System reliability theory based decision supporting system for optimizing construction costs and quality

Abstract

The construction industry is accustomed to failing in the delivery of quality products and is facing many quality problems. On the other hand, participants in the construction industry have been under higher pressure than ever before to control construction costs. As a result, proper actions should be undertaken to maximize the quality level of construction projects while minimizing their resource consumption. Such actions are addressed by system reliability optimization (SRO), one of the most creative tasks in the domain of construction management. Although the concept of SRO is not new and has been a thought-provoking subject for a long time, little success has been recorded in its application to the construction industry. This study aims to develop a "system reliability theory based two-stage decision supporting system (DSS)", which consists of two critical and interrelated sub-models, namely, the fuzzy flexibility assessment model of reliability improvement potential (MODEL I) and the particle swarm optimization (PSO) based SRO model (MODEL II), to assist decision makers (DMs) in identifying and selecting the optimal cost-quality trade-off solution for construction projects. For MODEL I, based on an extensive literature review, a set of assessment criteria of flexibility level is developed. Meanwhile, the whole construction project is broken down into several work packages by employing the work breakdown structure (WBS) approach. Second, expert interviews are carried out to collect the assessment data, including the subjective scores for each work package in relation to each criterion, the fuzzy IF-THEN rules and their confidence degree levels of making each judgment. Third, fuzzy logic is applied to process such quantitative and qualitative data, and crisp outputs are generated. Lastly, in accordance with the confidence degree levels, the crisp outputs are combined to form the reliability improvement flexibility index (RIFI) for each work package. For MODEL II, based on the outputs from MODEL I, including the RIFI of each work package and the WBS, the system reliability theory is applied to quantify project quality. As a result, the nonlinear cost-reliability function and system reliability structural function are set up. Second, the objective functions, i.e. total construction cost (TCC) minimization, and system reliability maximization, along with the optimization constraints can be defined. Third, SRO is carried out by employing PSO as the optimization tool. Lastly, a set of Pareto-optimal solutions is generated, from which DMs can identify and select the final cost-quality trade-off solution according to the post-analysis preference information. At the accomplishment of this study, it is expected that this two-stage DSS for optimizing construction costs and quality can be developed for practical applications which could revolutionize the project planning, quality management and optimization concepts in construction. The feasibility and workability of the proposed DSS are tested and verified by two case studies, one belonging to an addition and alternation (A&A) project and the other a concrete and finishing project. The results show that the proposed DSS can effectively and efficiently assist DMs to identify and select the optimal cost-quality trade-off solution, and fully prove its validity and practicality. Finally, the advantages, limitations and further study potentials of the proposed DSS are highlighted and recommended.

Date of Award	16 Jul 2012
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Chi Ming TAM (Supervisor) & Ka Chi LAM (Supervisor)