IrelandLogistics and transport2025 – ongoingArdmore Shipping
Hybrid machine-learning vessel performance models and voyage optimisation
Ardmore Shipping said machine learning raised usable vessel data by 77% across its 22 tankers
Problem
Ardmore needed reliable vessel performance analysis across its fleet of 22 tankers, but much of the raw onboard data could not be used. Its fleet performance manager also warned that 30% of the shore workforce was expected to retire within a decade.
Approach
Ardmore combined traditional grey-box ship models with machine learning in a hybrid model, using Albis performance monitoring systems. It used DeepSea Technologies' Pythia system for voyage speed optimisation. Building a ship-specific model took one or two months.
Results
Takeaway
Better data handling gave Ardmore more sea-days it could analyse. The figures were presented by the company at a conference and not independently checked.
Categories
- Industry
- Logistics and transport
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and MLOptimization
- Adoption
- Vendor solution
- Evidence
- Company's own claim
Sources
Source summaries
Report on a talk by Ardmore's fleet performance manager at the International Chemical and Product Tanker Conference 2025. It describes the hybrid model, Albis monitoring and DeepSea Pythia, and gives the 77% and 30-31% figures.
Full details are in the original sources.
