superior MHES cost efficiency.