International Choice Modelling Conference, International Choice Modelling Conference 2015

Font Size: 
The role of dynamic needs in tourists’ activity choice: design and results of a stated choice experiment
Theo Arentze, Astrid Kemperman, Petr Aksenov

Last modified: 11 May 2015


Advanced ICT and social media developments provide possibilities to give more personalized advice and recommendations to tourists. A typical user of a ‘smart routing’ recommender system is a tourist who is interested in exploring a city and wants to make a tour around. Such a tour comprises a scheduled list of attractions, museums, heritage sites, shops, parks, or other points of specific interest and the trips to travel from one point to the other. To recommend a program for a tour requires knowledge about various needs, preferences and expectations that the tourist may have regarding the content and duration of the tour. Existing choice models have an important limitation in this regard. Existing approaches generally assume that tourists’ preferences for activities depend exclusively on static attributes of the person and choice alternatives. This stands in contrast to a central notion of need-based theories of choice behavior which state that individuals’ motivations to engage in certain activities are at least partly dependent on needs that change depending on previous activities. Dynamic needs give rise to saturation effects and variety seeking. Process utility emerges from the satisfaction of the needs people have and it generates positive and negative substitution relationships between activities contributing significantly to the value of an overall tourist experience.

To obtain a better understanding of the role of dynamic needs, insight is needed in how tourists trade-off dynamic and static preference components. Decision components relate to the various attributes tourists take in consideration when they select points of interests. First, the general attraction value of the point of interest is relevant: the extent to which the point of interest is special, worth a special trip, or even the primary reason to visit the city. For example, in many travel guide books some kind of rating system is used to distinguish a top attraction from attractions of less importance. Second, the extent to which the point of interest matches a person’s interest in specific objects is a consideration. For example, some persons may be fascinated by cathedrals whereas others find them boring. Attraction value and personal interest relate to static preferences. In terms of dynamic needs, six dimensions have been found to play a role in leisure activity choice  (Nijland et al. 2010). These include needs for entertainment, relaxation, new experiences/-information, socializing, physical exercise, and being in the open air and green environment. A dynamic need determines what the person would like to do currently, given the state s/he is in. Finally, monetary costs, such as entrance fee, and effort to reach the location are relevant. A choice of an activity/point of interest generally requires a trade-off between all these considerations.

In this paper, we present the design and results of a stated choice experiment specifically designed to measure the relative impact of dynamic needs and variety seeking on tourists’ activity choice. In the experiment respondents are asked to plan the activities they like to participate in during a day trip to a hypothetical large city. They are assumed to travel with a friend/relative with similar interests. In the choice task, respondents choose an activity or point of interest from a set of optional activities. The activity options vary in terms of the considerations discussed above: the extent it matches static needs, the extent it fits current (dynamic) needs, the travel costs and entrance costs. The context description is varied across choice tasks to manipulate the current needs at the time of the choice. In each context description respondents are asked to imagine a situation where particular needs are activated as a consequence of earlier activities. For example, a typical context description could read as: “imagine a situation where, given the previous activities conducted, you currently have a [strong, moderate] need for [entertainment, relaxation ... ]”. Both the context and choice tasks are varied independently based on orthogonal designs. By varying the size and nature of the current needs in each choice task the relative importance of dynamic needs for each need dimension separately is identified.

A sample of individuals from a national representative on-line panel participated in the experiment and survey. To allow heterogeneity in taste preferences and identify classes of tourists, a latent class model is estimated based on the choice data. Class membership is explained based on background variables including a respondent’s  rating of importance he or she assigns to each need as a personal predisposition. The LATUS model recently proposed by Arentze (2014) offers a tool to generate optimal activity programs taking into account dynamic needs. In the paper we describe the design and results of the experiment and show how the parameters can be used as part of a personalized recommender system. The system supports tourists in their activity choices during a day tour in a city for the first time taking into account their dynamic needs.



Arentze, T.A. (2014) LATUS: A Dynamic Model for Leisure Activity-Travel Utility Simulation. Paper prepared for presentation at the 94th Transportation Research Board Annual Meeting, January 2015, Washington, D.C.

Nijland, L., T. Arentze, H. Timmermans (2010) Eliciting the Needs That Underlie Activity-Travel Patterns and Their Covariance Structure: Results of Multimethod Analyses, Journal Transportation Research Record, 2157 / 2010, 54-62.

Conference registration is required in order to view papers.