Reimagining service operations with AI and data insights

By Yuval Stein, AVP Technologies, TEOCO
Yuval Stein, AVP Technologies at TEOCO, outlines the value of data-driven decision making for data centre providers

For network operators, service operations are frequently cited as one of the most complex aspects of the business to run. And for those that get it right, they are likely to emerge as the undisputed market leaders in the long term. Today’s fast-paced digital environment demands a simultaneous, coordinated, and dynamic approach across their various business units which have traditionally operated in silos. In recent years however, artificial intelligence (AI) has revealed the potential to simplify multiple tasks by optimising the various functions that make up operations within each business unit and their coordination. Operators are slowly beginning to home in on that promise and are now finding success with AI solutions that optimise and enhance service operations journeys.

AI and data insights offer operators a 360-degree awareness of the network, including potential failures which translates into anomaly detection and prediction. AI applications used within the context of service operations can employ advanced algorithms to look for patterns within data, allowing operators to detect and predict anomalies in the network. It also enables them to proactively fix issues before customers are negatively affected and their services are interrupted. AI can be used to analyse, optimise and correct errors in real-time. An uninterrupted service creates a self-organising network that can be optimised and configured automatically. Additionally, AI can predict whether a similar problem will occur in the future and then take steps to prevent and solve it in advance. This then improves performance and allows operators to deliver an optimal experience to customers.

Reimaging service operations at a time of technology evolution, digital transformation, virtualisation of cloud-native functions and 5G is important in the current climate: with network complexity gradually increasing and manual processes reaching breaking point, operators are under continuous pressure to reduce costs. Meanwhile, throwing more people at the problem is prohibitively expensive. This is where advancements in automation, self-healing networks, self-organising networks, zero-touch service operations centres and so on might allow operators to ensure their services remain reliable and scalable to keep up with ever-increasing customer demand.

AI: empowering intelligent operations 

The promise of AI in the context of service operations is that it has the power to boost network efficiency, reduce OpEx, and improve customer experience. This is due to the fact that when applied to service assurance, AI has the ability to perform human-like reasoning and achieve more contextual decision making, at speed and at scale. AI and machine learning (ML) can be used to detect patterns, flag anomalies, trigger root cause investigations, and recommend or implement remedial actions. ML can also establish a log of best practice actions for similar future events. The advantage of this approach is that it allows network operators to establish a service operations model that is both dynamic and self-improving. As more data sets are continually analysed, the operator is able to achieve higher levels of accuracy and operational improvements. 

Remediation of network and service problems as reflected by symptomatic network faults can also be optimised: Assurance triggers a troubleshooting workflow, a process that runs through a pattern of checks in order to isolate the cause of an issue. If such a problem is detected, a ‘remedial action’ is then triggered to fix or put a workaround in place. Overall, the introduction of this process to service operations contributes to a more systemic procedural change in operations: these heightened levels of performance management and fault monitoring enhance operational efficiency and improve customer experience, as service operations go from being reactive to predictive. Root cause analysis that can be engaged pre-emptively to prevent a customer-impacting fault is a significant step forward for service operations. 

Adding value: an automated network environment 

The application of AI and ML to support consistent and reliable operations becomes the catalyst for a fully automated network environment that is self-configuring, self-healing, self-optimising and self-evolving. In other words, a network that runs itself and that requires little to no human intervention. Such an eventuality is desirable for network operators because thanks to the arrival of 5G and its associated technology ecosystems, systems have become so complex, dynamic, and fast-paced that the ingenuity of engineers can no longer be depended upon to repair even minor network errors as and when they occur. The preponderance of network faults arising from this growing network complexity means that there are now too many to manage, and response times need to be immediate. As a result, automation needs to be the default, with network engineers stepping in to address the bigger issues as a last resort..  Using AI insights naturally requires tighter cooperation between the network software and management software (the Operation Support System, OSS), and this is being made possible thanks to newer architectures from standards bodies like 3GPP, O-RAN and TM Forum.

Additionally, the cost savings that can be achieved through automation are also a key motivating factor for network operators. The industry has been under considerable cost pressure for the last 20 years, and it’s always a challenge to maintain the pace of investment in the network in the face of diminishing returns. It is welcome news for many in the industry then that a 20% increase in automation investment could reduce labour costs for operators by 90%. The move to an autonomous network not only promotes greater efficiency and reduces operating costs, but it also allows for the freeing up of resources which allows operators to pursue more value added tasks.

A question of lifecycles 

For network operators, creating, analysing, and working with AI and ML data is very different from traditional software. Traditional assurance solutions simply present information on screen for an operator to monitor and act on. Such a capability is expressed from day one i.e., a turnkey solution. This isn’t generally the case with AI and ML where the system goes through a learning curve. These systems have shorter lifecycles and are best defined as a hybrid mix of both a product and a service. They require regular re-training and updates as they are continuously learning from new data. Having the right support structure in place for the ‘care and feeding’ of these new systems will be critical to their long-term success.

Transforming service operations  

AI can offer valuable predictive insights when it comes to anticipating network failures, troubleshooting incidents, and uncovering opportunities for optimisation and efficiency gains. At the same time, the economics of investing in AI and ML systems need to strike a balance between achieving longer-term cost savings versus showing immediate benefits – namely operational efficiencies. 

The practical application of AI and ML is set to transform existing systems and processes as well as the role of service operations operatives. The success of which lies in properly embedding AI/ML results in the service operations lifecycle.


Featured Articles

AWS Data Centre & Cloud Computing Leader to Invest in Italy

AWS to invest US$2.2bn in Italian data centre expansion by 2029, as the global data centre leader continues to grow across Europe and Asia

Huawei Cloud Expands Global Footprint with New Cairo Region

Huawei Cloud is expanding its footprint with the launch of a new cloud region in Egypt, becoming the first major public cloud provider in the country

Onnec: Building Future-Proof Data Centre Strategies

Data Centre Magazine speaks with Onnec’s Niklas Lindqvist and Matt Salter about the future of sustainable data centres and how to harness the power of AI

STT GDC Vietnam Expansion to Fuel Digital Transformation

New OVHcloud Data Centre in Sydney Powered by Liquid Cooling

The Datacloud Congress is back for 2024