Nokia launches upgraded SON software to drive 5G
Nokia cognitive SON removes the traditional operator console and replaces it with an objective driven dashboard, allowing real-time solution deployments and increased productivity. This dashboard allows operations stakeholders, from chief experience officers to market engineers, to easily determine how their objectives are achieved with SON.
Accelerated by machine learning concepts like clustering, classification and reinforcement thinking, Nokia cognitive SON oversees standard and adapted models to automatically detect, categorise and solve network problems that leverage insights to improve the solution itself.
It is expected that this will mean a substantial decrease of manual work and extensive technical analysis that network operators would usually be required to carry out. In addition, by overseeing the workflow from end-to-end and introducing objectives into SON for the first time, Nokia cognitive SON optimises activities with an operator’s end goals in mind.
The software upgrades and enhances automatically through cognitive functionalities without the need for the operator to manually activate the request. The software also provides programmability with Software Development Kit and open APIs to further scale speed and flexibility.
Brian McCann, Chief Product Officer of Nokia Software, said: “This cognitive upgrade to our Nokia SON solution massively reduces the need for manual work and technical expertise when optimising radio networks, allowing us to deliver our promise for a much more efficient and error-free process that will ultimately result in better network quality and reliability. It is a timely product and one that reinforces Nokia’s software innovation leadership.”
Predictive Monitoring for Continuous Operations Management
Unplanned downtime and poor maintenance procedure can cost companies a lot of time and money.
For companies looking to set specific targets for cost efficiency, production output and quality control, the ability to predict how certain variables affect machines can aid success in reaching these targets.
Monitoring and remediation are important steps to optimize production, by adopting Predictive Monitoring, organizations can receive the ideal support to keep operations running efficiently, and ongoing maintenance to keep machines running at their best.
Predictive Monitoring is an AI driven method of production analysis. It uses metrics such as temperature and vibrations to determine when machines are working outside of their optimum conditions.
Around 98% of organizations report that a single hour of downtime can cost them over US$100,000, which highlights a significant cost implication that can be avoided with AI driven analytics.
TwinThread applications work with Predictive Monitoring to provide a network of data, which ensure machines work within their optimum conditions, for the best output.
Providing the Digital Tools
According to PwC’s ‘Digital Factories 2020’ report, “manufacturers’ adoption of machine learning and analytics to improve predictive maintenance will increase by 38% by 2022.” One of the main reasons for this, given by 98% of respondents, is to gain more efficiency through investment into digital factory solutions.
By learning what potential issues may occur if machines are not working to the correct standard, Predictive Monitoring systems work with TwinThread’s Predictive Asset Reliability application, which conducts an “automated root cause analysis” enabling the operator to analyze how the machine has fallen from its optimum conditions. Ultimately, any issues will be addressed much faster when predictive technology uses data to monitor variables.
“TwinThread’s Predictive Operations Center is making a big difference to our process engineers, giving them real-time feedback on the stability of our production,” said Domenic Verte, Manufacturing Application Manager at Toray.