Blog: Leverage AI to Augment the User Experience at the ATM

It’s a fact that the artificial intelligence (AI) revolution is on. Leveraging AI, along with its subsets of machine learning and advanced data analytics, to increase the performance of the ATM channel and ensure a positive user experience began years ago. The outcome is proven: the use of AI so far has truly helped optimize operations and maximise ATM uptime for a large number of financial institutions. As new capabilities like generative AI continue to develop, further advancements are expected, and their implementation will help reach new frontiers.

In 2024, AI is certainly top of mind of financial institutions around the world. There are numerous opportunities to unlock, and the stakes are high, including deeper customer relationships, sales growth, and simplifying operations to reduce costs, just to name a few. Examples of the proven power of AI are diverse: optimizing pricing for increased margins, improving fraud detection to minimize financial losses, generating additional revenue through personalized offers, digital cross-selling via social media, and enhancing client service while reducing support costs with in-app chatbots. Some challenges remain, but this is only the start: the McKinsey Global Institute recently estimated that generative AI could create up to $340 billion in annual value for banks across the globe¹.

Although it is not widely known yet, the ATM channel is one banking area where the use of AI is already at scale and proving to help financial institutions increase uptime. When ATMs function as true Internet-of-Things (IoT) devices, detailed, technical data can be regularly collected from sensors and data points. This information is then analyzed within cloud computing platforms using machine learning. This enables the identification and monitoring of patterns that generally occur through the devices’ lifecycles, continuously building and enriching a technical knowledge base, and even establishing a precise personality profile for every single ATM device operating in the field.

In the event of a device failure, typically the most likely root cause can be automatically established within a matter of seconds, eliminating the need to send a technician to diagnose the fault. Consequently, an automated recommendation can be issued for the precise fix, the required level of skills and experience of the technician, the spare parts needed, and the time the repair should take. As a result, repairs can be better planned, completed faster and earlier, and with very high first-time fix rates.

An AI-driven approach can also enable the shift from a reactive to a predictive service model where incidents can be preempted. By analyzing data patterns, trends, leading indicators and other key data points, impending failures can be identified, so maintenance activities can be scheduled at a time of low customer usage to avoid an unplanned future outage and maximize the uptime of a device.

More than 240,000 devices across over 90 countries are connected to Diebold Nixdorf’s data intelligence platform powered by AI. DN AllConnect^SM Data Engine builds on a unique combination of decades of engineering experience and a continuously augmented global technical knowledge base, as well as the application of the latest developments in IoT, cloud computing and storage, machine-learning technologies, and AI. It Is the enabler of our truly data-driven service model, which augments our employees’ expertise to resolve technical incidents faster and earlier, and, in some instances, predict failures and prevent them from occurring. Diebold Nixdorf maximizes uptime of all fleets signed up to their availability management services, reducing downtime by 50% in some cases.

To discover more about our AI-driven service model, visit DieboldNixdorf.com/DataEngine.


Sources: ¹ https://www.mckinsey.com/industries/financial-services/our-insights/been-there-doing-that-how-corporate-and-investment-banks-are-tackling-gen-ai#/