The vital difference between unstructured and contextualized data

The buzz around big data is declining. After some disillusionment around its capabilities, there now follows a period where businesses view data utilization as a necessary part of their business models, and as a foundation for digitalization efforts. But for data to bring business value, you need to know what to look for. In the maritime sector, most of the data that is collected today is sensor data from various systems and equipment on board the vessel. Thousands of data tags are collected every minute. These data can be used in software made specifically for optimizing maritime operations, like asset lifecycle management, operational excellence and energy management. But, and this is vital, only if they are contextualized. Allow us to illustrate.

Getting unstructured data from sensors on board a vessel is like looking at a text with thousands of letters in no specific order, no sensible punctuation and no spacing. In other words, useless. What the recipient needs is something that puts the letters, spaces and punctuation marks in order so that the text becomes readable. You need to have it contextualized. The exact same thing is true for data to become valuable.

But data is still flowing to vessel operators in an unstructured form, and there is a good reason why. Data contextualization requires an asset structure, and those can be tricky and expensive to map out on large and complex assets or systems, like vessels. Because no two vessels are alike, you also cannot replicate the work done on vessel A to vessel B. The process does not scale and has therefore been too expensive. And as a vessel operator you need the asset structure to contextualize data, which in turn allows you to utilize value adding software.