| Cart Total:
Menu
dP-FMEA: An innovative Failure Mode and Effects Analysis for distributed manufacturing processes
  • Quality

dP-FMEA: An innovative Failure Mode and Effects Analysis for distributed manufacturing processes

Publication:
Quality Engineering
Date:
July 2020
Issue:
Volume 32 Issue 3
Pages:
pp. 267-285
Author(s):
Maisano, Domenico A., Franceschini, Fiorenzo, Antonelli, Dario
The copyright of this article is not held by ASQ.

Abstract

The Failure Mode and Effects Analysis (FMEA) is a powerful tool to design and maintain reliable systems (products, services or manufacturing processes), investigating their potential failure modes from the threefold perspective of severity, occurrence and detection. The Process FMEA, or more briefly P-FMEA, is a declination of the FMEA for manufacturing processes (or parts of them). Being progressively characterized by decentralized networks of flexible manufacturing facilities, the current scenario significantly hampers the implementation of the traditional P-FMEA, which requires the joint work of a group of experts formulating collective judgments. This paper revises the traditional P-FMEA approach and integrates it with the ZMII-technique – i.e. a recent aggregation technique based on the combination of the Thurstone’s Law of Comparative Judgment and the Generalized Least Squares method – allowing experts distributed through organizations to formulate their judgments individually. The revised approach – referred to as “distributed-Process FMEA” or more briefly dP-FMEA – allows to manage a number of experts, without requiring them to physically meet and formulate collective decisions, thus overcoming a relevant limitation of the traditional P-FMEA. The dP-FMEA approach also includes a relatively versatile response mode and overcomes several other limitations of the traditional approach, including but not limited to: (i) arbitrary formulation and aggregation of expert judgments, (ii) lack of consideration of the dispersion of these judgments, and (iii) lack of estimation of the uncertainty of results. The description is supported by a real-life application example concerning a plastic injection-molding process.

ALREADY A MEMBER?    REGISTER