TASK-WORKER-STATION ASSIGNMENT AND SCHEDULING ON MULTIMANNED ASSEMBLY LINES

Abstract

The paper investigates the Multi-Manned Assembly Line Scheduling Problem (MALSP), where multiple workers operate concurrently at each station with the goal of improving the productivity of the line. The problem is approached by a mixed integer linear programming (MILP) model to capture the relations of precedence, station capabilities, and feasibility between workers and tasks. Since MILP becomes computationally expensive for medium and large-scale instances, Genetic Algorithm (GA) is proposed as a heuristic approach to obtain near optimal solutions with signifcantly shorter running time. Validation on small benchmark datasets shows that both the MILP and GA produce feasible schedules that satisfy all constraints, demonstrating the correctness of the formulations and operators. Further computational experiments indicate that while the MILP achieves optimality for small instances, the GA provides high-quality solutions and scalable performance for larger tests. The results underscore the effectiveness of combining exact optimization and metaheuristics to solve complex multi-manned assembly line scheduling problems aimed at minimizing makespan.