Utility-based neurofuzzy approach for engineering performance assessment in industrial construction projects
Abstract
Engineering performance is a prime determinant of the successful implementation of industrial construction projects. Outcomes of engineering activities have a far-reaching impact on the entire life cycle of the project and quality of the constructed facility. However, there is a major lack of understanding of the interaction between a project's environment and its resulting engineering performance. With the current surge in developing industrial facilities, there is a need to have analytical schemes that relate engineering performance to its driving project variables and methods to further quantify this performance. The research introduces a generic system, which incorporates neurofuzzy computing with multiple attribute utility functions, for the assessment of engineering performance in industrial construction projects. Because of their fault-tolerance, and ability to model non-linear relationships and to handle imprecision in variable description, neurofuzzy systems are used for relating the measures of engineering performance to the set of project input variables identified to have the largest impact on such performance. The use of multiple attribute utility functions allows the generic system to simultaneously assess various measures of performance and provide a collective evaluation of the total engineering performance in the project. Questionnaire surveys were used to acquire data necessary for the generic system implementation. The system is further verified and validated for a proper adaptability and functionality. Statistical methods are employed for comparison purposes with the developed generic system. The study demonstrates the use of the generic system in several practical applications such as prediction of engineering performance measures, assessment of total and relative engineering performance, recognition of the project variables having the largest impact on engineering performance, among others.
Degree
Ph.D.
Advisors
Chang, Purdue University.
Subject Area
Civil engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.