Web application with data centric approach to ship powering prediction using deep learning

• Main Authors: Khairuddin, Jauhari, Abdul Malik, Adi Maimun, Hiekata, Kazuo, Siow, Chee Loon, Ali, Arifah
• Format: Article
• Language: English
• Published: Elsevier B.V. 2022
• Subjects: Q Science (General); QA Mathematics; TJ Mechanical engineering and machinery; TK Electrical engineering. Electronics Nuclear engineering; TS Manufactures
• Online Access: http://eprints.utm.my/104129/1/AdiMaimunAbdul2022_WebApplicationwithDataCentricApproach.pdf; http://eprints.utm.my/104129/; http://dx.doi.org/10.1016/j.simpa.2022.100226
• Institution: Universiti Teknologi Malaysia

This work describes an AI-based web application to predict passenger ship powering requirements. The data centric approach is developed based on the actual passenger ship design data as a design tool to assist naval architects to quickly estimate the ship brake power. It emphasised on the preliminary design stage to minimise design tasks and laborious calculations. Based on the study, it is observed that the model shows good agreement to the existing empirical method results with 10% mean absolute errors. Significantly, this presents the approach ability to facilitate faster and effective preliminary design, and scalability for large and complex systems.