<![CDATA[Service quality in financial service companies depends on the ability to provide efficient and timely customer services. PT FIFGROUP Cikampek Branch utilizes the Integrated Self-Service Smart Queue Information System (ISSSIAP), which implements a sequential queueing system where customers may pass through different service stages with varying processing durations. This complexity makes customer service time prediction challenging and may affect resource allocation and queue management. This study aims to develop a customer service time prediction model using the Gradient Boosting Regressor algorithm based on historical ISS-SIAP data. A quantitative approach was applied through data collection, preprocessing, 10-Fold Cross Validation, model development, and performance evaluation using R 2 , MAE, MSE, and RMSE metrics. The experimental results show that the proposed model achieved an R 2 value of 0.866, MAE of 2.726 minutes, MSE of 15.308, and RMSE of 3.903 minutes. These results demonstrate that the Gradient Boosting Regressor provides good predictive performance in estimating customer service time. The proposed model provides practical support for improving operational efficiency at PT FIFGROUP Cikampek by enabling better service planning, resource allocation, and queue management decisions.]]>
Copyrights © 2026
