Vessel Intelligence / Seeking pilot partners
Safer operations.
More reliable vessels.
Earlier insight into machinery health. Better-informed decisions about vessel readiness.
We’re seeking vessel owners and operators to test whether everyday machinery data can inform earlier maintenance decisions.
- You contribute
- Access to a suitable vessel, operating data, and maintenance-team feedback.
- We propose
- An initial 90-day baseline, findings reviews, and input into the tools we build.
Scope, costs, data use, and installation are agreed before participation. Safety and reliability lead; cost management is a supporting benefit.
Could your vessel be a fit? ↗Proposed pilot · Illustrative dashboardSample data only. Not measured pilot results.
Safety & reliability
An example of what we’re testing
When a fluctuation
becomes a pattern.
Imagine a machinery signal drifting beyond its usual range at a comparable load. Could that change help your team investigate a developing issue before it affects vessel readiness?
- Normal rangeLearn the vessel’s everyday behavior.
- Persistent changeLook for a pattern worth investigating.
- Maintenance reviewCompare the finding with service history.
Where AI earns its place
A change is a clue.
Your team makes the call.
AI models are being developed to learn your vessel’s normal operating patterns and flag meaningful changes. Your team’s observations help us test whether the findings are useful.
Two different jobs for AI
Anomaly detection: what changed?
Statistical methods and machine-learning models compare machinery signals with the vessel’s baseline. The pilot will test whether those changes provide useful early indications, and how often they create false alerts.
AI explanation: what should we review?
A separate language-model layer is planned to explain flagged changes using engine documentation and maintenance history, and draft a work order. It does not make the anomaly detection decision. People review the findings and decide the action.
Does this control the vessel?
No propulsion control is proposed. The design calls for listen-only engine-network connections and onboard analysis through connectivity gaps. Installation and safeguards will be reviewed for each pilot vessel.
A practical exchange
Your vessel. Your insight.
Help shape a useful tool.
Contribute operating data and maintenance feedback. The proposed partner package gives you a clearer view of your machinery and a voice in what we build.
A machinery baseline
A picture of how the monitored machinery behaves across normal operating conditions.
Findings reviews
Periodic conversations about observed changes, data quality, and what merits investigation.
Input into development
Help shape the reports and maintenance workflows your team would actually use.
Proposed partner package. Deliverables, timing, and costs would be agreed for each pilot.
Start with observation
Log quietly.
Build confidence.
We’re seeking owners and operators who can provide access to a suitable vessel and feedback from their maintenance team. We’ll begin by assessing machinery, available signals, and installation needs together.
Discuss your vessel’s fit ↗Proposed pilot sequence
- Log quietly
Build a baseline over an initial proposed 90 days. No crew-facing alerts.
- Review findings ashore
Compare flagged changes with crew observations and maintenance records.
- Evaluate crew alerts
Consider a rollout only after reviewing usefulness and false alerts.
What would we agree before starting?
Installation responsibilities, cost, timing, partner reporting, and support—as well as data access, retention, and use for model training. These terms are still being developed and would be agreed before participation.
Start a conversation with Arto.
Arto Altincatal · Data Systems & AI
Discuss your vessel, available machinery data, and maintenance workflow. Start by exploring pilot fit and the questions to resolve before participation.