Implementasi Algoritma Random Forest untuk Prediksi Keterlambatan Produksi pada PT Pelangi Grafika

Abstract

Production delays are one of the problems that can affect operational effectiveness and customer satisfaction in the printing industry. PT Pelangi Grafika still identifies potential production delays based on experience, making it necessary to develop a system capable of providing more objective predictions. This study aims to implement the Random Forest algorithm in a web-based application to predict production delays using historical production data. The dataset consists of order quantity, product type, deadline, estimated production time, number of machines, number of operators, difficulty level, number of revisions, and production queue. The research stages include data preprocessing, training and testing data splitting, Random Forest model construction, and model evaluation using a Confusion Matrix with accuracy, precision, recall, and F1-score metrics. The results indicate that the proposed model achieves excellent classification performance in predicting production delay status. Furthermore, the developed application assists the company in identifying potential production delays at an earlier stage, thereby supporting more effective decision-making.