PENGEMBANGAN METODE PENJADWALAN PRODUKSI PADA AREA SHEET METAL DI PT DIRGANTARA INDONESIA DENGAN ALGORITMA SIMULATED ANNEALING

• Main Author: Alexandra, Natasya
• Format: Final Project
• Language: Indonesia
• Subjects: Teknik (Rekayasa, enjinering dan kegiatan berkaitan)
• Online Access: https://digilib.itb.ac.id/gdl/view/74497
• Institution: Institut Teknologi Bandung

PT Dirgantara Indonesia (PTDI) is one of the state-owned enterprises engaged in the manufacturing industry of aircraft. One of the areas in PTDI is the sheet metal area which has high workload. In 2022, there were 92 parts in this area that couldn't be completed on time. The sheet metal area is divided into 5 cells, and one of them is cell 2, which has the largest contribution in the sheet metal area. About 70% of the parts that enter the sheet metal area are processed in cell 2. Therefore, minimizing the makespan in cell 2 can have a significant impact on reducing bottleneck and production delays. One way to minimize the makespan is by developing scheduling methods in cell 2. Simulated annealing algorithm (SA) is used to develop the scheduling method in cell 2. SA algorithm consists of two parts: the outer loop algorithm that used for running the general simulated annealing algorithm and the inner loop algorithm that used for generating new solutions and performing assignment procedures. The development of the simulated annealing algorithm is translated into a Python program to obtain the desired result, which is the production scheduling that minimizes the makespan. The algorithm is experimented using the full factorial design method to evaluate the makespan values. Based on the experimental results, significant factors are identified, including the final temperature (????????), cooling rate (????), and the number of iterations per temperature (????). Furthermore, recommendations were obtained for significant factor values that can be used to achieve minimum makespan, namely ???????? with a value of 1, ???? with a value of 0.95, and ???? with a value of 20. Based on the computational results using the recommended factors the makespan was improved by 24.61% compared to the priority scheduling method.