Freight Transport Models

Globalization, economic development, and new information technologies have been spurring consumption and trade among nations. Available forecasts predict significant growth in freight demand [Page 600]in many regions worldwide in the coming decades. Such growth is likely to generate significant impact on the current transportation systems (in terms of congestion and accidents) and the environment (in terms of increase of emissions of greenhouse gases and depletion of natural resources). There is therefore the need to increase the efficiency and promote the sustainability of freight transport systems and to mitigate the conflicts with other users (including private cars, bikers, and even pedestrians).

The identification of adequate solutions often requires field experimentation of new concepts and multiple alternatives. However, field experimentations in transportation require a significant amount of resources (human, equipment, capital, etc.) without any real assurance of return, besides inevitably introducing disturbance or disruption into the transport systems. It is at this moment that models can be quite useful. Models are cost-efficient alternatives to field tests and provide fully controllable environments for testing alternative hypotheses and scenarios.

A model is a simplified representation of a real-world situation (although it can also be used to study imaginary conditions). It contains the minimum amount of detail (in terms of variables, features, or detail) to adequately permit the study of the problem. The simplification is required for reducing the complexity of the real world to manageable levels for an efficient development of the model. However, simplification also denotes that particular aspects of the real world are not considered in the analyses. Therefore, the outcomes of a model are by definition inaccurate or, in other words, have a degree of uncertainty, which tends to increase with the level of simplification. Even so, models have proven very useful to forecast demand for transport services and to estimate the impact of changes in the transportation system.

Passenger transport models are more established and developed than freight transport models. Understandably, the primary political and societal interest was on improving the transport of people, which eventually drove academic endeavor. Indeed, the initial freight transport models were inspired by the passenger models. Yet people and freight exhibit significantly different transport patterns and behaviors, and before long, the need for tailor-made models was recognized.

Freight transport models have been applied to a wide variety of geographical areas, ranging from the city district level up to the continental level. The size of the geographical area is relevant in the modeling process because it influences the zoning. Zoning is the process of dividing the geographical area of study into a set of smaller regions—zones. The modeling parameters are considered homogenous within each zone. They are normally equal to the average value of some of the individual elements of each zone. Zoning thus reduces the inherent complexity of the real world to a manageable level. As it introduces errors in the analysis, the number and shape of the zones should be carefully considered.

The European Union–cofunded research project SPOTLIGHTS-TN identified a total of 65 freight transport models and 29 joint passenger and freight transport models in a universe of 222 transport models. Compared with the passenger side, freight transport modeling lags behind considerably. Even so, substantial progress has been achieved and there is now a rich assortment of models. Typically, classification is done according to their primary purpose, as discussed below.

...