New ECO Insight hull degradation computation method goes beyond ISO 19030 standard
Hull performance determination
Hull Degradation is one of the most important vessel performance levers in shipping, but very often overlooked. But why? It is so hard to compute it properly! Only a proper computation allows a shipping company to go away from current practice, which is either trouble shooting (e.g. after a charter party consumption claim) or do simple planned maintenance with regular cleaning of hull and propeller, no matter whether it is needed. Usually the speed-consumption development is monitored and not the hull index itself.
For this reason industry experts, including ourselves, have developed an ISO norm (ISO 19030) to describe how to come to proper results. This norm serves as a good basis for assessing the condition of the hull based on collected data. Having more than 2.000 vessels using DNV GL’s ECO Insight Performance solution we have started seeing ISO 19030 shortcomings though:
- the standard method (ISO 19030 part 2) needs high frequency auto logging data, which most vessels do not have today
- the current computation in most cases shows a dependency on speed, the faster the vessel sails, the lower the computed degradation
- the computed hull condition is also changing with larger weather changes, which is also not the case in the physical world
Overall ISO 19030 is meant to measure performance improvements not the absolute level of performance and that is not very satisfactory. You will see some X % improvement of performance after dry dock, but do not know whether you came back to 100% performance.
To understand the significant improvements we have recently implemented, we will explain the basic methodology for hull performance assessment in detail. As a main KPI you can start looking at the power increase required to move the fouled ship forward or at speed loss when keeping constant power. Both look at the same speed-power relationship. ISO 19030 uses speed loss, we work with both but prefer power increase, due to the exponential relationship between speed and power. To come to a hull degradation assessment (I) you measure power, (II) then correct it for environmental factor (weather, water, temperature etc.) and compare this value (III) to the ideal needed power at the same Speed, draft and trim as the measured power.
To get an easy KPI, we bring the measured and ideal power in relation that a clean hull with new paint is approx. 100% hull performance and a degraded one falls below e.g. 80%.
Doing hull degradation assessments for so many ships on a continuous basis, we not only recognized above mentioned issues, but also had the capabilities to solve them as the first movers in this subject.
- To be able to use also low frequency data we have introduced snapshot reporting to the industry, which means that not average noon data are taken but speed, draft, trim, power, weather data from the same point in time. Also we have applied smarter filters, that are not reducing the amount of available measurements so much as in the very strict ISO norm. Example: our wind filter takes into account where the wind is coming from and its real influence on the vessels, so we also look at the superstructure geometry.
- To get away with the speed dependency you could model all the difficult effects like drifting at slow speeds, other Propeller inflows etc. This is possible but very costly. Having so many real performance data from actual sailing vessels at hand, we could use a modern machine learning approach to solve this issue.
- To reduce the weather dependency we used a similar approach. The ISO 19030 wind correction factor uses the ISO standard for sea trials, which have been introduced by ship yards and tend to over correct high winds. This is an obvious advantage for the ship yard. Here we have also used machine learning on our whole 2.000 vessel data base to come to more realistic wind corrections.
At the end we would like to show hull degradation as an absolute level and as degradation trend as shown in below picture, which will be re-set by a dry-docking or cleaning event.
Above explained development work took about 1 year. How do we know our new approach is better than the existing ISO one?
- We have significantly reduced the scatter of our results, which is a statistical measure for confidence. Also speed and weather dependances which we saw before have disappeared.
- We continue comparing computed hull performance indexes with visual inspections, e.g. from dry docks which we get from customers and our coating partners all over the world. So our method also allows for assessment of the absolute level of hull performance.