A port choice model with logit models: a case study for the Spanish container trade

Earlier research by the authors showed that transport costs, port specific dummy variables and proxies for quality of service aspects have a statistically significant impact on port choice. The objective of this paper is to improve the earlier models through the use of a nested logit model and further by including new variables. The results show that the nested model is an improvement, that the variable indicating the inland transport balance has a statistically significant impact and that the split of variables into distance dependant and independent components gives no improvement. The model outcomes can be used to assess the impact of interventions in ports, and thereby to serve as a basis for the economic and financial evaluation of port projects.