The Facility Layout Problem (FLP) is a well-known NP-hard problem in the field of mathematical programming. This study aims to enhance conventional approaches for solving FLP using a genetic algorithm (GA) from three critical perspectives: (1) using polygons to simulate facility areas, (2) optimizing the total distance as the objective function, and (3) developing algorithms to identify spatial and temporal interferences in actual processes as constraint conditions. The simulation results for a layout including six facilities demonstrated no spatial or temporal interferences, with maximal individual fitness converging as the iterations progressed. Moreover, by evaluating the total distance and time required for the ground maintenance process of an armed helicopter, it was demonstrated that the individual fitness function in the GA is a standard method for evaluating layout efficiency. Based on these findings, it could be concluded that the improved GA is practical for optimizing layouts containing central obstacles and facilities.