A practical composable data analytics experience for complex subjects under data problems

Happy new year! As we step into 2026, our journey continues: we stay focused on (i) modernization of existing services and systems, (ii) solution development for process/automation/intelligent monitoring, and (iii) data analytics/ML. Throughout the journey, solving complex problems has always been challenging, but also rewarding. Each journey is an opportunity for us to work alongside our customers and partners.

Last year, we had chance to work in an exciting problem: energy analytics in a telco network. The existence of data problems and the complex structure of the base station drives the need to have appropriate analytic approaches. Therefore, we proposed a composable methodology which presented in our recent released article:

Composable Analytics for Complex Domain Subjects: The Case of Energy Analytics in Mobile Networks.

“Abstract—Creating suitable data products characterizing real-world industrial subjects often faces various challenges. We have a compound issue due to multitude of measurements about the analytics subject and, at the same time, the lack of required data of the subject’s constituting components. This work discusses a practical composable data analytics experience where data associated with a system of Base Transceiver Stations (BTSs) is analyzed to produce data products. We describe the multi-facet data problems coupled with the complex structure of the subject. Our approach leverages the composable data analytics to achieve data products of meaningful insights for the requirements from the mobile network operator via a use case of profiling of BTSs configurations for energy efficiency management.”

Enjoy reading!