Existing models of commercial traffic are not sufficiently sensitive against many policy measures. In fact, there was a similar situation for passenger transport modelling until about a decade ago. Here, activity-based demand modeling, in conjunction with dynamic traffic assignment has made important steps towards addressing these issues. This is not the case for freight transport, since freight actors are not endowed with sufficiently reactive behaviour in these simulation studies. It is, however, to be expected that in most cases upstream logistics decisions are much more flexible than their passenger transport counterparts. Thus, freight traffic is one of the most promising parts of traffic in urban areas in terms of mitigating negative impacts without having to accept unnecessary limitations to the accustomed levels of service.
The behaviour of freight transport agents is connected with complex optimisation tasks. Multi agent (MA) modelling addresses the challenge to map the behaviour and the interactions of agents pursuing their goals and possessing solution strategies for complex planning problems. These are exactly those characteristics which are needed to analyse the effects of policy measures on various types of logistics agents.
The project extends an existing MA framework for freight transport by modelling the demand side of the transport market, i.e. shippers and recipients as well as logistics service providers, in a detailed way. It will probably be the only existing freight demand model that includes several decision dimensions, makes use of complex optimisation, interacts with time-dependent traffic flows, and is available as an open source program that has been tested by various other researchers.
11/2017 – 10/2020