Product Quality Control Forecast Using Machine Learning Algorithms: A Case Study

In today’s highly competitive global market, industries require near-perfect quality to be profitable. Therefore, the prompt prediction of a defect product has become an issue of paramount importance and an opportunity for manufacturing companies to move quality standards forward. This article presents a real case study based on Machine Learning algorithms suggested to create a knowledge based intelligent supervisory system to predict defect products on a fashion industry. Defect detection is formulated as a binary classification problem and several Machine Learning algorithms have been compared for a clearer understanding of which algorithm is best suited to the available data. The predictions have been obtained performing the algorithms on pre-processed dataset which was made available by the industry. The Random Forest, LightGBM and C5.0 algorithms performed similarly well on the data. However, the Random Forest algorithm has been selected as the best one since it allows to reduce the rate of false negative (i.e. the proportion of defected-free products wrongly classified as defected ones), that is the goal of the analysed industry.