LWP - Diurnal Variation Forecast of Truck Waiting Times at Logistical Nodes

In the project a prototype for predicting the diurnal variation forecast of truck waiting times at logistical nodes was developed - including a pre-arrival notification system.

Project duration01.01.2014 - 31.12.2016
Project funding
Industrial Collective Research project funded by Federal Ministry for Economic Affairs and Energy
Our status
Responsible AiF reseach institute
Contact personMarvin Kastner
Project partnersCompanies accompanying the project:
  • Addicks & Kreye Container Logistik GmbH & Co. KG
  • Eduard Meyer GmbH & Co. KG Spedition
  • E.L.V.I.S. Europäischer Ladungs-Verbund Internationaler Spediteure Aktiengesellschaft
  • Hamburger Container- und Chassis-Reparatur-Gesellschaft mbH (HCCR)
  • Hamburg Port Authority AöR (HPA)
  • Oetjen Logistik GmbH
  • Sächsische Binnenhäfen Oberelbe GmbH (SBO)
  • SGKV - Studiengesellschaft für den Kombinierten Verkehr e.V.
  • Stapelfeldt Transport GmbH
  • Survey Compass GmbH
  • SWS Seehafen Stralsund GmbH


Load peaks in the truck handling of logistic hubs lead to highly fluctuating waiting times for arriving vehicles and prevent an optimal disposition of the trucks at the forwarding companies. This causes higher costs and also leads to a higher demand for equipment and personnel on the part of the logistics nodes concerned. This is associated with economic losses. Factors influencing the waiting time include the use of resources at the respective logistics node, traffic connections as well as individual local events (construction sites, weather, system failures, etc.).

Therefore there is a need for a simple and cost-effective solution that effectively supports both truck scheduling and organisation. In order to achieve this goal, the project pursued a four-part research approach with a corresponding number of work packages (AP): The forecast model for valid prediction of truck waiting times at logistical nodes is to be based on a two-stage procedure. In the first stage, classic forecasting methods based on historical data are used to produce a "basic forecast". In the second stage, concepts of artificial intelligence are used to improve this basic forecast. In this way, stochastic daily events (e.g. disturbances in the processing of the node) can be integrated into the forecast. In addition, voluntary truck registrations in advance for the next few hours are taken into account as an indicator of the estimated number of arriving vehicles.

Parallel to the development of the forecast model, a generic (IT-based) system solution is being designed, which needs to ensure the demand-oriented provision and storage of generated forecast data as well as collected operating data during implementation. The use of (standard) software components ensures that the IT solution created is manageable in terms of the expected investment and operating costs and can be transferred cost-effectively to various types of logistical nodes. In order to increase the attractiveness of voluntary pre-registration at the junction, possibilities of preferential truck treatment are also being investigated which support or provide incentives for appropriate behaviour on the part of freight transporting companies and truck drivers. These so-called preferential options are also examined with regard to their economic feasibility and the pre-registration behavior associated with them.

Finally, the determined research results are implemented as a demonstrator solution at a logistics node in order to test them on site in everyday operation concerning their effects on waiting time development and resource efficiency.

Presentations (excerpt)

  • Hill, Alessandro and Böse, Jürgen W. (2017). A decision support system for improved resource planning and truck routing at logistic nodes. Information Technology and Management. 18 (3), 241-251. [doi] [BibTex]

  • Hill, Alessandro and Böse, Jürgen W. and Meissner, Finn (2015). Forecasting-Based Truck Wait Time and Peak Workload Reduction at Logistic Nodes. [BibTex]

  • Hill, Alessandro and Meissner, Finn and Böse, Jürgen W. (2016). Umschlagprognose für maritime Leercontainerdepots mit Hilfe künstlicher neuronaler Netze. [BibTex]

  • Stamer, Martin and Hill, Alessandro and Böse, Jürgen W. and Jahn, Carlos and Krick, Ronald (2016). Ein generisches IT-Konzept zur Entscheidungsunterstützung an logistischen Knoten durch die Prognose von Lkw-Wartezeiten. 125-131. [BibTex]

The Industrial Collective Research project "Diurnal variation forecast of truck waiting times at logistical hubs" (17694 N) of the research association Bundesvereinigung Logistik e.V. was funded by

in the scope of the Industrial Collective Research project of the

on the basis of a resolution of the German Bundestag.