Development of an expert system to conduct automated HAZOP studies

Rahman, Shibly (2006) Development of an expert system to conduct automated HAZOP studies. Masters thesis, Memorial University of Newfoundland.

[img] [English] PDF - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (2893Kb)

Abstract

Process hazard analysis is an important step to identify risk in a process facility. Automation of hazard identification requires efficient search techniques with the aid of a knowledge base. It also requires an easy menu driven interface so that an ordinary user can interact with the system with minimal intervention from an expert. One of the models to implement automation of hazard analysis is HAZOP (hazard and operability) study. Fault propagation, an aspect of HAZOP analysis, defines the propagation of deviation among equipment in a process facility. To identify all the possible hazards and their faster access, it is necessary to develop an efficient fault propagation algorithm with a knowledge-base. The existing tools performing automated HAZOP analysis does not provide any means to identify the propagation of deviation to all downstream equipment. Also some of the developed tools are slower in data extraction, require an expert to interpret the analyzed result, focuses more on causes of deviation in a process facility than the consequences, and is specific to process facility structure. -- This thesis focuses on development of an expert system to perform automated HAZOP analysis with a unique fault propagation algorithm that will identify propagation of deviation to all downstream equipment in a process facility. Furthermore, the expert system has a knowledge base that identifies all general causes and consequences of equipment failure in a process facility and enables effective and efficient decision making tool for the user of the system.

Item Type: Thesis (Masters)
URI: http://research.library.mun.ca/id/eprint/11307
Item ID: 11307
Additional Information: Includes bibliographical references (leaves 64-66).
Department(s): Science, Faculty of > Computer Science
Date: 2006
Date Type: Submission
Library of Congress Subject Heading: Expert systems (Computer science); Hazardous substances--Risk assessment.

Actions (login required)

View Item View Item

Downloads

Downloads per month over the past year

View more statistics