Reference Case
Client profile
Client: Nokia
Industry: Telecom
Process: Operations
Nokia is a leading provider of mobile network infrastructure, software and services, powering communication networks around the world.
Anomaly detection solution to prevent chipset failure in telecommunication network systems
This project was a successful milestone for our Nokia team in diving deeper in successful cooperation between data scientists and product experts.
The challenge
Nokia sells telecommunication network systems, that are full of sensor-equipped chipsets and regulate operating temperature using fans. Failing to detect broken chipsets and failure of the systems can result in costly system maintenance and replacements that include repair and travel costs, with later detection leading to even higher costs. As a result, it is crucial to detect any small degradations or deviations from standard operating behavior (so-called anomaly detection) before it is too late.
The solution
To address Nokia’s challenge, Bright Cape developed a solution that involved modeling the impact of exogenous factors, like power consumption, fan speed and ambient temperature, on the operating temperature to gain an understanding of the expected operating state of chipsets. A proof of concept of an automated anomaly detection solution was subsequently developed. This model indicates deviations from the expected operating state before chipset failure thresholds are reached, aiming to prevent costly maintenance and replacements. At Nokia’s request, Bright Cape developed a modular and customizable development framework that serves as the foundation for the first anomaly detection pipelines and facilitates future development. Moreover, a graphical representation of complex dependencies was provided, which empowered Nokia to find root causes of deviating behavior.
The impact we created
By creating the first version of an anomaly detection system and developing a modular framework for its implementation, we enabled Nokia to take the first steps towards predictive maintenance of their telecommunication network systems. The developed framework provided Nokia’s data scientists with the flexibility to update existing pipelines and develop new ones, paving the way for continuous improvement in their anomaly detection capabilities. Our Bright Cape solution has the potential to save Nokia significant costs and increase the overall reliability and efficiency of its telecommunication network systems, making it a valuable asset in their ongoing efforts to deliver high-quality services to their customers.
Results
Automated anomaly detection
Based on continuously monitored aspects
Insights and understanding
In expected operating behavior
Modular anomaly detection framework
Allowing for extensions in the future
Other projects
Case Study
Creating insights in machinery performance & enabling failure analysis
Client: Settels Savenije Industry: High-Tech Process: Reliability Testing of Complex Machinery
Case Study
An optimization power play: enabling a tomato grower reduce energy costs by smart allocation of energy usage and production
Client: Vereijken Kwekerijen Industry: Agriculture Process: Operations
Tell us your challenge.
We are here to help you.
Get in touch with our experts
