<![CDATA[The MSME-scale batik industry in Sidoarjo plays a vital role in the regional economy, but product quality consistency remains a challenge due to variations in production processes and the minimal application of data-driven quality control. This research develops a batik product defect prediction model using the Logistic Regression algorithm as an accurate and easily interpretable Machine Learning approach. The dataset consists of 1,250 observations collected over three months of production, covering variables such as dyeing temperature, wax temperature, dye concentration, dipping duration, room humidity, drying duration, operator experience, stamping pressure, and number of products per shift. The results indicate that the Logistic Regression model performs very well, achieving an accuracy of 86.4% and an AUC of 0.91. The variables that most significantly influence defects are room humidity, dyeing temperature, operator experience, and stamping pressure. These findings form the basis for developing a data-driven quality control system that can help MSMEs establish optimal process parameters, manage production capacity, and improve operator skills. This research proves that the Machine Learning approach can be effectively applied in the MSME context with significant practical and operational benefits. Keywords: Machine Learning, Logistic Regression, Batik MSMEs, Prediction Model]]>
Copyrights © 2025
