TK Elevator advances global field service with agentic AI on Azure

When elevators work as designed, people move smoothly through offices, apartments, transit hubs, and hospitals. When they don’t, the impact is immediate. Across industrial environments, the challenge is familiar: aging assets, strict safety requirements, and the need to keep complex equipment running without disruption. For TK Elevator (TKE), uptime is more than an operational metric—it is fundamental to success.

“Elevators are mission-critical infrastructure. Uptime is a key measurement for our equipment because it reflects safety, reliability, and customer trust all at once,” says Philipp Cramer, Head of Global Field Operations at TKE.

As one of the world’s largest elevator and escalator manufacturers and service providers, TKE maintains about 1.4 million units—its own and third‑party equipment—across more than 100 countries, supported by more than 25,000 field service professionals. At this scale, even small inefficiencies translate into significant uptime risk.

Reduce unplanned site visits and improve uptime without compromising safety

Traditionally, elevators operated as stand‑alone systems. “Service calls required manual preparation to assemble the customer documentation, service history, and latest technical updates. Additionally, service technicians relied on their technical expertise and onsite visits to fix errors and maintain equipment,” explains Cramer.

TKE embarked on a journey aimed to improve technician access to real-time data for readiness. The company aimed to reduce unplanned service visits, improve safety and technical compliance, and increase elevator uptime to improve customer and passenger satisfaction.

Connect everything

To help meet these needs, 10 years ago TKE combined mechanical engineering expertise with technical expertise. The company created a connectivity platform for the elevator, using MAX connectivity solutions based on Microsoft Azure. The next evolution created digitally native elevators, including the EOX elevator, launched in 2022, for the low and mid-rise segment, and HELIX elevator, launched in 2026, for high-rise elevators. These proprietary elevator platforms are cloud connected, IoT enabled, and AI ready by design. Built on Azure, telemetry flows through Azure IoT into governed analytics via Azure Databricks, creating a trusted backbone that links field service and global Digital Operations Centers (DOCs) for real-time support and remote intervention. This governed, real-time data foundation gives agentic AI the trusted context it needs to transform daily operations.

Today, technicians can arrive at a service call with real-time information from a customer’s connected elevator. Technicians can access customer-specific information plus a real-time global technical database to learn from and build on in addition to their knowledge. The platform enables TKE and service technicians to instantly access knowledge from their global colleagues that aggregates individual training, experience, and available documentation. Technicians are better equipped to evaluate and learn across the installed connected elevator base to increase uptime for customers and passengers.

“The value of leveraging AI is integrating it with TKE’s deep domain knowledge broadly into service operations. Our new service delivery model is supported by regional Digital Operations Centers, our engine for scaling digital service excellence. To operate our data-driven digital and AI solutions safely and reliably at our scale, we need a global technology partner that can grow with us. We have that in Microsoft,” says Matthias Gohl, Chief Digital Officer at TKE.

Turn insights into uptime

With cloud and data infrastructure in place, TKE can now connect proprietary data, operational workflows, and technician experience globally with new agentic AI modules. The company is able to analyze and share knowledge and service-relevant insights more efficiently throughout its technician community globally. Technicians are prepared before they arrive onsite. The agentic AI approach prompts and guides decision making in the field and captures post-visit summaries, transforming verbal entries to text, so the AI gets more informed, and the global database can be updated with each visit.

Elevators vary in age and configuration—some more than 40 years old—while contracts and response expectations shift by building, time of day, and situation. “Technicians need to have a lot of things on hand to make a good service visit: service history, current state, spare parts requirements, SLA and warranty, not to mention the expertise and knowledge,” says Cramer. Agentic AI helps to assemble that context in advance. Drawing on telemetry, service history, contractual terms, and prior technician insights, the agent gives technicians real-time information before they arrive.

“Our goal is to harness the insights from our 25,000 global technicians and use the insights in our local operations. Our TKE technicians can arrive better informed as our Digital Operations Centers and the Azure AI–powered briefing agent brings the global expertise and current context together,” shares Gohl.

Once onsite, data-informed and AI-generated guidance reduces system switching and supports diagnostics. Afterward, voice‑guided debriefing captures actions taken and lessons learned—creating a learning loop that improves the next visit.

Protect every ride

In the United States, where TKE introduced AI-supported workflows, this approach helped avoid about 20,000 unplanned service visits and reduced callbacks more than 40% where structured DOCs-supported interventions were executed per unit in the first year of deployment. It not only strengthened safety and uptime but also generated meaningful cost and CO₂ savings.

As expectations rise for always-on, safety-critical infrastructure, TKE’s approach shows how uptime can be engineered into daily operations. Agentic AI with humans at the center can transform field service into a smarter, more resilient global system—one that keeps people moving safely every day.

Discover more about TK Elevator on Facebook, Instagram, LinkedIn, X, and YouTube.