The tourist's drives : GIS oriented methods for analysing tourist recreation complexes

See also theweb sitebased on this thesisTourism is a product of diverse composition. An increasing number of people pursue their own specific wishes and combine various products which may or may not be intended for tourists; they create their own individual holiday package. In order to determine how this trend of combining elements influences the use of (tourist) products in a region, it is necessary to gain insight into tourist time-space behaviour. Time, space and context are important domains for describing tourist time-space behaviour. People differ, situations constantly change and a particular interaction depends on the circumstances (personal and topological) in which it takes place. The analysis of tourist time-space behaviour might provide an explanation for this combinatory behaviour. This type of analysis requires specific personal data about time spent, places visited, routes chosen, information used, perception and motivation. Not only the visible tourist time-space pattern is important, but also the process involved.To date, most researchers have attempted to analyse spatially related tourism data using statistical methods. The data structure needed for such a statistical analysis requires data for each period considered and for each possible location and road in a region. However, a maximum of only 1% of these data is likely to be significantly related to one person. Furthermore, the enormous size of the data set makes it difficult to uncover spatial relations. Geographical Information Systems (GIS) are capable of handling spatial relationships. Four main data groups can be distinguished:(1) tourist related characteristics;(2) perception of space and of activities undertaken, and observed time-space behaviour;(3) spatial objects;(4) specific (tourism) codes added to these objects.The constructed tourist recreation complex can be understood as an interwoven structure of several different network s. None of these networks prevails or determines tourist behaviours exclusively. A methodology consisting of two steps is proposed for the analysis of tourist time-space behaviour:(1) Survey the use of the physical environment by tourists, using exploratory spatial data analysis techniques and dynamic visualisation. Determine clusters of product elements and a possible typology of tourist groups.(2) Deduce, describe and analyse tourist recreation complexes using graph and network analysis techniques, and statistical methods. The individual network is based on products and product-clusters and tourist time-space behaviour in relation to the use of the environment and the tourist's perception of it. Execute pattern analysis using graph techniques and accessibility studies for the links and nodes in the network.Data visualization is used to make patterns in scientific data visible. The application of dynamic cartography adds a new dimension to the visualization process: data can be interactively explored for errors and patterns. The Cartographic Data Visualizer for Time-Space data (CDV-TS) can be used to make a coherent analysis of the use of space, the time distribution and the context of time- space behaviour. GIS is an instrument which is particularly suited to the analysis of clearly limited physical elements. Current GIS software can be applied to obtain a static overview and to perform spatial analyses of the use of a region at a certain moment in a specific context. The storage of time-space data within the GIS data structure is more efficient than the data storage for a statistical application. However, the statistical uses of current GIS are limited to descriptive forms. A linkage between GIS and statistical software creates a powerful instrument. The current generation of commercial GIS software is not capable of dealing with time. Applications were developed to approximate this. A GIS has few network capabilities for supporting tourist time-space behaviour analyses. Network pattern recognition and comparison is not possible at all, and network indices cannot be calculated within a GIS. A newly developed morphologic pattern describer seems appropriate for comparing different constructed network patterns.Two data sets were used to illustrate how the applications and approaches developed can describe a tourist recreation complex in a tourist region. The applications otter a wealth of opportunities for the interactive examination of time- space oriented data, and to search for different tourist combinations of products supplied. A main drawback of the applications is the amount of data that has to be processed