Knowledge Modeling Tool for Evidence-Based Design

Abstract

Aim of the Study:

The aim of this study is to take evidence-based design (EBD) to the next level by activating available knowledge, integrating new knowledge, and combining them for more efficient use by the planning and design community. This article outlines a framework for a performance-based measurement tool that can provide the necessary decision support during the design or evaluation of a healthcare environment by estimating the overall design performance of multiple variables.

Background:

New knowledge in EBD adds continuously to complexity (the “information explosion”), and it becomes impossible to consider all aspects (design features) at the same time, much less their impact on final building performance.

Research Questions:

How can existing knowledge and the information explosion in healthcare—specifically the domain of EBD—be rendered manageable? Is it feasible to create a computational model that considers many design features and deals with them in an integrated way, rather than one at a time?

Approach:

The found evidence is structured and readied for computation through a “fuzzification” process. The weights are calculated using an analytical hierarchy process. Actual knowledge modeling is accomplished through a fuzzy neural tree structure. The impact of all inputs on the outcome—in this case, patient recovery—is calculated using sensitivity analysis. Finally, the added value of the model is discussed using a hypothetical case study of a patient room.

Conclusion:

The proposed model can deal with the complexities of various aspects and the relationships among variables in a coordinated way, allowing existing and new pieces of evidence to be integrated in a knowledge tree structure that facilitates understanding of the effects of various design interventions on overall design performance.