Managing inventory levels and time to market in assembly supply chains by swarm intelligence algorithms

The proposed work addresses the problem of placing safety stock under the guaranteed-service model when a set of supplying, manufacturing and delivery stages model the production system. Every stage has a set of options that can perform the stage and every option has an associated cost and time. Hence, the problem is to select an option per stage that minimises the safety stock and lead time at the same time. We proposed solving the problem using two swarm intelligent meta-heuristics, Ant Colony and Intelligent Water Drop, because of their results in solving NP-hard problems such as the safety stock problem. In our proposed algorithm, swarms are created and each one selects an option per stage with its safety stock and lead time. After that, the Pareto Optimality Criterion is applied to all the configurations to compute a Pareto front. A real-life logistic network of the automotive industry is solved using our proposed algorithm. Finally, we provided some multi-objective performance metrics to assess the performance of our approach and carried out a statistical analysis to support our conclusions.