Fuzzy Expert System for Melamine Moulding Compound Dye Feasibility Identification: A Case study on Plastic Industry

Abstract

This study addresses a practical decision problem within the melamine industry. Production managers are tasked with determining whether a new Melamine Moulding Compound (MMC) dye is suitable for use, based on testing parameters such as boiling point, processing time, and pressure. Given that these values frequently fall between established expert categories, manual decision-making can be challenging to justify and replicate. Accordingly, this research develops a web-based Fuzzy Sugeno expert system to assess the feasibility of MMC dye. The model incorporates three input variables, each characterized by low, medium, and high fuzzy sets. Expert knowledge is formalized into 27 rules employing zero-order Sugeno consequents for three classes: not feasible, conditionally feasible, and feasible. The system has been implemented as a PHP and MySQL application and is accessible via a publicly available login page. An illustrative example involving MMC103—boiling point of 165 °C, processing time of 65 seconds, and pressure of 55 bar—indicated medium and high membership values across all three parameters, activating eight rules. The aggregate firing strength was calculated as 3.00, the weighted consequent sum amounted to 80.00, and the final Sugeno score was 26.67. This score categorizes MMC103 as feasible. The results demonstrate that the model not only provides a classification label but also displays memberships, active rules, rule consequents, and the final computation, thereby enabling verification by the production manager. Furthermore, the study includes an English version of the application interface with privacy masking features for user data.