Need help to prioritize correctly?
CUSTOMER-CENTRIC SERVICE PERFORMANCE
MACHINE LEARNING SETS AUTOMATIC THRESHOLDS
REAL-TIME PLOTTING SERVICE METRICS ON GEO-MAP
INTUITIVE ANALYSIS OF PROBLEM SEVERITY & SIZE OF IMPACT
NocMap – the EKG of your network
When you’re working in the Network Operations Centre (NOC) / Service Operations Centre (SOC) it is absolutely critical to always be on top of current service performance. Systems often give data per node making it difficult to see the full picture and what service users actually experience. If certain issues are not captured in real-time, it can have a huge impact on service performance. Besides seeing all the specific details in the network, it is also important to have an overall view. This could be, seeing what effect a storm or fire has on a large area or estimating how many users are affected by an issue.
In these situations it is also important that everyone has the same view and sees the same trends. Most importantly however is that all this information must be delivered instantly. If a major issue occurs in the network, calculated KPI:s and alarms cannot appear late when customers have already been affected and the problem may have grown.
NocMap gives an overview of the entire network where service performance is presented in real-time. You can zoom in on certain areas and sites for more detailed information and graphs can be viewed from network- down to cell level where you can further drill down to see specific events. Machine Learning is used to learn network behavior and set relative alarm thresholds to pick up any deviations in service performance.
Machine Learning is used to learn network behavior
The AI platform makes NocMap self-configuring on setup and the application will then learn the network behavior in order to pick up abnormalities based on current and past performance. If, for example, a site is placed next to a tunnel where there is no coverage, NocMap will learn that KPI:s for this site will be different compared other sites when triggering alarms. Machine Learning is used to learn the network behavior and set relative alarm thresholds to pick up any deviations in service performance.
Benefits from Machine Learning
The logic of handling service trends and patterns for a network can be done by a person. However, if a network has tens of thousands of trends and patterns to be watched in real-time times thousands of cells, the number of extra staff that would be required becomes unrealistic. This is why NocMap application uses Machine Learning for self-configuration, threshold handling and automation of tedious tasks.
Artificial Intelligence is a supplement to human insight, not substitute
- Edsger W. Dijkstra
A complete overview of your network from an aggregated to a very detailed view in real-time
NocMap gives an overview of the entire network where service performance is presented in real-time. Colors for alarms let you see instantly if something has a bigger impact on the network, (e.g. bad weather coming in) and how many users are affected. Map can be zoomed in to see performance per cluster or, per individual site. By clicking a site, detailed information appears. Historical data can be selected, otherwise it is presented in real-time.
Real-time KPI graphs
Detailed real-time KPI graphs can be viewed from network level down to individual cells. From graphs you can drill down to see specific events. Historic graphs can also be presented.
ALARMS & AUTOMATION
Alarms to 3rd party systems/individuals.
Machine Learning will set the alarm thresholds.
Plotting service quality metrics on geographical map for quick response and faster resolution-time
Indicate problem severity and size on service / subscriber – e.g. during an outage and helps to prioritize correctly
Telenor Sweden used NocMap during successful HLR migration
Telenor Sweden did an HLR (Home Location Register) migration as part of modernizing their core network. The two existing HLR:s were to be consolidated to a new one, with minimum customer impact. NocMap was used during the migration to monitor the service performance of the network in real-time.
6M SIM cards moved in HLR migration
14h to successfully migrate one HLR
Real-time monitoring of service performance during migration