Assessing Production Line Risk using Bayesian Belief Networks and System Dynamics

Sudhir Punyamurthula, Fazleena Badurdeen

Research output: Contribution to journalConference articlepeer-review

5 Scopus citations

Abstract

Increased complexity in product design, strict regulations and a changing market make risk assessment critical for successful operations. Failure in responding quickly to raw material shortages, downtimes, deteriorating equipment conditions or other operational issues can prove to be an expensive affair. A company-wide risk assessment includes both external and internal operations. However, external/supplier risk assessment has been of major interest. Even though the scope of risk assessment at the production line level is not as broad as it is at the supply chain level, assessing risk would help recognize vulnerable areas of the production line, which would in turn help reduce damage caused when risk events occur. In this research, a method for production line risk assessment is proposed by considering operational risks affecting the line. Operational risks and their causal relationships are represented using Bayesian Belief Networks (BBN). The impact of these risks is observed using a simulation model of the production line using System Dynamics (SD) approach. The combination of BBN and SD assists in developing a versatile methodology, which can capture the dynamic causal mechanisms in a complex system, the uncertainties amongst risk events and the long-term impact of operational risks on the production line.

Original languageEnglish
Pages (from-to)76-86
Number of pages11
JournalProcedia Manufacturing
Volume26
DOIs
StatePublished - 2018
Event46th SME North American Manufacturing Research Conference, NAMRC 2018 - College Station, United States
Duration: Jun 18 2018Jun 22 2018

Bibliographical note

Publisher Copyright:
© 2018 The Author(s). Published by Elsevier B.V.

Keywords

  • Bayesian Belief Networks
  • Risk Assessment
  • System Dynamics

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Assessing Production Line Risk using Bayesian Belief Networks and System Dynamics'. Together they form a unique fingerprint.

Cite this