Shipping: tighter fuel rules, unused onboard control and data

Process control on board and at project sites lacks control engineers

What happens. Marine contractors run plants on ships and on project sites far from the shore: ballast and cooling water, power management, handling equipment. The plant is commissioned in a hurry, with the settings of the supplier. The few engineers who can tune a loop are in the office or on another vessel.

What it costs. Plants that run with oscillating loops or with large reserves, waiting time for the vessel or the project and trips of the specialist that cost more than the fix.

Why the usual fix fails. Training the crew in loop tuning is slow, and the crew changes. A permanent control engineer for each project is not available. Remote support by phone fails when nobody sees the data.

How a fixed-price remote diagnosis checks it. From recorded trends of the plant (setpoints, valve positions, measured values, alarms) I check which loops cause the problems and what a stable setting would look like. The fixed-price remote diagnosis answers what can be solved remotely, and what needs a visit.

Engine and generator models should define normal behaviour and flag deviations

What happens. Engines and generators on board produce trends of temperatures, pressures and speeds. Alarms trigger at fixed limits, when the problem is already there. A slow change, say of a cylinder temperature compared with the load, goes unnoticed because the load changes all the time.

What it costs. Unplanned repairs at sea or in port, maintenance by calendar instead of by condition and downtime that comes as a surprise.

Why the usual fix fails. Fixed alarm limits are wide because the load varies, and then they come too late. A comparison with the last voyage ignores the sea state and the cargo. A monitoring service built on generic statistics produces many alarms that crews learn to ignore.

How a fixed-price remote diagnosis checks it. From engine data over some voyages I check how well a model describes the normal behaviour at any load and which deviations it shows early. The fixed-price remote diagnosis answers whether a model-based early warning is useful for your engines and how many false alarms to expect.

Background: Computing unmeasured states for control

Fuel rules for ships make every tonne per voyage count

What happens. European rules on maritime fuel now require large ships to lower the greenhouse-gas intensity of the energy they use, with penalties for the shortfall. The ship manager sees a fuel bill and in addition a compliance balance that depends on speed, route, weather and onboard energy use.

What it costs. Fuel cost per voyage, plus penalties or purchases to cover a shortfall. Planners have to decide speed and routing with a second target next to arrival time and fuel.

Why the usual fix fails. Slow steaming everywhere saves fuel and breaks the schedule. A weather-routing service chooses a good route but does not know the compliance balance or the onboard power plant. A spreadsheet at the end of the year shows the result too late.

How a fixed-price remote diagnosis checks it. From voyage logs (speed, fuel use, weather, onboard power demand) I check how much of the use depends on speed choice and onboard energy management, and what an optimisation of both against the rules would involve. The fixed-price remote diagnosis answers whether voyage planning with a model of consumption is useful for your fleet.

Background: Economic predictive control