Measuring Sustainability Performance of Supply Chain Management Practices Using Fuzzy Inference

Date
2015-10
Authors
Talebzadehhosseini, Seyyedmilad
Journal Title
Journal ISSN
Volume Title
Publisher
Faculty of Graduate Studies and Research, University of Regina
Abstract

Supply Chain Management (SCM) is critical for manufacturing operations, to increase their efficiency, productivity and quality. The sustainability performance of a SCM practice involves measuring its impact with respect to the three dimensions of sustainability: economic, social and environmental. In this work, the sustainability performance of a SCM practice is measured using an Intelligent System (IS) technique with the aim of not only measuring the sustainability performance simultaneously with respect to the three dimensions; but also the capability to incorporate user’s experience and knowledge, as well as accommodate ambiguous or vague information. The introduced IS technique is a Fuzzy Inference System (FIS) also enables users to decide whether it is necessary to improve the sustainability of fields, dimensions and overall. In order to consider expert knowledge and human thinking (by defining the consequences of rules), a Mamdani FIS was adopted. Mamdani FIS has the advantages of being intuitive, describes expert knowledge in a human-like manner and uses a defuzzification technique to provide a crisp output. The economic, environmental and social dimensions were each divided into five fields. The economic fields were: reliability, responsiveness, flexibility, financial performance and quality. The environmental dimension was given the fields: environmental management, use of resources, pollution, dangerousness and natural environment. For the social dimension the five dimensions were: work conditions, human rights, societal commitment, customers issues and business practices. Each field is in turn assigned some indicators or sub-fields. In this work, the overall sustainability performance was measured in three phases: in Phase I Nineteen Mamdani FIS modules (Modules-PI) were introduced to ii measure the impact in each sub-field, and in Phase II three Mamdani FIS modules (Modules-PII) were employed to measure the performance of each field. Finally in Phase III one Mamdani Fuzzy Inference System module (Module-PIII) utilized to measure the overall sustainability performance of a SCM practice. The functionality of this method was tested and the results were analyzed to demonstrate that the proposed IS technique can measure the sustainability performance of a SCM practice in manufacturing operations. The results showed that the introduced FIS modules enable managers and decision makers to measure the sustainability performance of a SCM practice in a manufacturing operation with respect to three dimensions individually and at once. Also, the method allows handling different levels of expert knowledge by adjusting the standard deviation of a Gaussian output membership function. Vague (ambiguous) data can also be accommodated. The fuzzy outputs have distributions that can be used to decide whether further effort is needed to increase the sustainability of fields, dimensions and overall. measure the impact in each sub-field, and in Phase II three Mamdani FIS modules

(Modules-PII) were employed to measure the performance of each field. Finally in Phase III one Mamdani Fuzzy Inference System module (Module-PIII) utilized to measure the overall sustainability performance of a SCM practice. The functionality of this method was tested and the results were analyzed to demonstrate that the proposed IS technique can measure the sustainability performance of a SCM practice in manufacturing operations. The results showed that the introduced FIS modules enable managers and decision makers to measure the sustainability performance of a SCM practice in a manufacturing operation with respect to three dimensions individually and at once. Also, the method allows handling different levels of expert knowledge by adjusting the standard deviation of a Gaussian output
membership function. Vague (ambiguous) data can also be accommodated. The fuzzy outputs have distributions that can be used to decide whether further effort is needed to increase the sustainability of fields, dimensions and overall.

Description
A Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Master of Applied Science in Industrial Systems Engineering, University of Regina. xv, 191 p.
Keywords
Citation
Collections