Use Cases
Omdia has identified seven key telecom AI use cases, which are outlined below. The lines between many
of them are blurry. There are many commonalities between cybersecurity and fraud prevention, and some
AI solutions might be used for both use cases.
Network Operations Monitoring And Management
No area in telecom will be more critical to automate with AI-driven solutions than network operations.
Telecom networks must automate to survive. Next-generation networks will be impossible to run without
significant AI-driven automation.
AI-driven network management solutions providers are productizing tools to design networks, predict global
functions like load balancing, reduce handover between cells, and automate node slicing, as well as offer
broader systems aimed at delivering autonomous networks. But network management poses challenges
for AI-driven solutions and many network automation specialists believe that deterministic, rules-based
automation solutions will remain critical and, in many ways, more important to network automation.
Network management solutions will collectively be the largest AI opportunity in Africa between 2018 and
2025, with cumulative spending of more than $541 million, or almost 70% of total investment in AI solutions, during that time period. However, rollout of these solutions in the region will be relatively slow compared to
other regions because of delayed implementation of NFV, SDN, and 5G networks. Omdia believes most
of the investment in AI-driven network solutions during this period in Africa will be focused on solutions that
can help automate network functionality that is in place today in non-NFV, SDN, and 5G networks.
Predictive Maintenance
Telecommunications services typically rely on heavy equipment, machinery, lines, boxes, and a range of
other infrastructure to maintain connectivity and reliability. When parts go down, costs incurred are manifold:
costs of machines, costs of maintenance required (e.g., labor, emergency rates), costs of downtime, and
costs of customer frustration and loss, especially when customers are businesses. The ability to manage
so much capital outlay is critical. As in other industries, CSPs are also applying AI to predictive
maintenance.
Spending on AI-driven predictive maintenance will represent a very small number in Africa over the forecast
period, $2.4 million. The reason for this is low labor costs in the region.
Fraud Mitigation
To detect telecom fraud schemes, AI, ML, natural language processing (NLP), and deep learning (DL) are
being explored in ways that do not solely rely on pre-programmed rules or models based on historical data.
The goal for such systems is to become self-learning, with models continuously updating individual profiles,
threat profiles, payment methods, situations, behaviors, and other parameters.
The IoT and other technologies will create a very large number of endpoints for telecoms. The nature of the
network will change with 5G (e.g., high-density urban deployments). This will create a large number of
patterns, difficult for humans to ascertain. Because AI is self-learning, it will also be used to help manage
these patterns, fraudulent or otherwise.
AI-driven fraud mitigation solutions are being used in Africa today. Spending on these solutions in Africa
over the
Cybersecurity
ML and DL are used to aid in learning from threats and predicting optimized protection for all types of
telecommunications infrastructure, assets, and networks. Specifically, telecom companies can leverage
ML, DL, and machine reasoning (MR) to review massive amounts of data to detect suspicious behavior,
foresee equipment failure or downtime, identify threat types and profiles, and protect confidential
information.
AI development is now targeting how to respond to cyberattacks on networks, working to quickly block
suspicious communications and analyze malicious behavior and software—tasks still often allocated to
humans. When under attack, the system will be able to identify the entry point and stop the attack, as well
as patch the vulnerability.
AI-driven cybersecurity solutions are being used in Africa today. Spending on these solutions in Africa over
the forecast period will total $16.8 million, growing from $0.1 million in 2018 to $5.6 million annually in 2025.
Customer Service And Marketing Virtual Digital Assistants
Telecom operators were some of the first enterprises to widely embrace the use of VDAs, particularly for
customer service. Companies seeking efficiencies and automation for customer support and customer
service began experimenting more than 10 years ago with automated applications that leveraged NLP.
Most of these applications lived within an enterprise website, delivering smart search for frequently asked questions (FAQs), and then later as pop-up VDAs and avatars with more advanced capabilities. In the last
3 years, significant advances in combining NLP with other forms of AI, primarily ML and DL, have made
enterprise VDAs more intelligent and more useful.
Verizon Chatbot, Powered by Creative Virtual
(Source: Creative Virtual)
Spending on AI-driven customer service and marketing VDAs in Africa today is robust, even with relatively low customer service labor costs. Omdia estimates that spending in 2018 will reach $3.1 million. Total spending over the forecast period for Africa will be $127.9 million. By 2025, annual spending will reach $39.4 million in the region.
Intelligent Customer Relationship Management Systems
In telecom environments, AI can be applied to nearly every aspect of CRM, including personalized notifications and promotions, real-time responses to service requests, augmenting loyalty, and reducing churn. AI-driven CRM solutions are being used in Africa today. Total spending on these solutions in Africa over the forecast period will be $32.4 million, growing from $0.4 million in 2018 to $10.7 million annually in 2025.
Improve Customer Experience Management
CEM is used in telecommunications to support various elements of the customer experience that most CRM systems are not tooled to handle. Examples include auto-adjusting network parameters, service quality detection, website quality detection, and addressing network performance or security needs in real time. Mobile and network service providers are now leveraging AI in a number of ways that both enhance customer experience and help automate quality of service.
Improving CX is a universal phenomenon that will be important to CSPs both pre- and post-5G. AI-driven CX solutions are being used in Africa today. MTN Nigeria announced in July 2018 that they have launched a CEM platform with Nokia that will use AI to monitor and analyze network and end user device performance. Total spending on these solutions in Africa over the forecast period will be $24.1 million, growing from $0.3 million in 2018 to $7.9 million annually in 2025.
