Optimized selection of logistical operating strategies based on simulation models using statistical methodology

Background

Freight forwarding facilities as logistical facilities for road freight transport are the transshipment points for national and international general cargo and groupage traffic and basically consist of two subsystems, the yard and the hall. The primary task in operating a facility consists of controlling the vehicles in the yard ("yard management"), loading and unloading the shipments at the gates and the transportation, sorting and buffering processes within the hall. A comprehensive consideration of all functional areas of a transshipment facility in a simulation model, i.e. the consideration of internal and external processes and areas for the optimized selection of logistical operating strategies of such a facility, has not yet been carried out from a scientific point of view and will be undertaken as part of the research project. A comprehensive comparison of operating strategies is made more difficult by the fact that the simulation of all permissible strategy bundles is not practicable for logistics facilities due to the high time requirement of the simulation and the large number of results to be interpreted. The development of a methodology for designing an experimental plan for such an extensive simulation, which enables the handling of the high number of potential strategy bundles, is therefore necessary in order to be able to adequately investigate complex logistical systems.

The aim of the research project is to develop a methodology for the model- and simulation-based identification of logistical operating strategies for complex logistics systems and to test this procedure using the example of a freight forwarding system. The use of the effective process time approach and the statistical modeling of system loads is planned for the analysis and modeling of forwarding systems. In addition, methods for processing real data as input data for simulation models of logistics systems will be developed.

Approach

In the research project, the two specialist areas of logistics and statistics are closely interlinked. In a first step, processes, system loads and operating strategies are analyzed and modeled. It is also examined whether a model consisting of system loads and structural data can be developed from the available real data of forwarding companies, which serves as a reference model for the simulation. Based on this, evaluation models are developed and the logistical questions are converted into hypotheses. These results form the basis for the simulation studies. A cycle of experimental design, simulation and recursive partitioning is applied, which leads to conclusions about the choice of optimal strategy bundles.

Intended results

A "meta-week model" is developed, with the help of which typical weekly processes (e.g. arrival times of trucks, shipment structure, volume fluctuations) can be generated and used for the simulation of the forwarding system. By using such a meta-week model, the influence of artifacts in real data on the simulation result should be reduced, e.g. the influence of undetected peculiarities in the data when using a randomly selected real week. A further result of the research project is the development of a methodology for complex logistics systems to identify logistics operating strategies or strategy bundles (several operating strategies from different areas).

Contact: Prof. Dr.-Ing. Uwe Clausen

Funding and partners