System Dynamics Based Microscopic Freight Transport Simulation for Urban Areas – CEP Market Analyses –



System Dynamics Based Microscopic Freight Transport Simulation for Urban Areas – CEP Market Analyses –

Nominated for The Neil Mansfield Award

Authors

Carina Thaller, TU Dortmund University, Institute of Transport Logistics, Uwe Clausen, TU Dortmund University, Institute of Transport Logistics

Description

A System Dynamics based microscopic freight transport simulation is presented which enables an investigation of freight demand as well as the resulting freight transport demand at an urban level by means of long-term and point-in-time forecasts.

Abstract

Agglomerations and metropolitan regions are social and economic attractions, which have been growing in the last decades and will continue to rise in importance as human habitats and economic poles in the future. Further trends are the increasing share of e-commerce demands of private households as well as the reduction of storage capacity in the retail sector. These tendencies lead to a significant growth of freight transport demand in urban areas. Especially courier-, express- and package service providers (CEP services) benefit from the trends mentioned above. But the supply of freight demand requires high-frequency small-scaled deliveries and a growing share of trips of light-duty commercial vehicles. In consequence, the growing share of these small commercial vehicles has negative effects on environment on the one hand and on transport flows resulting from the frequent stops of these vehicles on the other hand. Due to these issues regarding the continuously growing freight transport demand of CEP services in urban areas, efficient solutions in transport planning and city logistics have to be identified to regulate this freight transport demand in a sustainable way.
For this reason, a System Dynamics (SD) based microscopic freight transport simulation is developed to examine the freight demand of private households and retailers and the resulting freight transport demand of CEP services as well as to identify efficient solutions to regulate this freight transport demand. This approach mutually uses the forecast capability of SD and the detailed resolution of microscopic transport simulation – in this case MATSim. By this methodological approach the degree of explanation and the accuracy of forecasting as well as the quality and informative value of the impact analysis can be improved and enhanced by conducting point-in-time and long-term forecasts. This appraoch is tested in a framework of a case study, which concentrates on Berlin, Germany.
Hence, this approach enables to investigate urban freight demand and freight transport demand in the course of time on the one hand and to analyse the impacts of the freight transport demand at infrastructural level on the other hand. SD long-term forecasts are carried out to show future developments in consumer behaviour of private households and retailers as well as the resulting freight transport demand of CEP services. In dependency of freight demand of clients, fleet size and transport capacity development the number of transports, road mileage for last mile, fuel consumption, transport costs and GHG emissions are calculated in the course of time. The outcome of the SD trend analyses are transferred to MATSim and the freight extension Jsprit by an adapter developed to adapt the synthetic population of the freight demand (e.g. private households, retailers) and to generate the freight transport demand - in this case carrier population of CEP service providers. By Jsprit, carriers of CEP service providers are generated, which execute their daily plans and deliver their packages to private households and retailers by MATSim. The spatially influenced parameters (e.g. number of trips, transport distance, transport lead time, transport costs, GHG emissions polluted) are derived by MATSim and transferred to the SD model in a feedback loop by the adapter to adjust the freight transport demand development for further SD trend analyses.
By this approach analyses are carried out to examine the development of freight demand and freight transport demand by long-term forecasts and to investigate their impacts by point-in-time forecasts. Within the SD modelling procedure the relevant submodules of the system urban freight transport are developed. The SD model describes the submodules in a transparent form (stock-flow chart) and works with high aggregated data. MATSim dynamically simulates the behaviour of individual agents within a day. Thus, the two instruments enhance and complement each other. By the linkage between these two methodological instruments stable systems are generated at an aggregated and a disaggregated level.

Publisher

Association for European Transport