<![CDATA[Workstation imbalance can constrain assembly throughput even when spare capacity exists elsewhere in the production system. This study evaluated workstation and operator allocation alternatives for selected motorcycle assembly operations at PT Astra Honda Motor’s Karawang Plant, Indonesia. A survey-informed discrete-event simulation framework used operational estimates from 120 purposively selected employees collected through Google Forms during a September 2025–February 2026 study period. Median minimum, most-likely, and maximum processing times parameterized triangular distributions for ten analytical workstations. Seven configurations were compared using 50 replications of an eight-hour terminating simulation. Baseline throughput averaged 408.42 units per simulated shift (95% confidence interval 406.78–410.06). Fastening and adjustment (WS6) accumulated the largest mean queue, at 35.75 units, and the longest station waiting time, at 2,023.37 seconds. Reallocating capacity from component preparation and final release to WS6 and functional testing (WS8) produced the highest mean throughput, 478.76 units (477.66–479.86), representing a 17.22% improvement while retaining 12 modeled resource units. Mean completed-unit flow time decreased by 45.79%. The integrated alternative produced comparable throughput, 478.62 units, but used 13 resource units. Variability reduction alone increased throughput by only 0.39%. These findings support evaluating coordinated reallocation before adding capacity at a single station. The results are conditional simulation estimates, not observed post-implementation improvements; production-log validation, resource feasibility assessment, and fuller implementation documentation are required before operational deployment.]]>
Copyrights © 2026
