La Ode Ichlas Syahrullah Yunus
Department of Mechanical Engineering, Faculty of Science and Technology, University of Sembilanbelas November Kolaka, Jl. Pemuda no 339, Tahoa, Kolaka, Kabupaten Kolaka 93511, Indonesia

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Artificial Neural Network–Based Forecasting for Airside Capacity Planning at a Data-Scarce Regional Airport: Case of Sugimanuru, Indonesia Panci Yocing; La Ode Ichlas Syahrullah Yunus; Rajab Jandipo Kaebansiha
Civil and Sustainable Urban Engineering Volume 6 - Issue 1 - 2026
Publisher : Tecno Scientifica Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53623/csue.v6i1.959

Abstract

Accurate demand forecasting was essential for sustainable airport capacity planning, particularly at small regional airports characterized by volatile and data-scarce traffic patterns. This study developed an Artificial Neural Network (ANN)–based forecasting framework to predict passenger demand and aircraft movements at Sugimanuru Airport, Indonesia, and to translate projections into airside capacity planning requirements. Researchers normalized historical operational data from 2017–2025 using the min–max technique before developing a feed-forward backpropagation ANN model. Model performance was evaluated using key indicators—the coefficient of determination (R²), root mean square error (RMSE), mean absolute percentage error (MAPE)—along with a review of residual patterns, benchmarked against linear regression and ARIMA for comparison. The ANN performed notably better, achieving R² > 0.90 while reducing prediction errors well below those of the alternative models. Long-term uncertainties prompted scenario analyses for low, medium, and high growth paths. In the medium-growth scenario, by 2035, passenger numbers could rise to 97,000, requiring 2–3 Code C stands and an apron expanded to approximately 12,000–13,000 m², in accordance with ICAO Annex 14 and FAA AC 150/5300-13A specifications. Overall, the study presented a straightforward, repeatable ANN setup suitable for under-resourced regional airports, highlighting AI’s role in guiding infrastructure development and supporting risk-informed planning strategies.