New insights show the value of open ocean data
We know that we must take better care of our planet and the oceans. And to get technological advances in these fields, we need to take better care of all the data describing them. But why, and how? As engineers and data scientists, we have to show that all this noise about digitalization and data, it’s not all fluffy business talk. We can drive more impact from data and analytics if we can show insights through good storytelling. Last November an interesting story on data usage practically sank onto our desks. As you might have read in the news the previous months, one of Norway’s newest and most modern frigates “Helge Ingstad” grounded and sunk after a collision with a tanker outside Bergen. Although some ripples of information leaked out, perhaps the truth of what happened will remain underneath the sea forever. But based on new analytics developed together with the Norwegian Coastal Administration, we are using data to understand how accidents like these can be avoided in the future.
There are many reasons why this case is interesting, but for data scientists and subject matter experts working with ocean data analytics here at DNV GL, there’s one in particular. Together with the Norwegian Coastal Administration, some of our analysts from the Maritime Advisory unit have done some really clever exercises to make a risk model of the Norwegian coastlines. Combining a myriad of data from high fidelity AIS data (ship positioning data) to weather data to information about cargo and accident statistics, our analysts have come up with a model that assesses risk over the whole Norwegian coastline. This will be used by the coastal authorities for better monitoring systems, responding more accurately to needs and enable better planning around transport and risk mitigation. The system can easily be expanded to cover other sea areas vulnerable to impacts from shipping activities, e.g. The Baltic Sea, EU waters and so forth.
The Norwegian Coastal Administration and DNV GL started this project in 2016 to develop a model for automated calculation of risk related to maritime traffic. The reasons for initiating the model development was the increasing complexity and change of ship traffic, lack of transparency to the previous decision models and costs of repeated manual calculations. Furthermore, cloud-providers like Microsoft/Azure have enabled the sufficient amount of data ingest and processing power so that we can utilize high resolution AIS data, shore and satellite based, to develop more dynamic and accurate risk calculation models than before.
The project has delivered a risk calculation model called AISyRisk, including a long-term data collection on probability of ship accidents and consequences for fatalities and oil spills for the sea areas under Norwegian interest. The results will be published on a web portal administered by the Norwegian Coastal Administration where aggregated results are publicly available.
The potential in an open industrial data platform
AIS- and metocean data only represent a small part of the datasets we currently offer or plan to offer on Veracity, DNV GL’s industrial data platform. What would an open data platform be without a rich set of high quality, relevant and fit for purpose data? The AISyRisk digital service is one of many examples where the upside is significant if data is treated as an asset. In DNV GL, we know there are many more use cases out there. Our goal is to work with the whole industry to become “data smart” together and make a global impact for a safe and sustainable future. This is why we are working hard on making quality ocean space data available on Veracity. We see how new digital tools in many ways are creating more complexity, and believe we have a role to play in making data accessible, trustable, and usable. To make value from the data, one has to trust the entire data value chain. Without trust in the data, forget trust in analytics. It’s that simple.
In Veracity, we also see that many players each hold pieces of the “data puzzle”, but that without combining, they will never be able to see the complete picture. In the example above, the risk analysis would never have been possible without the gathering of several important data threads from both DNV GL and the Coastal Administration.
Veracity is growing to be one of several significant players within open and proprietary industry data. This means enterprises, authorities, research institutions, NGOs and so on can keep full control, ownership and governance of their data, but that they can offer free or paid access under set terms and conditions in the Veracity Marketplace. Those looking for data, be it for research or commercial purposes, can find/purchase/cooperate on datasets in one place.
By simplifying the exchange of data and making it easier to access data with known quality, perhaps we can prevent minor and major accidents in the future, but also help in solving some of the other important issues facing our industry and most importantly our planet.