Abstract
Traditional approaches to a location allocation problem have focused on the allocation of customers to a distribution center (DC) according to some arbitrary geographical boundaries (e.g., administrative zones and census districts), which usually incurs underuse or overcrowding of these centers. Location allocation with balancing requirements (e.g., balanced workload of service among DCs) has therefore been addressed. A distribution strategy with balanced-workload allocation aims to be cost-efficient and to improve customer service. A novel bilevel programming model is presented that minimizes the cost of the total distribution network and at the same time balances the workload of each DC for the delivery of products to its customers. A genetic algorithm-based approach was developed to cope with the bilevel model, and it was tested on a best realistic data set. In addition to the most cost-efficient design, the bilevel programming model presents a picture to decision makers that shows the trade-off between the objective of cost minimization and the balancing requirements. It is also shown that the bilevel model offers a flexible framework that allows the incorporation of more requirements and constraints if necessary.
Get full access to this article
View all access options for this article.