A Roadmap from Digital Foundations to Supervised Automation

How does an operator actually get from advisory tools to supervised automation? This final insight in the series sets out the staged path, strengthening data integrity, embedding optimisation in everyday operations, running pilots, scaling with governance, navigating barriers, and tracking progress against a maturity model that points the industry towards a clearer direction of travel.

For many operators, the path forward is a process of staged progression from advisory systems to fully automated control.

This process of progression often starts with data integrity, then moves to embedding optimisation into everyday operations, and then to a point where supervised automation can be introduced in a controlled manner.

Establishing data integrity. Any plan of moving a maritime operator towards the use of automation in propulsion systems should begin with a focus on data. If the operational data used aboard vessels is unreliable, then optimisation recommendations will also be unreliable.

As Dr Timoleon Plessas, Founder at Sealion Engineering, put it, “if one link doesn’t produce the result, the whole chain is useless.” This is why stronger data validation and governance should come first. Without that, companies who move towards automation risk building on unreliable foundations.

Embedding optimisation into operations. Once data is reliable and trustworthy, the next challenge is operational. Optimisation can only deliver value when it is embedded in how vessels are run, rather than being a set of recommendations which sit alongside operations without fully shaping them.

Rather than being a criticism of ship staff, this is a reflection of operational reality. Optimisation only becomes routine when it is supported and followed up through the wider operating structure around the vessel. That process is likely to depend on a closer operational link between ship and shore than many fleets currently have.

Introducing supervised automation through pilots. Once the foundations for automation have been established, moving towards supervised automation can become the next step. However, fleet-wide deployment of autonomous tools should be preceded by a validation period. While scaling across the fleet can unlock the greatest value, rollouts are most effective when tested in controlled conditions with clear operating boundaries and measurable outcomes.

Scaling with governance. If a pilot project is successful, the question is how to scale it. This is where the importance of governance is central. Scaling automation is not just about deploying the same system across more ships and fleets.

Barriers and mitigation strategies. Often barriers to implementation are likely to be technical, cultural, and organisational.

A maturity model for propulsion automation. Together, these stages show how operators can gradually move from basic reporting and advisory tools towards more advanced optimisation and, eventually, supervised automation.

The final stage is when this is no longer limited to a handful of pilot projects or optional tools, but becomes part of the way the fleet operates more broadly. Not every operator will move at the same speed, and getting there will take time. Even so, the direction of travel is becoming clearer.

Put simply, fragmented reporting and weak data limit what optimisation can achieve. Stronger data and better-integrated workflows enable more advanced optimisation. From there, supervised automation can help close the gap between what is planned and what is actually delivered onboard.

In practical terms, the route to better performance is already visible. The industry does not need to wait for some distant autonomous future. It needs to strengthen its data foundations, make optimisation part of everyday operations, test automation in controlled conditions, and then scale what proves effective.