Evaluating smart manufacturing systems using fuzzy logic

• Main Author: Jennifer Grace John Alexander
• Format: text::Thesis
• Language: English
• Published: 2023

This research discusses about the adaptation of Smart Manufacturing Systems (SMS) in Industry 4.0. The increasing demand of digitally connected machines and knowledge-based manufacturing systems signifies the continuing and rapid growth of Smart Manufacturing Systems (SMS), thus increasing the demand from consumers for continuously innovative and higher quality products, and for affordable prices and product immediacy. These market players are constantly under pressure to make well informed decisions based on measured values by using systematic approaches, as to continuously stay ahead in the industry. The question as to how to measure the effectiveness of the configurations adapted in the manufacturing industry are constantly raised as the key indicators need to be identified and their values measured by using any systematic tools. Therefore, the objective of this research is to formulate a smart manufacturing systems’ framework configuration model to measure the effectiveness of the configurations, mainly focusing on how few identified indicators play a significant role in these configurations. The objectives were achieved by conducting thorough literature review on past research papers, to identify the research gap and the influencing indicators. Then, Fuzzy logic approach was proposed to test different types of configurations and this proposed model was implemented and analyzed by using MATLAB’s Fuzzy Logic Designer tool. Identified key indicators were chosen to be used as inputs to control the final output of the configuration model. Configurations were manipulated based on how indicators such as Quality, Leadtime and Cost affects the manufacturing Cost justification in multiple setups. Results obtained from various configuration tests were later presented to actual field engineers from manufacturing industry for evaluation and validation. Based on this research, it is proven that this proposed configuration model has a satisfactory rate of 83.7%, as this was achieved by significant feedbacks from field engineers. This research has significantly facilitated the identification of influential indicators and the measured relationship of the indicators in the formulated configurations, hence enabling the best configuration approach to be identified. Therefore, it can be concluded that a visualized and measured configuration system has the capability to influence decision makings in manufacturing industry, thus allowing manufacturers to stay competent by making well-versed decisions proactively.