<![CDATA[General background: Efficient production scheduling is essential for improving operational performance in multi-stage manufacturing systems with fluctuating demand. Specific background: At PT XYZ, Cement Bag production involves six sequential machines, yet scheduling remains manual, causing bottlenecks, long waiting times, and a makespan that exceeds production targets. Knowledge gap: Prior studies largely optimize a single performance indicator—typically makespan—and rarely address dual objectives in complex multi-machine plastic-bag manufacturing. Aims: This study aims to optimize Cement Bag production scheduling using the Particle Swarm Optimization (PSO) algorithm to minimize makespan and total waiting time simultaneously. Results: Implementing PSO on 12 jobs and 6 machines reduced makespan from 49,400 seconds to 34,520 seconds (32.77%) and lowered waiting time from 186,050 seconds to 115,870 seconds. The optimized job sequence balances machine workloads more effectively than the existing manual schedule. Novelty: The study integrates dual performance criteria in a real multi-process Cement Bag production system and applies PSO to an industrial context not previously examined comprehensively. Implications: Findings demonstrate PSO’s suitability as an adaptive AI-based scheduling approach, offering manufacturers a practical pathway toward improved resource utilization, reduced delays, and enhanced responsiveness to market variability. Highlights: Highlights the significant reduction of makespan and waiting time using PSO. Demonstrates balanced workload distribution across six machines. Shows the novelty of dual-objective optimization in Cement Bag production. Keywords: Production Scheduling, PSO, Makespan, Waiting Time, Manufacturing]]>
