Multi-person and multi-attribute design evaluations using evidential reasoning based on subjective safety and cost analyses

Wang

2015

Ocean Engineering

-This paper proposes a novel framework for analysing and synthesising engineering system risks on the basis of a generic Fuzzy Evidential Reasoning (FER) approach. The approach is developed to simplify the inference process and overcome the problems of traditional fuzzy rule based methods in risk analysis and decision making. The framework, together with the FER approach has been applied to model the safety of an offshore engineering system. The benchmarking between the new model and a well-established Rule based Inference Methodology using the Evidential Reasoning (RIMER) is conducted to demonstrate its reliability and unique characteristics. It will facilitate subjective risk assessment in different engineering systems where historical failure data is not available in their safety practice.Keywords -FMEA, fuzzy rule base, evidential reasoning, maritime risk, maritime safety. The linguistic variables used to describe X k (iM; kL). X kThe k th child attribute in a parent-child link.

Section: Input Transformation (Associated With P)mentioning

confidence: 99%

Section: Input Transformation (Associated With P)mentioning

confidence: 99%

Section: Input Transformation (Associated With P)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Use of fuzzy risk assessment in FMEA of offshore engineering systems

Wang

2015

Ocean Engineering

A multigranular hierarchical linguistic model for design evaluation based on safety and cost analysis

Martínez

et al. 2005

Int. J. Intell. Syst.

Before implementing a design of a large engineering system different design proposals are evaluated. The information used by experts to evaluate different options may be vague and/or incomplete. Although different probabilistic tools and techniques have been used to deal with these kinds of problems, it seems better to use the fuzzy linguistic approach to model vagueness and the Dempster-Shafter theory of evidence for modeling incompleteness and ignorance. In the evaluation of alternative designs, different criteria can be considered. In this article an evaluation process is developed in terms of Safety and Cost analysis. Both criteria involve uncertainty, vagueness, and ignorance due to their nature. Therefore, we propose an evaluation process defined in a linguistic framework where both criteria will be conducted in different utility spaces, i.e., in a multigranular linguistic domain. Once the evaluation framework has been defined, we present an evaluation process based on a Multi-Expert Multi-Criteria decision model that will be able to deal with multigranular linguistic information without loss of information in order to evaluate different design options for an engineering system in a precise manner. Accordingly, we propose the use of a multigranular linguistic model based on the Linguistic Hierarchies presented by Herrera and Martínez~"A model based on linguistic 2-tuples for dealing with multigranularity hierarchical linguistic contexts in multi-expert decision-making." IEEE Trans Syst Man Cybern B 2001;31~2!:227-234!.

Intelligent decision system for self‐assessment

Multi Criteria Decision Anal

2003

Many small and medium enterprises (SMEs) in the UK use the beta (Business Excellence Through Action) approach to the EFQM Excellence Model to conduct business excellence self-assessment, which is in essence a multiple criteria decision analysis (MCDA) problem. This paper introduces a decision support software package called Intelligent Decision System (IDS) to implement the beta approach. It is demonstrated in the paper that the IDS-beta package can provide not only average scores but also the following numerical results and graphical displays on: The IDS-beta package also provides a structured knowledge base to help assessors to make judgements more objectively. The knowledge base contains guidelines provided by the developers of the beta approach, best practices gathered from research on award winning organizations, evidence collected from companies being assessed and comments provided by assessors to record the reasons why a specific criterion is assessed to a certain grade for a company. Four small UK companies, the industry partners of the research project, have carried out the preliminary self-assessment using the package. The results and experience of the application are discussed at the end of the paper.