Elsevier

Tourism Management

Volume 24, Issue 1, February 2003, Pages 25-34
Tourism Management

Segmenting the market of West Australian senior tourists using an artificial neural network

https://doi.org/10.1016/S0261-5177(02)00050-XGet rights and content

Abstract

Measuring perceptions of customers can be a major problem for marketers of tourism and travel services. Much of the problem is to determine which attributes carry most weight in the purchasing decision. Older travellers weigh many travel features before making their travel decisions. This paper presents a descriptive analysis of neural network methodology and provides a research technique that assesses the weighting of different attributes and uses an unsupervised neural network model to describe a consumer-product relationship. The development of this rich class of models was inspired by the neural architecture of the human brain. These models mathematically emulate the neurophysical structure and decision making of the human brain, and, from a statistical perspective, are closely related to generalised linear models. Artificial neural networks or neural networks are, however, nonlinear and do not require the same restrictive assumptions about the relationship between the independent variables and dependent variables. Using neural networks is one way to determine what trade-offs older travellers make as they decide their travel plans. The sample of this study is from a syndicated data source of 200 valid cases from Western Australia. From senior groups, active learner, relaxed family body, careful participants and elementary vacation were identified and discussed.

Introduction

In Australia, the proportion of 65 year old and over has increased from fewer than 5 percent of the population at the beginning of the 20th century to more than 10 percent in the 1990s and is projected to grow to over 20 percent in the 2030s (Rowland, 1991). It is expected that this market segment will increase from 3 to 4 million in the next 10 years; this as opposed to the 15–44 age group that is expected to be stable in the next 50 years (ABS, 1998). The ageing of the population has resulted from increased medical knowledge, advanced technologies, healthier life style and improved sanitation, coupled with a lower or zero birth rate in nuclear families of the post modern era (ABS, 1990). With its propensity to travel, the ageing population is receiving more attention from the tourism industry and academics (Ananth, DeMicco, Moreo, & Howey, 1992; Environmetrics, 1991; Wei & Ruys, 1998). Seniors have the budget to travel: Australia's 2.97 million seniors spend $895 million on domestic travel annually and this is expected to grow to $2.3 billion by 2051 (Australian Bureau of Statistics (1994), Australian Bureau of Statistics (1998)). Seniors also have the time to travel; they have increased discretionary time that they spent on holiday travel (ABS, 1997).

While the economic importance of the senior market has now slowly being recognised, the specific tourism needs of the market have not been fully explored. Given the improved health and life expectancy of older people and the changes that each following cohort of seniors will display, it is important that the knowledge of those needs and expectations is regularly reviewed and updated.

An important segment in servicing the senior market is the intermediary function of the travel agent in the sales of the tourism product. The use of travel agents by customers has been discussed by several authors. For example, Meidan (1979) discussed the selection criteria of two groups of customers when purchasing a foreign package tour and discussed the differences in preferences between groups of respondents. Oppermann (1997) presented a comparative perspective of users and providers, focussing on the service attributes of travel agencies and their importance for potential customers. In this study, travel attributes include travel motivations and travel concerns.

Consumer perceptions of travel agencies was discussed by (Kendall & Booms, 1989). They found that travel agents needed to develop more sophisticated techniques to compete in today's market place. Richards (1995) studied how travel agents try to effectively provide their customers with information by bridging the information gap. He found that travel agents are increasingly being left behind by consumers who are able to obtain more specialised product knowledge.

Some research on senior tourism has focused on travel motivations and to the best of the authors’ knowledge, few have researched disincentives to travel. Such concerns include health-related, safety and financial issues. Looking at seniors’ travel motivations in conjunction with their concerns provides a better understanding of why and how seniors travel. This study focuses on segmenting the market of senior tourists by some selected travel attributes and demographics through the use of neural network techniques.

Past research on market segmentation in tourism has focused on travel motivations as criteria (Wei & Ruys, 1998). However, tourists, especially senior tourists, often have concerns for travel. These concerns often serve as disincentives for many seniors to get away from their home. In this study, the needs and wants of senior travellers in relation to travel attributes in terms of their travel motivations and concerns are included as criteria for segmentation.

In this study, a relatively new technique, neural networks, is employed for the segmentation of the senior market in tourism. The typical problem in marketing analysis is the complex interaction between individual variables, which can hardly be allowed for using normal linear statistical methods, and represents a particular problem in this respect. So the collection of customer data is associated with a considerable amount of respondent subjectivity. This leads to a situation where the basic structures and mechanisms of the studied customer relationship are overlaid with a partially significant fuzziness in the data. Furthermore, missing values are an unavoidable problem in data gathering. This can lie both in the reply behaviour of the respondents and the nature of the inquiry. That is, respondents are only asked about the quality elements which are within their own experience. Therefore, the model which is used must be particularly well suited for separating relevant structures from noisy data and an optimum model consequently works as independently as possible of missing values. In these perspectives, neural networks have shown to be particularly suitable for handling customer survey data with its typical features of nonlinearities, interactions, correlated variables as well as missing observations. Conventional methods often requires certain assumptions of the underlying variables. However, neural networks handle all variables without the requirement of certain distributional characteristics of the variables included. In addition neural network models do not require explicit evaluation of transfer coefficients, and need no model formulation.

Section snippets

Neural networks

Neural networks are massively parallel systems that rely on dense arrangements of interconnections and surprisingly simple processors. Artificial neural networks take their name from the networks of nerve cells in the brain. They originated in the artificial intelligence discipline where they are often portrayed as a brain in a computer. Neural networks are designed to incorporate key features of neurons in the brain and to process data in a manner analogous to the human brain. Although it is

Methodology

This study has defined senior travellers as people aged 50 and over. This is based on the eligibility criterion of the National Seniors Association, which is the largest organisation of its kind in Australia.

An independent marketing consultant was given the brief to carry out a marketing research project for a large travel organisation in West Australia and the authors identified senior-important travel agent attributes that were supplied to the marketing consultant. The consultant used the

Analysis and results

Details of the variables included in this study are presented in Table 4, Table 5, Table 6. About 63.5 percent of the respondents are females and 36.5 percent are males. The age group of 65–74 and married group represented the largest segments of the sample. The income groups were more or less widely represented.

A sample of 220 of the questionnaires was randomly selected from a larger number of responses and 200 valid responses were used to analyse the opinions of the senior persons on their

Discussion and conclusions

This article provides a short introduction to the use of Kohonen's SOM algorithm in tourism. It contains a brief description of an approach that has been widely applied in other fields but is still relatively novel in tourism. This article has also elaborated on a real-world application in the tourism industry. The motivations and concerns of West Australian seniors that are relevant to domestic and international holiday travel were investigated. The respondents could, based on their

Acknowledgements

JK gratefully acknowledges the research grant (UQ-ECRG-2000100168) funded by the University of Queensland in Australia. We wish to thank the anonymous reviewers and the editor for their helpful comments and suggestions.

References (49)

  • G Cybenko

    Approximation by superpositions of a sigmoidal function

    Mathematics of Control, Signals, and Systems

    (1989)
  • G Deboeck

    Trading on the edge

    (1994)
  • Environmetrics (1991). Needs and expectations of tourists aged over 55 years. Sydney, Australia: NSW Tourism...
  • D.P Garg et al.

    Artificial neural network based robot controlAn overview

    Journal of Intelligent and Robotic Systems

    (1996)
  • Häkkinen & Koikkalainen (1997). The neural data analysis environment. Proceedings of the workshop on self-organising...
  • S Haykin

    Neural networksA comprehensive foundation

    (1994)
  • J Hertz et al.

    Introduction to theory of neural computation

    (1991)
  • Hush D. R., & Horne, B. G. (1993). Progress in supervised neural networks: What's new since lippmann. IEEE Signal...
  • S Kaski et al.

    Tips for SOM processing and colorcoding of maps

  • Kendall, K., & Booms, B. (1989). Consumer perceptions of travel agencies: Communications, images, needs, and...
  • Kim, J. (2001). Connectionism: A new perspective for data mining problems. International conference on advances in...
  • C Klimasauskas

    Neural networksAn engineering perspective

    IEEE Communication Magazine

    (1992)
  • T Kohonen

    Self-organized formation of topologically correct feature maps

    Biological Cybernetics

    (1982)
  • T Kohonen

    Adaptive associative and self organizing functions in neural computing

    Applied Optics

    (1987)
  • Cited by (0)

    View full text