Treat data as an asset and benefit from data smart solutions (part 1)
This is the first part out of a mini-series of blog posts. The first part will discuss the terms “data as an asset” and “data smart”, while the second part will provide an example where we played around with machine learning techniques to potentially automate parts of our software support process.
Becoming “data smart” is at the core of many companies’ strategy going forward towards a digital and efficient future. In DNV GL, we treat data as an asset. When the term “manage data assets” is properly defined, managed, and used, data can be extremely valuable, improving every aspect of every business and leading to opportunities to improve a company’s competitive position. You can read more about data quality in a longer article, treating data as assets, on our web site. It is written by Thomas C. Redman, President of Data Quality Solutions, and goes a bit more into the details.
But, back to the term “data smart”. What is it about and can we describe it more specifically with some examples?
In our world, “data smart” might refer to items (or technologies) such as:
- Big data
- Advanced analytics
- Automation of knowledge work
- Internet of things
If we do it the right way, it will help us to become more efficient, hopefully gain new market shares, retain customers and generate new revenue streams applying new digital business models.
Big data is data sets that are so big and complex that traditional data-processing application software are inadequate to deal with them. Big data challenges include capturing data, data storage, data analysis, search, sharing, transfer, visualization, querying, updating, information privacy and data source. Our industrial data platform Veracity is designed for this purpose.
Advanced analytics, in this context, refers to technologies like artificial intelligence (AI), machine learning (ML) and deep learning (DL). The ideas with AI came as early as the 1950s, but have been accelerating with the achievements within cloud-computing. AI involves machines (computers) that can perform tasks that are characteristic of human intelligence. AI can be put in two categories, narrow and general. A narrow example could be the work DNV GL is doing on image recognition to detect issues in steel constructions, for instance corrosion, cracking or lack of coating. In such cases we create AI-algorithms that are far more specialized so we refer to it as machine learning. If we gather hundreds of thousands of such images and tag them with the correct answer (for instance type of corrosion or coating problem), we might train the algorithms/models to become very accurate. This is what we call deep learning.
Automation of knowledge work refers to automating the type of “boring” repetitive tasks that sometimes are carried out by too qualified people. Take an example of travel expense statements, where an employee would use a lot of time to “tag” each receipt with taxi, hotel, flights and so on, and then enter it correctly into a system. Such tasks could be highly automated by applying image recognition and machine learning. In fact, there are several companies offering this type of services, for instance the start-up Luca Labs in Norway. Another example is the way Iris.ai apply machine learning to scan through thousands of research papers and help researchers to go from a problem to a precise reading list in a very short time.
Connectivity means that everything becomes more connected. We get more standardized communication protocols, faster communication and much more data. There are many examples that could be mentioned here, but I will stick to a few ones. For instance ships are starting to be highly connected, both by automatic identification systems (AIS) and more operation monitoring for machinery, structures and so on. The same applies to wind farms, solar farms, cars, utility grids and much more.
Internet of things is the network of physical devices, vehicles, home appliances, and other items embedded with electronics, software, sensors, actuators, and connectivity which enables these things to connect and exchange data. In our Veracity data platform, we are mainly focusing on industrial use cases. Thus, we regard “things” such as ships, offshore structures, electricity grids, wind farms, solar farms, fish farms and much more. The number of IoT devices increased 31% year-over-year to 8.4 billion in 2017 and it is estimated that there will be 30 billion devices by 2020. Accordingly, terms like big data and treating data as assets will become more and more important.
If you want to learn more about a specific example on how to apply machine learning to relatively simple use cases, you can move on to part 2 of this blog post.