A bi-objective location-inventory model with capacitated transportation and lateral transshipments

Abstract

We integrate location and inventory decisions in a three-echelon distribution network model with multiple products, capacitated in-house fleet and proactive transshipments between depots. The problem was originally motivated by the real-life case of a leading company in consumers' household goods. The company offers a wide range of products which are different from demand pattern and shortage viewpoints. One of the product groups has seasonal trend with possibly lost sales during high seasons and backlog in low seasons, while the others lack any seasonal trend and get backlogged if any shortage occurs. A two-step process consisting of a new practical customer classification framework and a new mathematical model is developed. Total profit and customers' dissatisfaction in the form of a new service-level measure are considered as the objective functions. To study the effects of transportation capacity on the network's performance, a second scenario with inclusion of third-party transportation resources is presented. Due to unavailability or insufficiency of required objective data about production capacities and demands, they are assumed to be imprecise (i.e. possibilistic). After converting the original fuzzy model to its equivalent crisp formulation, an interactive solution procedure is applied to deal with the trade-off between the conflicting objectives. Useful managerial insights are also derived from the numerical experiments.

The New Generation of Operations Research Methods in Supply Chain Optimization: A Review

“…There is promising room for contributing to the literature of M-O LIP models, as only a few studies of this type can be found. Surprisingly, all of the existing M-O cases are by some means related to the optimization of customer service level [89][90][91][92]. This type also includes further relevant measures, such as the cost of lost sales by Aryanezhad et al [93] and delay by Javid et al [94,95].…”

“…In terms of stochasticity, Sadjadi et al [98] employ queuing theory to deal with uncertainty. Fuzzy approaches are also applied by several LIP papers for the same purpose [50,89,99]. Some other articles [23,29,65,91,93,94,100,101] used the advantage of the probability distributions to address stochasticity, and the remaining models are deterministic.…”

Solving a Novel Multi-Objective Inventory Location Problem by means of a Local Search Algorithm

“…Although less studied, MO-ILMs have also been proposed in the literature (see [1,3]). In [1] the authors design a distribution network model considering three-echelon and integrating location and inventory decisions.…”

“…In [1] the authors design a distribution network model considering three-echelon and integrating location and inventory decisions. The objectives considered in their model are the overall profit and the level of service as a measure of the customers' dissatisfaction.…”

“…Consequently, we denote the image of x ∈ X E in objective space as f (x)=ŷ , where ŷ is called nondominated point. The set of non-dominated points of the MO problem in (1) is Y N ⊆Y .…”