Multiattribute Selection from Alternative Designs of Infrastructure Components for Accidental Situations

Abstract

The design of infrastructure systems for "accidental" situations, i.e., abnormal service conditions, may include a comparison of alternative design solutions of system components. These solutions are compared by means of applying a universal methodology known as a multiattribute selection (MAS). The alternative designs are described by a number of attributes that characterize each of them. Failure probability and quantitative measure of the risk are used as safety-related attributes of the alternative designs. This article shows how to carry out the MAS in the presence of epistemic uncertainties in safety-related attributes. Distributions of epistemic uncertainties are specified by means of methods used for quantitative risk assessment. The MAS is carried out as a propagation of these uncertainties. The Monte Carlo simulation is used to propagate uncertainties through the algorithm of MAS. The approach proposed in this article is illustrated by two examples, which consider alternative designs of a protective barrier and alternative sprinkler systems.

Design with Uncertainty: The Role of Future Options for Infrastructure Integration

“…The future options approach is an effort to address such uncertainties and any significant risks associated with them via a new epistemic framework. This framework can complement and work alongside a wide range of existing models that aim to reduce epistemic uncertainties, such as interval modeling (Change et al., ), Bayesian modeling (Cheung and Beck, ; Yin et al., ; Yuen and Mu, ), chaos theory (Schoefs and Yanez‐Godoy, ), evidence theory (Dixon and Rilett, ; Zavadskas and Vaidogas, ), fuzzy modeling (Faturechi and Miller‐Hooks, ; Zhang et al., ), Monte Carlo simulations Jahani et al., ), and gray system theory (Tserng et al., ).…”

A Comparison of Methods Used for Fire Safety Evaluation

“…The frequency of relatively rare occurrences of fires and thus of outcomes o ir can be estimated by means of the classical Bayesian approach to quantitative risk assessment (Aven and Pörn 1998;Vaurio and Jänkälä 2006;Vaidogas and Juocevičius 2009). In the context of this approach, likelihoods l ir will be estimated in the form of epistemic uncertainty distributions related to true values of l ir (Zavadskas and Vaidogas 2009). Such estimating is usually carried out by propagating epistemic uncertainties through such logical models of quantitative risk assessment as the event trees shown in Fig.…”

Reliability Assessment with Fuzzy Random Variables Using Interval Monte Carlo Simulation

“…There are different approaches to model the epistemic uncertainty, such as interval modeling (Neumaier, 1990;Change et al, 2001), Bayesian modeling (Ang and Tang, 1975;Cheung and Beck, 2010;Tao et al, 2010;Yuen and Mu, 2011;Yuen and Katafygiotis, 2006), chaos theory (Schoefs and Yanez-Godoy, 2011), evidence theory (Dempster, 1967;Guan and Bell, 1991;Dixon and Rilett, 2002;Zavadskas and Vaidogas, 2009), and fuzzy modeling (Bandemer and Gottwald, 1995;Möller et al, 1999;Muhanna and Mullen, 1999;Zimmermann, 1992). In this article, incomplete knowledge about the distribution parameters is modeled using fuzzy numbers.…”