Suciana, Nita Yuanita, Alamsyah Kurniawan
Erosion is one of the most common problems in Indonesia's coastal areas. Mangrove forests are a natural form of coastal protection, however planting and establishing mangroves requires temporary protection from waves and currents, allowing mangroves to grow and function optimally. Before conducting field construction, physical model testing is necessary to know the effectiveness of temporary protection systems. The physical model provides water elevation data, with the advancement of technology, which is used to analyze and predict the wave transmission coefficient (Kt). The transmission coefficient is a key factor in evaluating the effectiveness of a breakwater structure. This research focuses on deriving an equation for Kt by analyzing relationships between variables from the physical model of a coastal protection system that includes variations in the weight and slope of geobag structures and different wave data. Data-driven models are increasingly being developed and used, with one such approach being genetic programming (GP). GPTIPS serves as an offline data-driven modeling tool to catch the trend of the wave transmission coefficient and then use it to predict the Kt equation. The results suggest that combining data analysis with GP models enhances forecasting accuracy by identifying significant predictive parameters. It was found that the Genetic programming for the Kt prediction error model can lead to substantial improvements when used as a data assimilation technique to update Kt predictions derived from primary physical models. The results show that the prediction equation of Kt was successfully obtained, with the error prediction formula Kt being less than 5%. © 2025 Institute of Physics Publishing. All rights reserved.
Department of Ocean Engineering, Institut Teknologi Sumatera, Indonesia; Ocean Engineering Program, Institut Teknologi Bandung, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock