Analisis Peramalan Kebutuhan Obat Poli Jantung Di RSUD Dr Soetomo Menggunakan ARIMA dan Long Short Term Memory

Aninda, Ranti (2026) Analisis Peramalan Kebutuhan Obat Poli Jantung Di RSUD Dr Soetomo Menggunakan ARIMA dan Long Short Term Memory. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043221069-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

RSUD Dr. Soetomo sebagai rumah sakit rujukan nasional terbesar di Indonesia Timur menghadapi tantangan dalam manajemen stok farmasi di Pusat Pelayanan Jantung Terpadu (PPJT). Tingginya fluktuasi jumlah pasien sering mengakibatkan kekosongan stok obat vital seperti Asam Asetil Salisilat, Furosemid, dan Spironolactone, yang berdampak negatif pada pelayanan pasien. Penelitian ini bertujuan untuk membandingkan kinerja metode peramalan Autoregressive Integrated Moving Average (ARIMA) dan Long Short Term Memory (LSTM) guna mendapatkan metode terbaik untuk perencanaan pengadaan obat yang efisien. Data yang digunakan adalah data kebutuhan mingguan obat Asam Asetil Salisilat, Furosemid, dan Spironolactone pada periode Januari 2023 hingga Juni 2025. Hasil peramalan kebutuhan tiga jenis obat di RSUD Dr. Soetomo menunjukkan metode ARIMA secara konsisten lebih akurat daripada metode LSTM untuk semua kasus. Model ARIMA(1,1,1) terpilih sebagai model terbaik untuk meramalkan kebutuhan Asam Asetil Salisilat (MAPE 76,502%) dan Spironolactone (MAPE 64,261%). Sementara itu, model ARIMA(0,1,1) paling akurat untuk obat Furosemid dengan MAPE 19,198%. Berdasarkan nilai error (MAE, MAPE, RMSE) yang lebih rendah, model ARIMA secara konsisten mengungguli LSTM dalam ketepatan peramalan. Dengan demikian, metode ARIMA direkomendasikan untuk perencanaan stok obat-obatan kritis ini di RSUD Dr. Soetomo ke depannya.
====================================================================================================================================
RSUD Dr. Soetomo as the largest national referral hospital in Eastern Indonesia faces challenges in managing pharmaceutical stocks at the Pusat Pelayanan Jantung Terpadu (PPJT). The high fluctuations in the number of patients often result in a shortage of stocks of vital drugs such as Acetic Acid Salicylic, Furosemid, and Spironolactone, which negatively impact patient services. This study aims to compare the performance of Autoregressive Integrated Moving Average (ARIMA) and Long Short Term Memory (LSTM) forecasting methods in order to obtain the best method for efficient drug procurement planning. The data used is data on the weekly needs of Acetyl Salicylic Acid, Furosemide, and Spironolactone drugs in the period January 2023 to June 2025. The results of forecasting the need for three types of drugs at RSUD Dr. Soetomo show that the ARIMA method is consistently more accurate than the LSTM method for all cases. The ARIMA(1,1,1) model was selected as the best model to forecast the needs of Acetyl Salicylic Acid (MAPE 76.502%) and Spironolactone (MAPE 64.261%). Meanwhile, the ARIMA(0,1,1) model is the most accurate for the drug Furosemid with a MAPE of 19.198%. Based on the lower error values (MAE, MAPE, RMSE), the ARIMA model consistently outperforms LSTM in forecasting accuracy. Thus, the ARIMA method is recommended for planning the stock of these critical drugs at Dr. Soetomo Hospital in the future.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Asam Asetil Salisilat, ARIMA, Furosemid, LSTM, Peramalan, Spironolactone, Acetic Acid Salicylic, ARIMA, Forecasting, Furosemide, LSTM, Spironolactone.
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD30.27 Business forecasting
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Ranti Aninda
Date Deposited:	13 Feb 2026 06:12
Last Modified:	13 Feb 2026 06:12
URI:	http://repository.its.ac.id/id/eprint/132446

Actions (login required)

View Item