Decision Support for Location Routing with Relocation Aspects

Due to drops in orders and decreasing production rates, demand is declining seriously in many logistic sectors like e.g. air cargo, ship and less-than-truckload (LTL) transports. Hence, logistic companies are looking for new ways to improve their transportation network efficiency. On the operational level, costs depend mainly on planned delivery tours. In our current research project KolOptNet1 (Schwind and Kunkel 2009), we focus mainly on this aspect and develop an integrated software system which determines optimal delivery tours. The system exploits historical data as well as the demand information of the current day to calculate sorting plans for the packages and the delivery tours. The historical data is used to develop a prognosis function that analyzes changes in demand over a given period of time in order to forecast the future delivery volume of each geographical area. Moreover, our software automatically transfers sorting plans and vehicle routes to wireless mobile package delivery scanner devices which in return provide our system with collected data about the delivery process. In the long term, the location of a depot is a critical success factor because it strongly affects delivery costs. Therefore it is advisable to evaluate and improve depot locations before the route planning process. To this end, we integrate a relocation decision support system (RDSS) into our software. This tool analyzes the existing depot structure, i.e. location of depots and allocation of customers to depots, and checks if a relocation of existing depots would enhance the cost-efficiency of the delivery network. Besides this one-time determination of optimal depot locations, the relocation tool is coupled with our operational routing module to allow continuous verification of the network efficiency. As soon as changes in demand make modifications of the network structure reasonable, the system will suggest benefi-