Destination choice decisions of retail travellers: results from discrete choice modelling in Brisbane

Abstract

Retail trips account for approximately 20 precent of all trips in Australian cities and are extremely car dominated. Decisions about retail trip-making are driven by a range of factors such as the attractiveness of shopping destinations (their size, composition, level of service etc.); their accessibility (including travel costs, parking, public transport, walking and cycling access); traveller characteristics; and the nature of the shopping trip itself. Many retail trips are part of trip chains including trips for other purposes, such as shopping trips made on the way home from work, which further affects destination and mode choice decisions. The flexible nature of retail trips in comparison to other major trip types makes it challenging to understand how decision makers' preferences, destination attributes and accessibility affect destination choice. This paper applies discrete choice modelling, using multinominal logit-based models, to explore how the attributes of individual travellers and destinations choice alternatives influence travellers' choice of shopping destination. The model uses the 2009 SEQ Household Travel Survey dataset from the Queensland Department of Transport and Main Roads. Shopping centre locations and characteristics for greater Brisbane were geocoded and assigned using secondary data. The model's results are based on the level of utility associated with each choice for different types of shopping trip including groceries, clothes, eating or drinking, alcoholic drinks and household goods. Seven explanatory variables have been included in the model: car distance, number of retail jobs at site, site area and the existence of food court and socio-demographic characteristics of travellers, namely age, gender and income. The results show significant relationships between site attributes and socio-demographic characteristics for most of these trip types. The model results are the first step towards calculating the probability that particular types of customers will choose particular shopping destinations for their retail trips. This knowledge will help transport and land use planners to better target interventions, including possible retail planning strategies, to encourage more sustainable travel behaviour in South-East Queensland - a rapidly expanding conurbation.