Implementasi Reinforcement Learning Dengan Algoritma Deep Q-Learning Untuk Penentuan Reorder Point Optimal Dalam Manajemen Persediaan: Studi Kasus pada PT XYZ

Maulana, Muhammad Wibi (2025) Implementasi Reinforcement Learning Dengan Algoritma Deep Q-Learning Untuk Penentuan Reorder Point Optimal Dalam Manajemen Persediaan: Studi Kasus pada PT XYZ. Other thesis, Institut Teknologi Sepuluh Nopember.

Text Tugas Akhir Final Complete.pdf - Accepted Version Restricted to Repository staff only Download (4MB)

Abstract

Manajemen persediaan merupakan aspek krusial dalam operasional perusahaan yang berperan penting dalam memastikan kelancaran proses produksi dan distribusi sekaligus mengoptimalkan biaya operasional. PT. XYZ, sebagai pelaku industri rokok dengan produk yang bervariasi, menghadapi tantangan signifikan berupa fluktuasi permintaan, sehingga menimbulkan kesulitan dalam merencanakan persediaan secara akurat dan menjaga biaya operasional tetap rendah. Pengelolaan persediaan menggunakan metode konvensional seperti Economic Order Quantity (EOQ) dan Reorder Point (ROP) memiliki keterbatasan dalam menghadapi dinamika pasar dan variabilitas supply chain yang tinggi, sehingga diperlukan pendekatan yang lebih adaptif dan responsif. Seiring pesatnya perkembangan teknologi, metode-metode baru seperti Deep Q-Learning (DQL) muncul sebagai alternatif inovatif yang menjanjikan untuk diterapkan dalam sistem manajemen persediaan. Penelitian ini melakukan perbandingan performa antara metode konvensional dengan metode DQL berdasarkan analisis biaya total manajemen persediaan yang dihasilkan. Hasil penelitian menunjukkan bahwa penerapan metode DQL mampu menurunkan total biaya manajemen persediaan hingga sebesar 20,21% dibandingkan dengan metode EOQ dan ROP konvensional, membuktikan efektivitas pendekatan ini dalam meningkatkan efisiensi pengelolaan inventaris pada kondisi yang penuh ketidakpastian dan fluktuasi permintaan yang tinggi.
====================================================================================================================================================
Inventory management is a crucial aspect of a company's operations, playing a vital role in ensuring smooth production and distribution processes while simultaneously optimizing operational costs. PT. XYZ, as a player in the tobacco industry with a diverse product range, faces significant challenges due to demand fluctuations, which create difficulties in accurately planning inventory and maintaining low operational costs. Conventional inventory management methods, such as Economic Order Quantity (EOQ) and Reorder Point (ROP), have limitations in addressing the dynamic market conditions and high variability of the supply chain, thus requiring a more adaptive and responsive approach. With the rapid advancement of technologies, new methods like Deep Q-Learning (DQL) have emerged as promising innovative alternatives for implementation in inventory management systems. This study compares the performance of conventional methods with the DQL approach based on an analysis of the total inventory management costs incurred. The results demonstrate that applying the DQL method can reduce total inventory management costs by up to 20.21% compared to conventional EOQ and ROP methods, proving the effectiveness of this approaches in enhancing inventory management efficiency under conditions of high uncertainty and fluctuating demand.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Reinforcement Learning, Deep Q-Learning,Reorder Point, Manajemen Persediaan, Optimasi Rantai Pasok
Subjects:	T Technology > T Technology (General) > T174 Technological forecasting T Technology > TS Manufactures > TS155 Production control. Production planning. Production management T Technology > TS Manufactures > TS161 Materials management.
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Industrial Engineering > 26201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Wibi Maulana
Date Deposited:	22 Jul 2025 04:17
Last Modified:	22 Jul 2025 04:17
URI:	http://repository.its.ac.id/id/eprint/120423

Actions (login required)

View Item