Norwegian AI Directory
Description:
Today, 98% of the world's ships are performing daily maintenance routines in a reactive and preventive manner. Reactive maintenance is defined as post-failure repair which introduces high risks of downtime, while preventive maintenance involves predefined maintenance intervals. Such intervals are based on the component supplier's recommendations. Since ship components are operated differently from ship to ship due to unpredictable environmental and operating conditions, they are subjected to random degradation patterns that the intervals fail to detect. Therefore, too much maintenance is performed, as a safety measure, based on the component supplier's conservative assumptions. Accordingly, today’s best practice, of regular maintenance based on predefined intervals, constitutes as much as 14% on average of the ships' total operating expenses.
Today, machine learning can be utilized to analyze the actual conditions of ship components to recommend just-in-time maintenance. The main goal of this project is to convince shipowners that a predictive maintenance system based on machine learning can provide a direct influence on ships' daily maintenance routines to reduce maintenance costs. Our self-learning diagnostics and prognostics system is based on extensive research results from Ph.D. students at the Department of Ocean Operations and Civil Engineering at NTNU in Ålesund. This project will further develop it in a commercial setting. If we manage to capture 10% of the Norwegian maritime market, our customers are estimated to save a total of NOK 388,8 million annually.
Project leader: Andre Listou Ellefsen
Started: 2020
Ends: 2021
Category: Universiteter
Sector: UoH-sektor
Budget: 277297
Institution: Institutt for havromsoperasjoner og byggteknikk
Address: Ålesund