Contextualization of data might not be top of mind when starting out with digitalizing vessels. However, if you intend to understand what the operational data you receive is – or even think about comparing vessel data across a fleet – you are completely dependent on data contextualization.
In essence, contextualization is about putting the cryptic data series you get out of an onboard system into a structure you can understand, e.g. giving them context and a place in a hierarchy.
Our experience with contextualization has revealed that many vessel systems produce dataseries with names, tags, that are very har do make practical sense of and that require contextualization. Across a fleet you may have tens or hundreds of thousands of dataseries available. You will often find that the same dataseries (e.g. exhaust gas temperature) will have many different names across the various vessels in a fleet, which makes it all but impossible to compare and use the data efficiently. Any third party application or dashboard is built on the premise of receiving clean and structured data as input. This can only be achieved with proper contextualization.
If you for instance want to compare the data series of “Main Engine Shaft RPM” across a fleet of vessels, you want to ensure that you are comparing apples to apples. This is only true as long as the naming of the data tags have been standardized. The same analysis would be fairly tricky if you for each vessel would need to consider if “ME_RPM”, “ShaftRevolution”, “ME RPM”, … etc would represent the same data series. Not to mention that the unit of the data series may differ from vessel to vessel, as well as the scaling of the value.
Locating the right data series can also be a rigorous task, this is why it makes sense to name all the data series according to a proper asset hierarchy that is humanly understandable, structured and available for drill down. This makes it easier to work with the data. To ensure interoperability, this asset hierarchy needs to follow commonly agreed standards.
The importance of tag standardization and contextualization cannot be overstated, as without it you will find it hard to get value from the data. Even worse, your data use cases will be hard to scale beyond the first vessel you have tested it on, as the complexity rises sharply with the number of vessels involved.
To get contextualization right requires experience and deep domain expertise. Reach out to us, and we are happy to walk you through how we approach contextualization and ensure the availability of meaningful data for your digital initiatives. - Contact us
Siri Bjørkedal Øvregård
Data Engineering and Quality Lead
Siri leads the team responsible for making sure vessel data is accurate, consistent and fit for purpose.
Meet the Raa Labs team →
