COMBINED LOGIT MODEL OF DESTINATION AND MODE CHOICE FOR LEISURE TRAVEL

Abstract

Transportation planning is a science that seeks to study the problems that arise in providing transportation facilities and to prepare a systematic basis for the planning of such facilities. Travel behaviour study is a vital link in overall transportation planning and involves understanding of the decision making behaviour of travellers with respect to choice of destination value to be visited, mode of travel, as well as, the time of travel. However, unlike work and education related trips, such analysis in leisure travel context is challenging on account of the vast heterogeneity in the preferences of tourists as well as the variability in destination value and mode selection criteria. Out of the varied destinations values, the trip maker takes decision while considering the budget, time available, previous visits, personal or house hold characteristics, travel mode and trip characteristics, etc. This work is an attempt to study the travel behaviour . of domestic tourists regarding their choice of travel mode and destination value. The tourists may have to consider several influencing attributes before an alternative is chosen. Behavioural aspects influencing the mode and destination value choices were also incorporated. Data collected through on-site personal interviews at selected tourist destinations from having varied values such as historical heritage (Delhi and Hyderabad), religious (Tirupati and Haridwar) and waterfronts (Goa and Kochi) were used for the analysis and building of the models. The models were based on the utility maximization hypothesis which assumes that an individual's choice is a reflection of underlying preferences for each of the available alternatives and that individual selects the alternative with the highest preference or utility. The Nested Logit (NL) model structure was used to model the decision making patterns of choices, as it avoids the lIA property of the multinomial logit models (MNL). Different forms of nested logit structures, say destination value at upper level and mode choice at lower level were formulated and the best fit structure was found among them. The variables considered for model development include household characteristics such as household income, vehicle ownership, and number of aged or children in the family; trip characteristics such as travel time, travel cost; behavioural intents based on importance rating for identified attributes such as social aspects, destination values, travel and miscellaneous aspects. Behavioural aspects were also included in the form of latent factors influencing destination choices, from factor analysis. Combined logit model of destination value and mode choice model were developed for the leisure value. The results would be helpful in understanding and predicting the individual decisions so as to forecast the leisure travel demand. This would also aid in improving the " destination attributes and modes of transport, there by mitigating traffic congestion and will aid in optimum allocation of resources. Key words: Mode choice, Destination Value, Nested Logit, Leisure travel and India iv