CSPs can strengthen their networks with closed-loop management

This article was written by members of TM Forum's Closed-loop Anomaly Detection and Resolution Automation project. Authors include: Aaron Boasman-Patel, Vice President – AI & Customer Experience, TM Forum; Emmanuel Otchere, Chief Architect, Huawei; Mathews Thomas, Executive IT Architect, IBM; Satishkumar Sadagopan, Associate Partner / Executive Architect, IBM; Utpal Mangla, VP & Senior Partner, IBM; Vikrant Bhargava, Deputy General Manager, Bharti Airtel. Please also register for the webinar – Enabling the networks & services of tomorrow using AI closed loop automation – for further insights. If customer experience in the field of telecoms was a term few years back, it has now become a core for driving engineering and network operations today. The challenge is not only to provide superior customer experience, but to do it using a complex interwoven microservices architecture. This requires the use of a framework as a key lever for automating engineering and network operations. Closed-loop Anomaly Detection and Resolution Automation (CLADRA) refers to a TM Forum framework that contains a reference architecture and related collateral to enable communications service providers (CSPs) to transform network operations by using AI driven closed-loop automation to detect anomalies, determine resolution and implement the required changes to the network within a continuous highly automated framework.

Network challenges

Engineering and network operations are extremely challenging for the following key reasons: The above challenges make it extremely difficult for engineering and network operations to be able to quickly correlate and analyze multiple application performance metrics to solve complex emerging problems before they impact end-user experience. There needs to be a framework which can drive efficiency in the network, predict problems, drive automation and ultimately improve the user experience.

• Servicing customers is no longer just about applications or devices, but about business service which includes end-to-end observability
• An enormous volume of data is now being generated across ecosystems in terms of logs, alerts, metrics etc.
• Network management is no longer about fixing the failure but about predicting the potential failure. Reactive is outdated – it’s all about predictive

Closed-loop automation capability framework

Closed-loop automation helps solve any problems before they even become issues. The following are some of the key capabilities of closed-loop automation to transform engineering and network operations into a predictive and automated operations model: TM Forum is currently conducting a study on CLADRA and the following are some use cases members of the study have deployed: Further details of the above and other use cases can be found here.

• Anomaly detection: This enables the capability to ingest and process large data in real time. Also, anomaly detection makes use of time series data to analyze applications, networks, operating systems, database metrics etc. This gives anomaly detection the capability to identify patterns and anomalies and raise awareness towards predictive actions.
• Intelligent alerts: In a general operations environment, 20% of overall alert volume is false positive. These false positives add to the overall load and volume of operations teams. closed-loop automation uses machine learning models to create the patterns for the series of alerts so that those can be bound to causes and known actions and corrected accordingly.
• Root cause analysis: Closed-loop automation leverages data to intelligently identify all anomalies in the service path and use AI to map it to find the most likely cause for a particular incident. During the course, it makes use of various AI algorithms to ensure accuracy of root cause identifications and implements the required remediation steps
• Predictive planning: With this capability, CSPs can predict how application and network behaviors are dependent on seasonality and other factors, to ensure appropriate corrective actions are taken, thereby permitting systems to perform optimally.
• Traffic flow optimization: Identifies network traffic issues by analyzing past and current traffic issues; uses AI to identify solutions and implements the correction using the management and orchestration layer to achieve a self-healing network and efficient network optimization.
• Alert correlation for operations: Identify and correlate alerts using machine learning so that they can be bound to known causes and be used for remediation of alerts
• System performance prediction: Real time capacity forecasting for hardware components on the basis of application parameters considering seasonality and network loads to ensure capacity is available.

Closed loop automation overview

Closed-loop automation can address issues of varying complexities and the implementation will vary depending on the problem being addressed. The figure below provides an overview of closed-loop automation addressing issues of varying complexity:

• Some issues are relatively simple to resolve and the network can be modified easily to implement the desired changes. This is often done by analyzing the data and invoking the orchestration engine to make the appropriate changes to the network components.
• In other cases, it is important to predict issues which could happen in the future. The appropriate data is analyzed by various predictive models which then make a recommendation on the change to be made to the orchestration layer which implements the change.
• In complex cases it is important to combine the predictive insights information with additional AI systems to determine a resolution. The AI system has been trained to resolve these issues and is integrated with a robotics automation system to automate the process. If the AI system determines it has a high confidence that the suggested resolution is correct, it will invoke the orchestration engine to implement the solution automatically. If not, a trouble ticket is generated, and an engineer will resolve the issue.

Benefits

The following are some of the benefits of closed-loop automation We hope you have a better understanding of the importance of closed-loop automation, its associated use cases and the benefits of implementing it. The next blog will take a deeper dive into the key components behind closed-loop automation and a high-level architecture. We invite you to join our current study on closed-loop automation, with further details to be found here.

• Improved network reliability through automation built on AI
• Superior customer experience leading to reduced customer churn and improving the bottom line
• Manual tasks are reduced through automation hence raising workforce productivity
• Mean time to resolution for incidents are decreased providing improved network services, better network performance and a faster rollout of new services