AI-DRIVEN COST ESTIMATION FOR MARINE STEEL IN TUGBOAT CONSTRUCTION

Abstract

Accurate cost estimation for new tugboat construction is a critical challenge in the shipbuilding industry, where conventional methods based on historical data fail to account for the real price at real time because of volatility of key materials. The paper addresses this gap by developing a data-driven framework for marine steel, a primary cost driver, which leverages a specialized Artificial Intelligence (AI) model. The research aims to enhance the application of advanced digital technologies in shipbuilding to support design, assembly, and logistical planning processes within shipyards. The framework, built using Python, features a Graphical User Interface (GUI). Through this interface, the system accepts internal project data from the shipyard and integrates it with live market data synthesized from the real-time web aggregation capabilities. This combined dataset, including steel plate specifications, supplier origin, certification, and logistics, is processed to calculate the true landed cost and automatically generate detailed cost reports. The model was applied to a tugboat construction case study in the Middle East. The AI-driven framework demonstrated a significant improvement in estimation accuracy compared to conventional methods, where the program can save about 20% for steel plates and 13% for steel stiffeners compared with the actual case study. It successfully enabled dynamic cost benchmarking of suppliers all over the world, while substantially reducing the time required to generate detailed steel cost breakdowns. The findings confirm that AI- enabled integration with e-marketplace data provides a practical solution for enhancing budgetary reliability, strengthening procurement strategies. Integrating these technologies contributes to more efficient review processes and reducing rework rates and achieving time and cost savings.