Usulan Penjadwalan Produksi untuk Meminimasi Makespan dengan Menggunakan Metode Simulated Annealing pada PT XYZ

Abstract

PT XYZ is a furniture manufacturing company that implements a flow shop production system with a make-to-order approach, where production is based on customer demand. In practice, the company often experiences delays in fulfilling orders due to the use of the First Come First Serve (FCFS) scheduling method, which has proven insufficient, particularly in October and December 2024, and January and February 2025. This study aims to propose improvements to the production scheduling system by applying the Simulated Annealing (SA) algorithm to minimize total job completion time (makespan). This method is chosen for its ability to explore a wide solution space and avoid local optima traps. Data were collected through direct observation, interviews, and historical company documentation, focusing on the production of tables and racks. The actual scheduling resulted in a makespan of 128.21 hours, while scheduling using the SA algorithm reduced the makespan to 107.32 hours. This reflects an efficiency improvement of 16.29% and a scheduling performance increase with an Efficiency Index of 1.19. Thus, the Simulated Annealing algorithm is considered effective for optimizing production scheduling and is recommended as an alternative scheduling method for PT XYZ.