Completed
Closed. Monitoring 22 vessels across two fleets.
KeelsonThis project has been verified
Predictive maintenance for marine vessels, built from real engine telemetry.
$560,000 raised so far, confirmed by both parties (82%)
Details
- Category
- AI / Machine Learning
- IP type
- Patent
- Stage
- Revenue
- Ask type
- Fixed amount
- Listed
- Sep 16, 2026
- Views
- 0
Description
Keelson ingests marine engine and hull sensor telemetry to predict maintenance needs weeks ahead of failure, helping fleet operators schedule dry-dock time around predicted issues instead of reacting to breakdowns at sea.
Problem
Marine engine failures at sea are far more costly than a scheduled repair, but most fleets still run reactive rather than predictive maintenance.
Solution
A failure-prediction model trained on real marine engine and hull telemetry, integrated with existing onboard sensor systems.
Target market
Mid-size commercial shipping and fishing fleet operators.
Business model
Per-vessel monthly monitoring subscription.
Investments
Wei Zhang invested $380,000
7% equity
Kenji Watanabe invested $180,000
3% equity
Discussion (1)
Wei Zhang
Scheduling dry-dock time around predicted issues instead of reacting at sea is a real operational and cost win.
Bjorn EriksenVerified account
Innovator·🇳🇴 Norway
Marine engineer, ran maintenance for a mid-size shipping fleet.