Modelling carrier type and vehicle type choice of small and medium size firms

Abstract

The objective of this paper is to study freight transportation outsourcing and vehicle type choice decisions made by small and medium size firms. Carrier type and vehicle type choices are modelled independently using multinomial logit models and jointly using nested logit models. Mixed logit models are also developed for the defined structures to study the choice correlation among the choices made by the same firm, resulting in development of six different models. A commercial travel survey for the Greater Toronto and Hamilton Area is used to estimate the models. The models consider characteristics of the firm (employment, industry classification) and shipment (location, type). The results show significant improvement for the carrier type choice model using a mixing distribution while the vehicle type model performs slightly better using mixed logit model than using multinomial logit model structure. The nested logit model results show that firms are more likely to substitute vehicle type choices before substituting carrier type choices. The correlation of vehicle type choices is found to be highest (96%) among own-account carriers and driver-own carriers (90%) and least for the vehicles owned by a for-hire company (82%). Similar to the mixed vehicle type choice model, the mixed nested logit model also shows a slight improvement in model performance. Based on the strong correlation indicated by the nested logit model, a model reflecting the joint choice of carrier type and vehicle type is suggested.