Issue |
RAIRO-Oper. Res.
Volume 51, Number 4, October-December 2017
|
|
---|---|---|
Page(s) | 1077 - 1100 | |
DOI | https://doi.org/10.1051/ro/2017010 | |
Published online | 24 November 2017 |
A stochastic approach for failure mode and effect analysis
1 Department of Industrial Eng., Ataturk University, 25240, Erzurum, Turkey.
elif.kdelice@atauni.edu.tr
2 Department of Industrial Eng., Baskent University, 06810, Ankara, Turkey.
gfcan@baskent.edu.tr
Received: 11 August 2016
Accepted: 11 February 2017
This study presents a novel approach combining Failure Mode and Effect Analysis (FMEA) and Multi-Attributive Border Approximation Area Comparison (MABAC) method based on a stochastic evaluation process to prioritize potential failure modes (FMs) in an assembly line. The aim of the proposed approach is to improve the performance of FMEA by eliminating its shortcomings addressed in the study. In this context, firstly the risk factor (RF) importance weights and the performance values of the FMs for the RFs are determined by generating random numbers having uniform distribution in a range of minimum and maximum value of a limited number of expert evaluations. In this wise, the number of experts are increased to improve effectiveness of the risk evaluation process. Diverse opinions of experts are also assessed more precisely. Secondly, the priorities of the FMs are identified by implementing MABAC method. MABAC is a practical and reliable tool which provides stability for solutions. Finally, a comparative analysis is implemented to confirm the effectiveness of Stochastic FMEA-MABAC approach.
Mathematics Subject Classification: 90B50
Key words: FMEA / Stochastic / MABAC / Uniform Distribution / MCDM
© EDP Sciences, ROADEF, SMAI 2017
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.