Multi-criteria decision making using Fuzzy Logic and ATOVIC with application to manufacturing

In this paper multi-criteria decision making (MCDM) is investigated as a framework for classification of part quality in a manufacturing process. The importance of linguistic interpretability of decisions is highlighted, and a new framework relying on the integration of Fuzzy Logic and an existing MCDM method is proposed. ATOVIC, previously developed as a TOPSIS-VIKOR-based MCDM framework is enhanced with a Fuzzy Logic framework for decision making - Fuzzy-ATOVIC. This research work demonstrates how to add linguistic interpretability to decisions made by the MCDM framework. This contributes to explainable decisions, which can be crucial on numerous domains, for example on safety-critical manufacturing processes. The case study presented is the one of ultrasonic inspection of plastic pipes, where thermomechanical joining is a critical part of the manufacturing process. The proposed framework is used to classify (take decisions) on the quality of manufactured parts using ultrasonic images around the joint region of the pipes. For comparison, both the original and the Fuzzy Logic-enhanced MCDM methods are contrasted using data from manufacturing trials and subsequent ultrasonic testing. It is shown, that Fuzzy-ATOVIC provides a framework for linguistic interpretability while the performance is the same or better compared to the original MCDM framework.

[1]  Been Kim,et al.  Interactive and interpretable machine learning models for human machine collaboration , 2015 .

[2]  Leila Baccour,et al.  Amended fused TOPSIS-VIKOR for classification (ATOVIC) applied to some UCI data sets , 2018, Expert Syst. Appl..

[3]  Carlos Guestrin,et al.  Model-Agnostic Interpretability of Machine Learning , 2016, ArXiv.

[4]  Dimitris E. Koulouriotis,et al.  A fuzzy digraph method for robot evaluation and selection , 2011, Expert Syst. Appl..

[5]  Lalana Kagal,et al.  Explaining Explanations: An Overview of Interpretability of Machine Learning , 2018, 2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA).

[6]  Georgios Athanasopoulos,et al.  A decision support system for coating selection based on fuzzy logic and multi-criteria decision making , 2009, Expert Syst. Appl..

[7]  Brandon M. Greenwell,et al.  Interpretable Machine Learning , 2019, Hands-On Machine Learning with R.

[8]  Abhijit Majumdar,et al.  Determination of quality value of cotton fibre using hybrid AHP-TOPSIS method of multi-criteria decision-making , 2005 .

[9]  Zachary Chase Lipton.  The mythos of model interpretability , 2016, ACM Queue.

[10]  Rita Gamberini,et al.  A new multi-objective heuristic algorithm for solving the stochastic assembly line re-balancing problem , 2006 .

[11]  Andrzej Piegat,et al.  Comparative Analysis of MCDM Methods for Assessing the Severity of Chronic Liver Disease , 2015, ICAISC.

[12]  Been Kim,et al.  Towards A Rigorous Science of Interpretable Machine Learning , 2017, 1702.08608.

[13]  Morteza Yazdani,et al.  A state-of the-art survey of TOPSIS applications , 2012, Expert Syst. Appl..

[14]  Edmundas Kazimieras Zavadskas,et al.  State of art surveys of overviews on MCDM/MADM methods , 2014 .

[15]  Francisco Herrera,et al.  Interpretability of linguistic fuzzy rule-based systems: An overview of interpretability measures , 2011, Inf. Sci..

[16]  Paulo J. G. Lisboa,et al.  Making machine learning models interpretable , 2012, ESANN.