Development of a hierarchical fuzzy model for the evaluation of inherent safety

Date

2004-11-15

Journal Title

Journal ISSN

Volume Title

Publisher

Texas A&M University

Abstract

Inherent safety has been recognized as a design approach useful to remove or reduce hazards at the source instead of controlling them with add-on protective barriers. However, inherent safety is based on qualitative principles that cannot easily be evaluated and analyzed, and this is one of the major difficulties for the systematic application and quantification of inherent safety in plant design. The present research introduces the use of fuzzy logic for the measurement of inherent safety by proposing a hierarchical fuzzy model. This dissertation establishes a novel conceptual framework for the analysis of inherent safety and proposes a methodology that addresses several of the limitations of the methodologies available for current inherent safety analysis. This research proposes a methodology based on a hierarchical fuzzy model that analyzes the interaction of variables relevant for inherent safety and process safety in general. The use of fuzzy logic is helpful for modeling uncertainty and subjectivities implied in evaluation of certain variables and it is helpful for combining quantitative data with qualitative information. Fuzzy logic offers the advantage of being able to model numerical and heuristic expert knowledge by using fuzzy IF-THEN rules. Safety is traditionally considered a subjective issue because of the high uncertainty associated with its significant descriptors and parameters; however, this research recognizes that rather than subjective, "safety" is a vague problem. Vagueness derives from the fact that it is not possible to define sharp boundaries between safe and unsafe states; therefore the problem is a "matter of degree". The proposed method is computer-based and process simulator-oriented in order to reduce the time and expertise required for the analysis. It is expected that in the future, by linking the present approach to a process simulator, process engineers can develop safety analysis during the early stages of the design in a rapid and systematic way. Another important aspect of inherent safety, rarely addressed, is transportation of chemical substances; this dissertation includes the analysis of transportation hazard by truck using a fuzzy logic-based approach.

Description

Citation