Prediction Of HIV Case Spread In Sukabumi Using Single Exponential Smoothing Based On Grid Search Optimization

anggi agusti (Information Systems, Computer Engineering and Design, Nusa Putra University)
Cecep Warman (Information Systems, Computer Engineering and Design, Nusa Putra University)
Rieska Rahayu Ayuningsih (Information Systems, Computer Engineering and Design, Nusa Putra University)

International Journal of Health, Engineering and Technology
Vol. 5 No. 4 (2026): In Press

Human Immunodeficiency Virus (HIV) remains a health problem of concern in Sukabumi City. The increase and fluctuation of HIV cases require control efforts through data-based planning. This study aims to analyze the development pattern of HIV cases, apply the Single Exponential Smoothing (SES) method, optimize the alpha parameter using the Grid Search method, and evaluate the accuracy of the prediction model. The data used were monthly HIV cases in the Sukabumi City Health Office area from January 2024 to December 2025. Forecasting was conducted using the SES method and implemented using Microsoft Excel and Python programming language. The alpha parameter was optimized using Grid Search with values ranging from 0.01 to 0.99. Model accuracy was evaluated using Mean Absolute Deviation (MAD), Mean Squared Error (MSE), and Mean Absolute Percentage Error (MAPE). The results showed that the optimal alpha value was 0.06, with a predicted number of HIV cases in January 2026 of 18.016 cases. The model obtained MAD of 2.597, MSE of 10.383, and MAPE of 13.71%. Based on the MAPE value, the model was categorized as good forecasting and can support planning and decision-making related to HIV case management.