Conjoint analysis (rank data and logit modelling were used to assess the preferences of exporters of frozen and refrigerated cargo for alternative sea transport services; the attributes considered were: fare, travel time, frequency, shipping mode (container or refrigerated chamber) and type of service (normal or multimodal), and their variation range provided by specialists of the shipping company. These values were used in turn to create alternative transport services that were ranked by a sample of the main Chilean exporters of non-traditional (i.e. excluding fruit) refrigerated and frozen foods. The rankings and the attribute values were finally used to estimate logit models with linear-in-the parameters utility functions. The method worked extremely well in one case (a particular frozen product) where we found that fare level, preference for container, travel time and frequency, were in decreasing order the most important attributes; it was also found that the introduction of a multimodal service-at the prices indicated in the questionnaire was not considered an attractive option. However, the method did not work in the case of refrigerated vegetables; it is felt that one important reason for this is that the exporters of these products were too heterogeneous in terms of their choices and restrictions (for example the price variety of these products in international markets is enormous). The model may be used to determine, for example, the percent variation in travel time, frequency or a combination of both, that would be necessary to compensate for a variation in fares; another application would be to determine what fare increase could be implemented without changing the service attractiveness, if a container service was offered instead of the normal refrigerated chamber service.
Abstract