mFUND Hegel Transport LogisticsHegel
Harnessing the potential of hypernetworks in freight transport and logistics
- The project aims to develop new solutions relating to the design and visualisation of linked transport networks (known as hyper-networks) in logistics.
- To achieve this, simulation and optimisation methods are being developed. The focus is on machine learning methods, especially in the field of deep reinforcement learning.
The market for freight transport services is not very transparent. This applies in particular to the interconnection of modes of transport such as road, rail and inland waterways. This lack of clarity leads to inefficiencies and makes it difficult to switch to more eco-friendly transportation solutions.
There is a lack of data bases and modelling methods for the representation and design of linked transport networks (known as hyper-networks). To obtain an overview of linked transport networks, data from various sources is combined. The focus here is on reducing the need for the companies involved to collect new data, and instead highlighting the benefits of existing data. The DLR MovingLab will be used to collect as yet unavailable data. It enables the simultaneous recording of spatiotemporal travel data for a journey and questions to be answered regarding the activities associated with the journey. In the case of goods transport journeys, for example, this includes information on the cargo and the loading and unloading processes at the individual stops that the vehicle makes along the way.
Simulation and optimisation methods are used to facilitate the design of hypernetworks in freight transport. This involves the use of machine learning methods, especially from the field of deep reinforcement learning. These methods make it possible to evaluate different network designs in advance and simplify the selection of suitable network options. The resulting models are used with the acquired data to develop intermodal transport options for promising segments of demand.
For this purpose, basic research into the modelling of hypernetworks is combined with the design of innovative services in the logistics industry, which consists largely of small and micro providers.
- Kühne Logistics University, Hamburg
- 4PL Intermodal GmbH, Rotenburg (Wümme)
- Ubilabs GmbH, Hamburg
- Fraunhofer‐Arbeitsgruppe für Supply Chain Services, Nürnberg
