Probabilistic Life-Cycle Costing: Beyond the Great Expectation

点击翻译按钮获取中文摘要

The backbone of our nation's economy, Canada's civil infrastructure has been long suffering. The reasons are systemic and multi-faceted, ranging from funding deficits and uncoordinated systems planning, to unrelenting extreme events and inefficient engineering design and management. Meanwhile, public expectations on infrastructure have moved from safety, health and mobility to the notions such as intelligence, resilience, sustainability and flexibility. Transparency in infrastructure-related decision makings is also an important factor in winning public support. To meet these challenges, a risk-informed life-cycle infrastructure engineering (RILCIE) theory that provides a holistic solution to modern infrastructure issues must be established. The applicant's long-term research goal is to advance RILCIE by developing theories, methods and tools that can be applied to support practical decision makings. Aligning with this long-term goal, the proposed research program aims to answer a burning question that arises in almost every stage of the infrastructure life cycle: How much does it cost and how much contingency is required to build a piece of infrastructure and uphold its services to a satisfactory level? Searching for a complete answer to this question goes much deeper than simple cost accounting. It turns out to be a scientific endeavor that involves several fundamental, intertwined research issues, including stochastic degradation modeling, time-dependent reliability analysis, lifetime prediction, optimal inspection and maintenance planning, time preference and discounting, and risk-informed decision making. The proposed research program focuses on the development of an advanced probabilistic life-cycle costing system that would address these issues with an emphasis on the modeling of uncertainties involved in each key component of the analysis. The use of the life-cycle cost information goes beyond conventional applications such as alternative comparisons. Rather, it is proposed to play a pivotal role in evaluating innovative project delivery models such as public-private partnerships, developing integrated optimal design and asset management plans, and guiding risk-informed structural integrity management - each representing a very different aspect of infrastructure issues at distinguishing phases of infrastructure life. The proposed uncertainty modeling methodology for a full probabilistic quantification - rather than the average or expectation only - of the life-cycle cost, and the way of using the cost information for decision making support are highly original and innovative. The proposed research is expected to have a significant impact on project evaluation, engineering design, and infrastructure asset management. The research program includes a comprehensive highly-qualified personnel training plan of six PhD and three MASc students.