Rijal, Samsul (2025) PREDIKSI JUMLAH PERMINTAAN OBAT MENGGUNAKAN RANDOM FOREST REGRESSION BERDASARKAN DATA FAKTUR PENJUAL (TA.10.25.006). Diploma thesis, UNIVERSITAS LOGISTIK DAN BISNIS INTERNASIONAL.
Full text not available from this repository.Abstract
ABSTRAK PREDIKSI JUMLAH PERMINTAAN OBAT MENGGUNAKAN RANDOM FOREST REGRESSION BERDASARKAN DATA FAKTUR PENJUALAN: STUDI KASUS PAS EXPRESS SEMARANG Samsul Rijal D-IV Logistik Niaga-EL, Universitas Logistik dan Bisnis Internasional Ketepatan prediksi permintaan obat menjadi faktor krusial dalam menjaga efisiensi rantai pasok farmasi dan mencegah terjadinya overstock maupun stockout. Penelitian ini bertujuan membangun model prediksi permintaan obat menggunakan algoritma Random Forest Regression berbasis data faktur penjualan dari PAS Express Semarang. Data yang digunakan mencakup periode Januari 2023 hingga April 2025, diolah melalui tahapan pra-pemrosesan, agregasi bulanan, dan feature engineering. Model dikembangkan dalam konteks time series supervised learning dengan variabel input seperti lag_1 dan lag_3_avg. Evaluasi model menggunakan metrik MAE, RMSE, dan R² menunjukkan tingkat akurasi tinggi dan kesesuaian prediksi terhadap data aktual. Hasil prediksi diintegrasikan ke dalam proses “Plan” dalam model Supply Chain Operations Reference (SCOR) untuk mendukung pengambilan keputusan pengadaan dan distribusi. Penelitian ini memberikan kontribusi dalam pengembangan metode prediksi permintaan berbasis machine learning serta integrasinya dalam perencanaan rantai pasok farmasi yang lebih adaptif dan berbasis data. Kata Kunci: Prediksi permintaan obat, Random Forest Regression, faktur penjualan, machine learning, SCOR model. viii Universitas Logistik dan Bisnis Internasional ABSTRACT DRUG DEMAND FORECASTING USING RANDOM FOREST REGRESSION BASED ON SALES INVOICE DATA: A CASE STUDY AT PAS EXPRESS SEMARANG Samsul Rijal D-IV Commercial Logistics, Universitas Logistik dan Bisnis Internasional Accurate drug demand forecasting is crucial for ensuring efficiency in pharmaceutical supply chains while avoiding overstock or stockout conditions. This study aims to develop a predictive model using the Random Forest Regression algorithm based on sales invoice data from PAS Express Semarang. The dataset, covering January 2023 to April 2025, underwent preprocessing, monthly aggregation, and feature engineering. The model employs a time series supervised learning approach with inputs such as lag_1 and lag_3_avg. Model performance was evaluated using MAE, RMSE, and R² metrics, indicating high prediction accuracy. Forecast results were then integrated into the “Plan” phase of the Supply Chain Operations Reference (SCOR) model to enhance decision-making in procurement and distribution. This study contributes to the advancement of machine learning-based demand forecasting and its integration into more adaptive, data-driven pharmaceutical supply chain planning. Keywords: Drug demand forecasting, Random Forest Regression, sales invoice, machine learning, SCOR model
| Item Type: | Thesis (Diploma) |
|---|---|
| Subjects: | H Social Sciences > HF Commerce |
| Depositing User: | PKL SATU SATU |
| Date Deposited: | 22 Jul 2026 08:44 |
| Last Modified: | 22 Jul 2026 08:44 |
| URI: | http://eprints.ulbi.ac.id/id/eprint/4145 |
Actions (login required)
 |
View Item |
