Congratulations on our new doctorate

Congratulations, Dr. Peter Allan Johnson, who successfully defended his dissertation, Visioning Local Futures: Agent based modeling as a tourism planning support system.

Abstract: Planning is process that operates in a complicated, poorly understood environment, one that is typified by multiple interacting processes. This makes the development of plans required to meet future desired outcomes a difficult and challenging task. An example of this complex environment is readily seen in tourism planning, where multiple industries operate across scales to influence a distinctly spatial phenomenon. The use of a planning support system (PSS) is one way in which planners can mediate these types of complexities. PSS are generally considered as any type of technology used to support one or more planning tasks. In looking to develop new methods of providing planning support, one type of computer simulation approach, agent-based modeling (ABM), has recently been used in academic research as an approach to simulating complex, adaptive, and heterogeneous systems. While there are many advantages and examples of successful application of ABM within academic research, comparatively little work has explored the use of ABM to support planning practice, especially with evaluation from planners themselves.

This research approaches the problem of complexity in tourism planning in three general steps; 1) the linkages and nature of tourism as a phenomenon are formalized within an individual-based tourism framework. This framework is used as a foundation with which to conceptualize how tourism functions in a way that is more analogous to reality compared to previous models. 2) This conceptual framework is formalized, using empirical data from a diverse range of sources, into an ABM-based PSS used to examine tourism dynamics in the Canadian province of Nova Scotia. Nova Scotia is a traditionally “have-not” province, where tourism represents a growing component of the economy. 3) This simulation model is evaluated by tourism planning professionals, to identify the specific planning tasks to which this model add greatest value and identify areas of adoption constraint. Tourism planners found that the model was most appropriate for research support tasks, such as communication, scenario development, and data visualization, and was not considered appropriate for use in a decision support role, due to a lack of predictive capacity.

First, issues of validation and representation are crucial modeling components that must be rethought in light of the disconnect between the needs of the planner and the requirements of an ABM approach. Second, from the perspective of tourism planners, the use of an ABM-based PSS is strongest in a research support role. Lastly, issues of prediction – related directly to validation, are acting to inhibit the confidence that planners have in using an ABM in a decision-support capacity. Further work on model validation is required, but it is noted that a lack of predictive ability does not negate the potential of this technology in planning, rather this research shows that there are indeed many non-predictive and valuable uses of ABM within tourism planning, and that further exploration of these areas provides a rich area for future work.