Development of an Improved Decision Support Tool for Freight Transport Planning in Norway

Development of an Improved Decision Support Tool for Freight Transport Planning in Norway


Inger Beate Hovi, Institute Of Transport Ecoonomics, Stein Erik Grønland, BI Norwegian Business School, Anne Madslien, Institute Of Transport Economics


This paper will focus on the development of an improved decision support tool for freight transport planning in Norway. The improved model is operational and provides a reasonably good match with reality (statistics).


Since 2005 a Logistics model has been developed for freight transport in Norway. The last years this model has been improved, where the main improvements are related to the following elements:
1. A method for better commodity flow representation in the model.
2. Differentiated terminal costs, representing different services and technologies offered by the terminals, as well as regional variations.

The last years, freight models have attracted substantial attention at an international level. As far as we know there are very few examples of national freight models at detailed zone level where mode choice is based on logistics elements, such as the determination of shipment size and the use of consolidation and distribution centres. One of the first applications of a logistics chain freight model was the SMILE model (Tavasszy et al., 1998).

A study done by Rand Europe and Transek (deJong et al., 2002) summarized the frontiers of knowledge of how transport modelling structure from passenger transport models fruitfully can be applied to freight transport models. Based on this, SAMGODS in Sweden and the NTP-group in Norway engaged Rand Europe (now Significance) and national research companies to establish new national freight models for the two countries, The Logistics models in Norway and Sweden (deJong et al., 2005 & 2008). The models can be described as aggregate-disaggregate-aggregate (ADA) model systems. In the ADA model system, the production to consumption (PC) flows and the network model are specified at an aggregate municipality level for reasons of data availability. Between these two aggregate components there is a logistics model that explains the choice of shipment size and transport chain, including mode choice for each leg of the transport chain. This logistics model is a disaggregate model at company level, which is the decision making unit in freight transport.

The research tasks that will be presented can be divided into two components:
Improved representation of commodity flows
The PWC flows between the production (wholesale) locations P (W) and the consumption (wholesale) locations C (W) are given in tons by commodity type. The consumption locations refer to both producers processing raw materials and semi-finished goods and to retailers. In lack of better data, the main commodity flow matrices in the previous model were based on economic data for production and consumption, converted from values to volumes, and where delivery patterns where computed by using advanced gravity models. In 2009 Statistics Norway carried out a commodity flow survey (CFS), where domestic commodity flows in Norway were mapped. In the project a method to utilize information from the CFS, statistics for different industries and the foreign trade statistics where developed to obtain more concise representation of the national commodity flows in the model. Data from the CFS were supplemented with other statistics from Statistics Norway, while mode specific statistics where used for validation of freight flows by mode at origin-destination (OD) level.

Terminal cost functions
There are different types of terminals implemented in the model; consolidation and distribution centres, ports, airports and railway terminals. In the previous model, there were implemented cost models with average costs only depending on transport modes and vehicles, and cargo groups. The enhanced functionality implies that we develop terminal cost models, where the costs are a function of: Terminal type, size and technology in the terminals, cargo handling system, and cargo types handled.

deJong, Ben-Akiva, Baak, 2008. Method Report - Logistics Model in the Norwegian Freight Model System. Deliverable 6A. Significance, Den Haag.
deJong, Grønland, Ben-Akiva and Florian, 2005: Specifications of a logistics model for Norway and Sweden. European Transport Conference 2005, Strasbourg
deJong, Gunn, Walker, 2004. National and international freight transport models: overview and ideas for further development. Transport Reviews 24 (1), 103–124.
Tavasszy, Smeenk and Ruijgrok, 1998. A DSS for modelling Logistic Chains in Freight Transport policy analysis. Intelligent Transportation Operational Research, Vol. 5, No. 6, 1998, pp. 447-459.


Association for European Transport