Keywords: multivariate regression, principal components, safety, database, artificial intelligence, fuzzy logic, safety index.

The safety evaluation of a constructed environment is often subjective and involves many components with differing degrees of impact on the environment. Many knowledge-based safety systems have been developed, but most focus on improving 'safety management', not 'safety' itself. Bias could exist if the standard of safety management evaluation is not well-developed.

This paper's goal is the development of a methodology that combines statistical methods and database design to assess the safety levels of a constructed environment. The principles of crime prevention through environmental design (CPTED) and fuzzy logic are used to identify the safety factors and build the knowledge base. Theoretically, fuzzy logic can be used solely, but only, but complexity involved in building the knowledge base for inference becomes enormous when the number of variables is too large. Fuzzy logic is basically a computing discipline that governs the calculations of words or combinations of similar expressions. One of its strengths is its ease to build an inference engine that can be used to assess certain evaluations.

The database plays an important role in developing this methodology because it has the capability to constrain the data structure and minimize the data collection. A relational database is developed for this program to provide the data for statistical analysis. To monitor the progress of the targeted environment, it is very important to store data and preserve the structure for future analysis. In addition, data can be retrieved with simple SQL queries, which provide a good functionality for presenting the result.

CPTED principles are experience-based rules that incorporate the environmental design with the user's awareness of safety. It is primarily used to identify factors that may potentially influence the user's safety in the environment. It also helps to characterize the physical components where the safety of the user is concerned.

An internet survey program is developed to collect users' input. This program includes a survey, virtual reality (VR) models, and a knowledge base which is embedded in the Web Server pages. Survey questions are related to the user's feelings. Another important aspect of this methodology is automation. An automated process and data simulation will enable a web-based survey for users to enter their input on safety. This in turn will provide estimates and information about the targeted components of the constructed facility. This automated system questions users about their feelings when using the facility. Their responses are then sent back to the web server for analysis. Results from the server can be displayed upon the user's request. With the help of those responses, the process can be simulated for the future reference and prediction. This serves as an evaluation tool for safety improvement and renovation of the facility. The application's design is based on Microsoft.NET structure. internet information service (IIS) is used to provide the web service for processing user's input/output. This automated system consists of three major designs: web-based graphical user interface for collecting and illustrating the information, a server-side scripting that performs the calculations, and a database that stores the data and maintains the data structure.

Proper use of statistics could change the behaviour of service provider such as the facility designers and consumers such as garage users. With the addition of fuzzy set theory and interactivity from the use of the WWW, this automation and simulation tool, which provides a real-time result, can be beneficial to the industry as well as to researchers in assessing and improving user safety for a constructed environment. Use of fuzzy logic helps in integrating the complex processes that are difficult to resolve, but has its limits. This paper presents a methodology and a computer program for safety evaluation and may serve as a practical solution to improving the user's safety.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)