PERANCANGAN PENUGASAN OPERATOR PADA LINTAS PERAKITAN PRODUK EXCAVA 200 DIVISI ALAT BERAT PT PINDAD MENGGUNAKAN ALGORITMA GENETIKA

• Main Author: Al Fattah, Viqri
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/77712
• Institution: Institut Teknologi Bandung

PT Pindad is one of the State-Owned Enterprises (BUMN) that focuses on producing products that support national defense, security, and industrial equipment. One of PT Pindad's products is an excavator called Excava 200. The production performance of the Heavy Equipment Division has been deemed unsatisfactory, especially for the Standard and Long Arm types of Excava 200 products. This is due to the failure to meet production targets for the past few years, which were set at 100 units for the Standard type and 50 units for the Long Arm type. Currently, the assembly line can only produce 75 units of the Standard type and 33 units of the Long Arm type. This issue arises because there is no proper operator work allocation method, coupled with the increasing variety of products. Therefore, it is necessary to design operator assignments for the assembly line of Excava 200 products for both types. This research begins with the development of a reference model for the analytical method. Subsequently, a metaheuristic method is developed due to the complexity of the problem. The metaheuristic method used is the genetic algorithm. This algorithm can solve problems with a broad solution search space for complex problems. The genetic algorithm starts with generating an initial population using a construction algorithm, followed by tournament selection, crossover, mutation, and elitism. The objective function used in this study is the assembly line cycle time. The genetic algorithm is capable of producing viable solutions for the actual problem in the research, resulting in an assembly line cycle time of 123.96 minutes. The proposed production capacity for the assembly line is 436 units per year for Excava 200 Standard products and 290 units per year for Excava 200 Long Arm products. The proposed operator assignments for the assembly line consist of 13 workstations with 31 operators. The cycle time generated using the genetic algorithm is smaller than that obtained through the analytical method due to the computation time constraint of 168 hours. The parameters that influence the minimization of cycle time in the actual research data are the number of generations (G), population size (P), crossover probability (Pc), and mutation probability (Pm).