Hybrid dynamic arithmetic city council optimization for improved rainfall prediction

• Published in: International journal of system assurance engineering and management Vol. 15; no. 7; pp. 3182 - 3192
• Main Authors: Lathika, P., Singh, D. Sheeba
• Format: Journal Article
• Language: English
• Published: New Delhi Springer India 2024; Springer Nature B.V
• Subjects: Algorithms; Arithmetic; Correlation analysis; Effectiveness; Engineering; Engineering Economics; Error analysis; Evolutionary algorithms; Logistics; Marketing; Municipal government; Optimization; Organization; Original Article; Performance prediction; Preprocessing; Quality Control; Rainfall; Reliability; Root-mean-square errors; Safety and Risk; Task complexity; City councils evolution algorithm; Open weather dataset; Dynamic arithmetic optimization algorithm; Correlation matrix; Rainfall prediction
• ISSN: 0975-6809; 0976-4348
• DOI: 10.1007/s13198-024-02324-9

In the meteorological department rainfall prediction is one of the complex tasks because it is directly linked to human life and the Indian economy. There is a significant demand for accurate and effective rainfall prediction methods to make better decisions regarding precautionary measures. To predict rainfall amounts effectively, this study proposed a novel rainfall prediction method named the Hybrid Dynamic Arithmetic City Council Optimization (HDACO) algorithm. The proposed HDACO method is a combination of two algorithms namely the Dynamic Arithmetic Optimization (DAO) algorithm and the City Councils Evolution (CCE) algorithm. The study utilizes preprocessing steps namely data cleaning, filling missing values, and data normalization. After preprocessing, the features closely related to rainfall prediction are selected by the computation of the correlation matrix. Finally, based on the features selected the HDACO algorithm predicts the amount of rainfall. The HDACO algorithm is evaluated using an open weather dataset and the effectiveness of the HDACO algorithm is validated using measures such as rainfall rate, Mean Absolute Error (MAE), coefficient of determination (R 2 ), and Root Mean Square Error (RMSE). As a result, the HDACO algorithm achieved RMSE of 0.272, MAE of 0.184, and R 2 of 0.97 respectively. The performance of the HDACO algorithm is compared with existing methods and the results demonstrate the better performance of the HDACO algorithm in rainfall prediction.