Metal manufacturing SMEs under Make-to-Order systems, such as CV. XYZ in Surabaya, often face a critical problem: excessive tardiness and overtime costs caused by an unstructured scheduling approach in a Flexible Job Shop environment. This study addresses this problem by developing and comparing two heuristic scheduling algorithms, the Active Schedule Algorithm with the Most Work Remaining (MWKR) rule and the Non-Delay Algorithm with the Shortest Processing Time (SPT) rule, aiming to minimize makespan, tardiness, and overtime cost. Historical order data from December 2025 to February 2026, covering 13 jobs across 7 workstations, were collected and validated through stopwatch time studies. Performance was evaluated using makespan, tardiness, and overtime cost, benchmarked against the company's existing scheduling condition. Active Schedule-MWKR achieved a makespan of 26,022 minutes, a 27.42% reduction from the 35,880-minute baseline, completely eliminating tardiness and overtime. Non-Delay-SPT yielded only marginal improvement (11.26%), retaining 6.80 days of tardiness and IDR 999,600 in overtime costs. Compared to actual company conditions, Active Schedule-MWKR delivered IDR 2,415,000 in cost savings with full overtime elimination. These findings establish Active Schedule-MWKR as the recommended scheduling policy for Flexible Job Shop environments with heterogeneous job types, offering a practical and replicable framework for Indonesian manufacturing SMEs.
Copyrights © 2026