Life-cycle optimization of structural systems based on cumulative prospect theory: Effects of the reference point and risk attitudes. (February 2022)

• Record Type: Journal Article
• Authors: Cheng, Minghui; Frangopol, Dan M.
• Is Part Of: Reliability engineering & system safety. Volume 218:Part A(2022)
• Publication Date: 2022-02
• Subjects: Cumulative prospect theory -- Life-cycle optimization -- Reference point -- Risk attitude -- Probability weighting function; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2021.108100
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: Conduct a survey among students and engineers in the field of civil engineering. Propose a model to depict the opposite risk attitudes towards failure loss. Select the best-fitting model using Akaike information criterion. Investigate the effects of the reference point and opposite risk attitudes on life-cycle management of civil infrastructure. Abstract: Reliability-, risk-, and utility-based life-cycle maintenance is a normative approach for rational decision-making regarding structural systems under uncertainty. However, since these indicators cannot reflect the preference of stakeholders, the obtained plans may not be optimal from their standpoints. One of the most popular models to describe people's decisions under risk is cumulative prospect theory (CPT). CPT can capture the attitudes towards high consequence and low probability associated with structural failure by the value function and probability weighting function, respectively. CPT can also handle different risk attitudes towards gains and losses by introducing a reference point. Previous studies on calibration of CPT were developed for economics, healthcare, and traveling rout choices. Their results may not hold for civil infrastructure. Therefore, the parameters of CPT (i.e. value function, probability weighting function, and reference point) are calibrated through a survey among students and practicing engineers in the field of structural engineering. The fitted model is applied to life-cycle maintenance of a steel girder bridge. The calibration results show that the reference point is set to a prescribed plan. It is shown by the illustrative example that the reference point and opposite risk attitudes can significantly affect optimal decisions when searching for riskier plans.