From Data Analytics to Data Management

Written by Timo Latvala
CSO & Director, Embedded Solutions

In the previous blog post, we discussed the benefits of proactive maintenance in railway infrastructure. Once the possible malfunctions are identified in advance from the data mass, maintenance can be scheduled precisely for those moments when it is really needed. This saves time, effort, and resources. Once the data has been accumulated, the next step is to leverage it using data analysis methods.

Our client, VR FleetCare, is a railway maintenance company that provides technological solutions for the proactive maintenance of railway infrastructure through SmartCare services. The switches in the railway networks, with which trains and trams are directed from one rail to another, are currently maintained according to a maintenance schedule. In order to further ensure the fluency and safety of rail traffic, VR FleetCare joined us in our customer-centric development toward proactive maintenance. VR FleetCare has measured the power data from the switches, and we put that information into use with data analytics.

Data Analytics and Design Puzzle

When there is a lot of data, dividing it into components helps to control the whole. We worked on the solution with our Design Puzzle, an agile method which helps structure even large amounts of data. In a step-by-step process, the work is broken down into smaller more manageable parts. The close cooperation with VR FleetCare ensures a thorough assessment of their needs and the utilization of customer knowledge. The work is adjusted to the resource and time frame needs of the client, and iteratively one area at a time is moved toward the solution.

Each project has its own specific characteristics. In the case of VR FleetCare, the data lacked information on what kind of switching each sample depicted. Millions of data samples had to be structured but no manual work was needed. We took advantage of SOM visualization, a method of data analysis with the strength of pre-sorting and visualizing data on a map. Even from the unclassified data mass, distinguishing and unifying features are highlighted, and attention can be focused precisely on those interfaces where normal samples encounter changes that predict switch malfunction. Based on this critical information, rules can be formed to predict the need for maintenance.

Data analytics leads to data management

The understanding and utilization of data enables predictive maintenance. At the same time, it opens the door to new ideas and business opportunities hidden inside the data mass.

Information management means decision-making based on up-to-date information. Every organization faces many important operational decisions every day. An up-to-date situational picture and an understanding of the consequences of decisions support thoughtful decision-making. When data generated in different systems proves its full potential through data analytics, the understanding of the data is deepened and the quality of business decision-making is improved, whether the decision is about proactive maintenance or something else.

Data can also be used in the development of new business opportunities. In this case, it is crucial to understand which conclusions from the data are so valuable that you can and should build a business on them. Data analytics methods, as well as the customer and his knowledge of the field, play a key role in the development work. Huld’s Design Puzzle is a functional way to reflect on possibilities systematically from the customer’s individual perspective and outlook.

If you are interested in new technologies for railway infrastructure, join us at VR FleetCare’s SmartCare seminar on 30 March 2022. Timo Latvala, our Business Unit Director, will participate in the closing panel discussion on new technologies as part of railway infrastructure maintenance. The discussion will focus on the challenges and opportunities brought by technology as part of the processes. Read more and sign up for the virtual event here. The event is in Finnish.