Pemanfaatan Predictive Analytics Untuk Peramalan Kebutuhan Barang Habis Pakai Pada Sistem Manajemen Inventaris Terpusat Klinik Multi-Cabang

Hutapea, Monika Damelia (2026) Pemanfaatan Predictive Analytics Untuk Peramalan Kebutuhan Barang Habis Pakai Pada Sistem Manajemen Inventaris Terpusat Klinik Multi-Cabang. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5027221011-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Klinik Gigi ABC yang memiliki enam cabang operasional menghadapi tantangan dalam manajemen inventaris akibat sistem manual yang tidak terintegrasi, menyebabkan ketimpangan distribusi stok antar-cabang dan pengadaan barang yang tidak efisien. Pendekatan distribusi seragam tanpa mempertimbangkan pola tindakan medis spesifik tiap cabang mengakibatkan penumpukan di satu lokasi sementara cabang lain mengalami stockout yang menghambat pelayanan medis. Penelitian ini mengembangkan sistem manajemen inventaris terpusat berbasis predictive analytics yang memanfaatkan data historis tindakan medis untuk meramalkan kebutuhan Barang Habis Pakai (BHP) secara akurat. Metode yang diterapkan adalah perbandingan tiga algoritma machine learning, yaitu Linear Regression, XGBoost, dan Long Short-Term Memory (LSTM). Evaluasi menggunakan metrik Weighted Mean Absolute Percentage Error (WMAPE), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), dan koefisien determinasi (R²) yang divalidasi melalui walk-forward validation sebanyak 48 langkah. Hasil pengujian menunjukkan XGBoost terpilih sebagai model final dengan WMAPE 17,90%, MAE 39,48, RMSE 51,44 dan R² 0,24. Prediksi volume tindakan medis dikonversi menjadi kebutuhan BHP melalui mekanisme Bill of Materials (BOM) dengan penerapan buffer 12%, safety factor 1,10, dan safety stock dua minggu. Sistem ini berkontribusi pada literatur penerapan machine learning dan BOM dalam manajemen logistik kesehatan di Indonesia.
======================================================================================================================================
ABC Dental Clinic, which operates six branches, faces challenges in inventory management due to a manual, non-integrated system that causes uneven stock distribution across branches and inefficient procurement. A uniform distribution approach that does not consider the specific medical service patterns of each branch leads to overstocking at one location while other branches experience stockouts that disrupt medical services. This research develops a centralized inventory management system based on predictive analytics that utilizes historical medical service data to accurately Forecast the demand for Medical Consumables (Barang Habis Pakai/BHP). The method applied is a comparison of three machine learning algorithms, namely Linear Regression, XGBoost, and Long Short-Term Memory (LSTM). Evaluation employs the Weighted Mean Absolute Percentage Error (WMAPE), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and coefficient of determination (R²), validated through walk-forward validation across 48 steps. The results show that XGBoost was selected as the final model, with a WMAPE of 17.90%, MAE of 39.48, RMSE 51.44 and R² of 0.24. Predicted service volumes are then converted into medical consumable requirements through a Bill of Materials (BOM) mechanism, applying a 12% buffer, a safety factor of 1.10, and a two-week safety stock. This system contributes to the literature on the application of machine learning and BOM in healthcare logistics management in Indonesia.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Manajemen Inventaris, Predictive Analytics, XGBoost, Walk-Forward Validation, Bill of Materials, Inventory Management, Predictive Analytics, XGBoost, Walk-Forward Validation, Bill of Materials.
Subjects:	Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) Q Science > QA Mathematics > QA336 Artificial Intelligence T Technology > T Technology (General) T Technology > T Technology (General) > T174 Technological forecasting T Technology > T Technology (General) > T57.5 Data Processing
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology
Depositing User:	Monika Damelia Hutapea
Date Deposited:	20 Jul 2026 04:24
Last Modified:	20 Jul 2026 04:24
URI:	http://repository.its.ac.id/id/eprint/135678

Actions (login required)

View Item