Open Collections

A Quality Function Deployment (QFD) Approach for Bridge Maintenance Management

University of British Columbia

Infrastructure age in the US/Canada are beyond half their expected service life. With billions of dollars invested annually, an increase in number of decisions towards maintenance, rehabilitation and replacement (MRR) activities are expected. Customer (infrastructure user) opinions are sometimes sought when major infrastructure-related decisions are made by conducting surveys, community meetings, etc. However, with consumers becoming more involved in economic, environmental, and social issues related to infrastructure, a process for ensuring customer demands are addressed would be valuable to all stakeholders involved. In this thesis, using bridge as an example, an innovative expert-based decision-framework has been proposed and developed using the Quality Function Deployment (QFD) approach. The framework comprises of three major elements. First a hierarchical evidential reasoning (HER) framework is proposed and developed for condition assessment of bridges by classifying bridge elements into Primary, Secondary, Tertiary and Life Safety-Critical elements. Respective indices are calculated in addition to an overall bridge condition index. The HER framework enables combining different distress indicators and propagating both aleatory/epistemic uncertainties using either Dempster-Shafer or Yager's rule. Importance and reliability factors (collectively termed "credibility factor") are introduced based on bridge element importance and reliability of collected data. Second, QFD implementation has been demonstrated with the following applications: (i) Inspection Prioritization (ii) Decision-Making between Replacement and/ or Rehabilitation scenarios. For inspection prioritization, an Inspection House of Quality is prepared for translating consumer demands (WHATs) into inspection requirements (HOWs) and demonstrated using data developed from Colorado Department of Transportation (CDOT) inspection manual. For the decision-making scenarios, a case study is furnished for a bridge located in Victoria, BC. Finally, the infrastructure-user's expectations are dynamic given the changing economic conditions, technologies, environmental regulations, etc. A hidden Markov model (HMM) is utilized for predicting such dynamic customer response by using probabilities of focus areas that are of interest to the infrastructure-user as hidden parameters. Using the 2005 California Transportation's customer survey, a case study is presented for demonstrating the application. This new expert-based framework has an ability to enhance decision-making by addressing uncertainty in collected inspection data, facilitating customer input into MRR procedures and by predicting customer expectations.

Text

2014-07-29

Attribution-NonCommercial-NoDerivs 2.5 Canada

http://hdl.handle.net/2429/48548

Civil Engineering

University of British Columbia

UBCO

http://creativecommons.org/licenses/by-nc-nd/2.5/ca/