New Open-Access Dataset for Marine Engine Fault Diagnostics Released

Ahmad BahooToroody, Oleksiy Bondarenko, Mohammad Mahdi Abaei, Niki Yoichi, Enrico Zio· July 23, 2026 View original

Summary

Researchers have released the Marine Engine Fault Dataset, an open-access collection of multi-sensor time-series data from a three-cylinder marine diesel engine. This dataset includes both reference performance and controlled fault scenarios, providing a valuable benchmark for developing predictive maintenance and anomaly detection models in maritime machinery.

A new open-access dataset, the Marine Engine Fault Dataset, has been made available to the public, addressing a significant scarcity of controlled fault experiment data for marine engine predictive maintenance. This dataset was collected from a turbocharged, intercooled three-cylinder marine diesel engine operating on a testbed. The experimental campaign included both reference performance measurements across various load ranges and scenario-based tests where five distinct anomaly classes were physically introduced. These anomalies, affecting subsystems like the cooling-water pump, air filter, and injection valve, allow for direct comparison between baseline and fault-affected behavior. The dataset comprises multi-sensor time-series data, offering a robust resource for research in anomaly detection, fault diagnosis, and degradation modeling for maritime applications.

Why it matters

This dataset provides a crucial resource for developing and validating AI-driven predictive maintenance solutions for the maritime industry, potentially reducing downtime, improving safety, and optimizing operational costs.

How to implement this in your domain

  1. 1Download and explore the Marine Engine Fault Dataset for potential use in predictive maintenance model development.
  2. 2Develop and benchmark anomaly detection algorithms using this new, controlled dataset.
  3. 3Collaborate with maritime industry partners to apply insights from this data to real-world engine monitoring.
  4. 4Integrate fault diagnosis models trained on this data into existing or new condition-monitoring systems.

Who benefits

MaritimeLogisticsIndustrial IoTManufacturingEnergy

Key takeaways

  • A new open-access dataset for marine engine fault diagnostics is now available.
  • It includes controlled fault scenarios and multi-sensor time-series data.
  • The dataset is valuable for developing predictive maintenance and anomaly detection models.
  • This resource can significantly advance AI applications in the maritime industry.

Original post by Ahmad BahooToroody, Oleksiy Bondarenko, Mohammad Mahdi Abaei, Niki Yoichi, Enrico Zio

"arXiv:2607.19444v1 Announce Type: new Abstract: Open-access datasets for marine-engine predictive maintenance remain scarce, particularly those from controlled fault experiments with documented operating conditions, subsystem-level interventions and system-level measurements. Thi…"

View on X

Originally posted by Ahmad BahooToroody, Oleksiy Bondarenko, Mohammad Mahdi Abaei, Niki Yoichi, Enrico Zio on X · view source

Want to go deeper?

Turn these trends into skills with Learnijoy's hands-on AI & tech courses.

Explore courses