English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The integration of sensorized vessels, enabling real-time data collection andmachine learning-driven data analysis marks a pivotal advancement in themaritime industry. This transformative technology not only can enhance safety,efficiency, and sustainability but also usher in a new era of cost-effectiveand smart maritime transportation in our increasingly interconnected world.This study presents a deep learning-driven anomaly detection system augmentedwith interpretable machine learning models for identifying performanceanomalies in an industrial sensorized vessel, called TUCANA. We Leverage ahuman-in-the-loop unsupervised process that involves utilizing standard andLong Short-Term Memory (LSTM) autoencoders augmented with interpretablesurrogate models, i.e., random forest and decision tree, to add transparencyand interpretability to the results provided by the deep learning models. Theinterpretable models also enable automated rule generation for translating theinference into human-readable rules. Additionally, the process also includesproviding a projection of the results using t-distributed stochastic neighborembedding (t-SNE), which helps with a better understanding of the structure andrelationships within the data and assessment of the identified anomalies. Weempirically evaluate the system using real data acquired from the vessel TUCANAand the results involve achieving over 80% precision and 90% recall with theLSTM model used in the process. The interpretable models also provide logicalrules aligned with expert thinking, and the t-SNE-based projection enhancesinterpretability. Our system demonstrates that the proposed approach can beused effectively in real-world scenarios, offering transparency and precisionin performance anomaly detection.

Enabling Smart Retrofitting and Performance Anomaly Detection for a Sensorized Vessel: A Maritime Industry Experience