Establishing an order allocation decision support system via learning curve model for apparel logistics

Apparel firms confront the problem with effective order allocation among factories when conducting global logistics operations. The problem is non-deterministic polynomial-time hard and often complicated to obtain an optimal solution. This research tried to solve, by heuristic method, the problem via applying the learning curve to the capacity planning to improve the decision-making on global logistics operations. It designed a two-stage order allocation decision support system. This two-stage model includes: (1) selecting and sorting factories and (2) dynamic learning of order allocation. The results demonstrated that the system could develop the most suitable order allocation model among multiple factories and minimize the order allocation cost. This research further provided a case study to help verify the designed system applied in the global logistics operations of apparel industry. The proposed system seems to be more conforming to the real-world production and could offer a better suggestion concerning order allocation.