NEC said Russia-based MegaFon has completed AI technology field tests on its live network to improve the efficiency of planning and maintenance of transport network resources.

NEC engineers have conducted AI tests in MegaFon Ural’s mobile network from October to November 2017. NEC’s AI algorithms carried out analysis of 150 radio links that are critical from a network construction standpoint.

MegaFon officials are confident that the analysis of data with NEC’s AI technologies — NEC the WISE – will assist business decision makers including senior telecom engineers to achieve significant improvements for the planning and operation of transport networks. The aim of this test is to improve customer experience of its subscribers.

Engineers of NEC and MegaFon verified the effectiveness of the two key features, “Demand Forecast” and “Preventive Maintenance,” using NEC’s AI technologies, NEC the WISE.
Demand Forecast, which enables timely planning of transport network resources through accurate prediction of network traffic, achieved over 97 percent accuracy for traffic prediction against actual traffic.

MegaFon, if it decides to deploy AI technology with NEC, will gain from efficient planning of transport network capacity and reduction in expenses for unnecessary infrastructure, cut in losses from untimely procurement of equipment.

The Preventive Maintenance feature provides optimal maintenance schedules according to packet loss prediction. The feature accurately predicts packet loss figures several months in advance for each network device and identifies those that are likely to cause disruptions in services.

MegaFon can utilize the Preventive Maintenance feature, as part of the AI technology, can prioritize the maintenance of potentially faulty devices.

“Our partnership with NEC aligns with our goals of efficiently improving the planning and maintenance of networks, which is becoming increasingly complex,” said Anton Sherbakov, technical director, MegaFon Ural.