Project Description
AIOPS ARCHITECTURE NETWORK OPERATIONS
TELECOM PROVIDER
BRIEF
Mediaan Conclusion assisted a Dutch and German telecom provider with the implementation of an AIOps platform. This AIOps solution covered a big data environment (data lake, data warehouse, data science), in combination with Robotic Process Automation (RPA). Customers often struggled with complex network monitoring and alert systems, leading to high operational costs. During this project, we supported the customer in setting up a central Big Data environment, now serving as the foundation for AI/ML-driven monitoring, predictive maintenance apps and capacity forecasting.
Our team provided support to the customer in the following areas:
Setting up data architecture, roadmap and aiding in AIOps platform implementation.
Robotic Process Automation (RPA)
- Identifying and implementing use cases for “autonomous” network operations.
- Converting their existing situation (with a large diversity of tooling) into a unified AIOps solution
THE CONCEPT
The idea of the project was to extract as much behavioral network data as possible, such as metrics and logging information from network elements. This is usually done by a wide variety of tools, so-called “stovepipes,” that prevent effective anomaly detection and root cause analysis. To fix this, all this data was collected in one place, called a Big Data Platform. You can then use AI and machine learning to create an intelligent end-to-end monitoring and alerting environment. A system that allows you to identify, classify and even prevent (future) incidents.
CUSTOMER CHALLENGE
Our client was lost in a jungle of monitoring and alerting systems for its network management. This resulted in a variety of problems:
– A very long learning curve for network operators
– It took a very long time to identify the cause of a failure
– It was very labor-intensive to classify network problems by urgency
– All of the above led to very high operational costs
MEDIAAN IN ACTION
By combining knowledge and experience in data architecture, data engineering, data science and RPA, we were able to help the customer step by step in the intelligent automation of their network operations. Our experience in the telecommunications market enabled us to identify and implement relevant use cases for “autonomous” network operations. We also supported our customers to transform their existing situation (with a wide variety of tooling) into an integrated and more homogeneous AIOps solution.
RESULTS
With our expertise in data architecture, data engineering, data science and RPA, we supported our client in the development of a vision and the step-by-step implementation of an AIOps solution. This also involved remediation of the jungle for tooling of network operations and, where necessary, the replacement by state of the art technologies. The result of the project is a central Big Data environment that acts as the foundation for AI/ML-driven monitoring, predictive maintenance and capacity forecasting.
WANT TO SEE MORE?
