Towards Structured Public Involvement in Highway Design: A Comparative Study of Visualization Methods and Preference Modeling using CAVE (Casewise Visual Evaluation)

The goal of this research is to develop a methodology that will improve public satisfaction with highway design and planning where visual presentation of design options is performed. A review of two visual assessment methodologies shows that these either cannot provide satisfactory design element guidance or are too resource-intensive to be useful in situations where many design elements are under consideration. A design principle called elemental decomposition is introduced and a Casewise Visual Assessment (CAVE) methodology is developed. Treating public preference as a complex non-linear system, CAVE employs fuzzy set theoretic modeling techniques to generate a preference knowledge base that can be used to investigate and quantify public preference for specific design strategies. CAVE’s robustness enables accurate modeling of public preferences even under conditions of partial knowledge. Using a combination of iterative focus group public involvement and an electronic scoring system to solicit rapid feedback the research team investigates the efficacy of three different visualization modes (2Dimensional, 3Dimensional and Virtual Reality) for a case study in the Bluegrass region of Kentucky. Public preference for these visualization modes is measured and the advantages of each mode are described. Using the preferred 3D visualization mode public preference for specific design elements is solicited and analyzed. Graphical preference surface models are presented and their features are discussed. Experimental shortcomings are summarized and future enhancements to CAVE are proposed. Emphasis is placed on the design and deployment of analytic methodologies for preference modeling within a framework of Structured Public Involvement.