Designing an auto-adaptive inventory replenishment system

In a periodic-review inventory replenishment system, two key questions are when to order (at which level of inventory) and how much to order (order up to which level). A periodic-review replenishment policy requires these two parameters (s,S)  be specified for each stock keeping unit (SKU). These parameters should be estimated based on the demand pattern of the corresponding SKU. However, in a volatile business environment a certain parameter set can be outdated very frequently. In that respect, there is a need for an auto-adaptive replenishment system, which can monitor the demand pattern of the SKU and update (“correct”) its key parameters accordingly on continuous basis.In this project, the team is expected to work with the problem owner (i.e. a sanitary ware producer) in order to design and evaluate such a system on the real demand data of a set of SKUs. The performance of the proposed auto-adaptive system will be compared with the static case where a predetermined parameter set is kept constant throughout the analysis time window. 

Skills Required:

Status: 
Open