AI boost for operators worldwide

15 Jan 2018

With telecoms operators sitting on a massive pyramid of diverse customer and network data, artificial intelligence (AI) and machine learning (ML) are gaining traction in the telecoms industry. AI technologies can be used to understand, optimize, and improve business and network capabilities.

Telcos are no stranger to AI, which has been widely used to improve customer services, and operators are making progress using analytics effectively in key areas that are linked to revenue growth, such as reducing customer churn and digital marketing.

Look out for the telcobots

See Also

Telecom Asia December 2017 / January 2018


LPWA IoT gains traction


Industry watchers anticipate that telcos will apply AI more broadly to revolutionize the way customer service is delivered and internal processes are managed. Some expect AI to enhance service personalization through offers and customer experience, while others anticipate more automation and advanced analytics to drive significant benefits on operator’s top and bottom lines, thanks to better cross-selling opportunities and optimized offerings-as well as opex reduction through automation, improved efficiencies and elimination of redundancies.

A new wave of telco virtual assistants, or “telcobots,” are emerging as telcos increasingly adopt AI to improve their customer services, says ABI Research industry analyst Sarju Vasavada in a statement.

Vodafone, for example, is looking at different AI technologies-from chatbots to conversational UI and smart assistants-to enhance customer experience across the group.

In April, Vodafone launched Tobi, its virtual assistant, to address their customer service woes after being fined £4.6 million ($6.14m) by UK regulators Ofcom for falsely charging more than 10k pay-as-you-go customers for top-up credit. They also had a record-breaking number of customer complaints until Tobi was powered-up.

The AI-based chatbot enables Vodafone to deliver a better customer experience by offering immediate, relevant support to resolve more than 70% of its customer queries.

“Traditionally, customers looking for help and advice are restricted to either connecting with an agent or using some of our self-service options to resolve their query,” says Vodafone. “But these options give customers little choice over how they can interact with us. AI blends these experiences by offering self-service capabilities with an agent experience.”

Vodafone says customers have reacted positively and the operator already started to expand the range of support Tobi can offer across basic and more complicated queries. It also plans to fully integrate Tobi with its systems so that the majority of interactions on its webchat service can be automated, leaving its agents to focus on higher quality interactions and commercial conversations with its customers.

Apart from Vodafone, several telcos are also leveraging AI, and NLP heavyweights, including IBM Watson, Nuance, Liveperson, and Ipsoft are building technology in-house.

Or do you prefer chatbots?

Spanish telco Telefonica is developing their chatbot, Aura, for 2018, and DT’s Tinka already averages 50,000 customers in Austria every month.

“The recent introduction of virtual assistants in customer service signifies the level of urgency within telcos to start emphasizing the importance of customer relationships and customer care management, something they have been taking for granted for decades,” says Vasavada.

While telcos are currently prioritizing these virtual assistants primarily to improve customer engagements and reduce churn rates, they are also positioning them to compete directly with the Siris, Cortanas, and Alexas of the digital world, the analyst says.

Vasavada says telcobots now assist customers with a variety of issues ranging from basic account inquiries to SIM purchases, service troubleshooting, and technical settings. However, the most important point is that telcos are realizing the advantages and benefits of adding virtual assistants to their arsenal for customer service delivery and are aggressive in introducing them throughout their footprint.

ABI Research forecasts AI investments by telcos will reach $14 billion by 2022 with a CAGR of 22.4%. Expect multi-talented chatbots from Orange and SK Telecom to be released in early 2018.

Virtual assistants will enable telcos to save as much as $1.2 billion on customer care management by 2022 with a CAGR of 17% over the next five years-particularly impressive given that telcos were not the early adopters of virtual assistants, says Vasavada.

“Telcos are slowly but steadily getting ready for prime time,” he says. “We are bullish on telcos making this Ônext-gen’ leap within the next five years.”

AI to smooth 5G/IoT integration

As telcos increasingly adopt technologies like SDN and NFV, AI will play a major role in smooth integration of these technologies and automating the networks, as the telecoms industry move towards the era of 5G and the massive IoT.

Counterpoint Research expects that AI application in mobile networks will focus on self optimizing networks, SDN/NFV, and the enablement of neural networks. Among these, the research firm says, SONs will be earliest, as SONs enable operators automatically to optimize network quality based on traffic information by region and time zone based on various machine learning algorithms.

As AI holds the key to optimizing and automating telecom networks, telcos, equipment vendors, standard bodies and industry groups are all embracing AI one way or the other, with the majority of activities still in experimental stages.

AI-assists in the OS

One of the pioneers in AI is SK Telecom, which is expanding the use of its AI-assisted network operation system Tango to all its telecommunications networks.

The operator has already been using Tango (the T advanced next generation operational supporting system) to help manage its fixed-line network, and is now extending the application of the system to the mobile network.

Tango uses machine learning to automate the optimization of network operation based on network traffic information broken down by area and period.

The system is also designed to enhance the accuracy of network management by measuring the quality of network operations delivered to customers, and incorporates virtualization capabilities to help mobile operators adopt new network capabilities including IoT and 5G.

“The AI-assisted network operation technology based on big data analytics will be essential in the 5G era,” says SK Telecom’s SVP and head of network technology R&D, Park Jin-hyo. “SK Telecom will continue to improve the functionality of Tango aiming at providing the best-performing network for customers to enjoy.”

In October, SK Telecom entered an agreement to provide the Tango platform to India’s largest operator: Bharti Airtel.

Enter the ITU

Meanwhile, the International Telecommunication Union (ITU) recently announced the formation of a new focus group to establish a basis for ITU standardization to assist machine learning in bringing more automation and intelligence to ICT network design and management.

The ITU says machine learning algorithms help operators make smarter use of network-generated data. These algorithms enable ICT networks and their components to adapt their behavior autonomously in the interests of efficiency, security and optimal user experience.

The ITU Focus Group on Machine Learning for Future Networks including 5G will lead an intensive one-year investigation where technical standardization can support emerging applications of machine learning in fields such as big data analytics, network management and orchestration, and security and data protection. The group will also explore and analyze use-cases of machine learning and underlying technical requirements as well as interoperability issues.

“Machine learning and artificial intelligence are finding promising applications in communications networking,” says the Focus Group’s chairman, Slawomir Stanczak of Germany’s Fraunhofer Heinrich-Hertz-Institut. “This Focus Group will establish a basis for ITU standards experts to capitalize on machine learning in their preparations for the 5G era.”

The European Telecommunications Standards Institute’s (ETSI) industry specification group has also formed a new industry working group, led by China Telecom and Huawei, dedicated to exploring the use of AI in the deployment and optimization of telecoms networks. ETSI says the purpose of the group is to “improve operators’ experience regarding network deployment and operation, by using AI techniques.”

“While SDN, NFV and network slicing technologies are helping networks become more flexible, the complexity of network management is not being reduced, but merely transferred from hardware to software,” says Wang Haining, vice-chair of the group. “Experimental networked intelligence helps to address this complexity.”

Ocean Sun, director of service strategy and architecture at Huawei’s Global Technical Services (GTS) unit under the Carrier Business Group, agrees that AI is still in an early stage for scale deployment, but he expects the technology will be increasingly deployed by telcos in near future.

“Operation and network efficiency will be a big headache for telecoms operators as they move towards 5G, and we expect to see more and more operators to deploy AI and data technologies to improve their network and operational efficiency,” says Sun.

The Huawei director says his firm will also spend $200 million over the next three years on technologies like AI and big data to further improve operators’ operation and network efficiency.

The company has launched six projects with telecoms operators in Asia and Africa to explore use-cases and verify the functionalities for AI. Initially these projects focus on using AI technologies to improve predictive antennas, field operations, network planning, maintenance and optimization and radio optimization.

This article first appeared on Telecom Asia December 2017 / January 2018

Related content

Follow Telecom Asia Sport!
No Comments Yet! Be the first to share what you think!
This website uses cookies
This provides customers with a personalized experience and increases the efficiency of visiting the site, allowing us to provide the most efficient service. By using the website and accepting the terms of the policy, you consent to the use of cookies in accordance with the terms of this policy.