Peramalan Kunjungan Harian Instalasi Gawat Darurat Menggunakan SARIMA: Evaluasi Multi-Horizon terhadap Model Pembanding

Fluctuating daily patient visits make it difficult for hospitals to plan staff and facilities. This study develops a Seasonal Autoregressive Integrated Moving Average (SARIMA) model to forecast daily emergency department (ED) visits at a hospital in Indonesia and compares it with four benchmark models. The data comprise 381 daily observations (March 2025–March 2026), split chronologically into training, validation, and testing sets. SARIMA(2,0,0)(0,1,1)₇ was selected based on parameter significance, the Ljung-Box test, and validation RMSE, and was evaluated at horizons of 1, 7, and 14 days. On the validation data, SARIMA achieved a MASE of 0.73, lower than the naive, seasonal naive, day-of-week mean, and Holt-Winters models (0.82–1.16). On the testing data, mean recorded visits fell from 69.9 to 32.3 patients per day, which the hospital attributed to incomplete recording of discharge dates. Under this condition, SARIMA forecasts exceeded the recorded values and were less accurate than the naive and Holt-Winters models at 7- and 14-day horizons (MAE 20.6–22.1 versus 12.2–14.5). On these data, SARIMA was more accurate for short-term forecasting while visit levels were relatively stable but did not adapt to level shifts, so its use requires monitoring, periodic re-estimation, and completely recorded data.

Authors

Institutions

Publication Details

Journal
METIK JURNAL
Published
2026-11-21
DOI
https://doi.org/10.47002/metik.v10i2.1704
Primary Topic
Forecasting Techniques and Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Peramalan Kunjungan Harian Instalasi Gawat Darurat Menggunakan SARIMA: Evaluasi Multi-Horizon terhadap Model Pembanding

Yudhi Fajar Saputra, Syekh Budi Syam, Sri Nur Hidayati, Bayu Pamungkas et al.
METIK JURNAL
Forecasting Techniques and Applications
article

Peramalan Kunjungan Harian Instalasi Gawat Darurat Menggunakan SARIMA: Evaluasi Multi-Horizon terhadap Model Pembanding

Yudhi Fajar Saputra, Syekh Budi Syam, Sri Nur Hidayati, Bayu Pamungkas, Milkhatun Milkhatun, M. Fajar Rizky
article en

Abstract

Fluctuating daily patient visits make it difficult for hospitals to plan staff and facilities. This study develops a Seasonal Autoregressive Integrated Moving Average (SARIMA) model to forecast daily emergency department (ED) visits at a hospital in Indonesia and compares it with four benchmark models. The data comprise 381 daily observations (March 2025–March 2026), split chronologically into training, validation, and testing sets. SARIMA(2,0,0)(0,1,1)₇ was selected based on parameter significance, the Ljung-Box test, and validation RMSE, and was evaluated at horizons of 1, 7, and 14 days. On the validation data, SARIMA achieved a MASE of 0.73, lower than the naive, seasonal naive, day-of-week mean, and Holt-Winters models (0.82–1.16). On the testing data, mean recorded visits fell from 69.9 to 32.3 patients per day, which the hospital attributed to incomplete recording of discharge dates. Under this condition, SARIMA forecasts exceeded the recorded values and were less accurate than the naive and Holt-Winters models at 7- and 14-day horizons (MAE 20.6–22.1 versus 12.2–14.5). On these data, SARIMA was more accurate for short-term forecasting while visit levels were relatively stable but did not adapt to level shifts, so its use requires monitoring, periodic re-estimation, and completely recorded data.

METIK JURNALVol. 10(2)
Widya Gama Mahakam University (ID)
Forecasting Techniques and Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon