What challenges did KONE face before AI implementation?

Technicians faced long waits for technical help desk support. KONE needed a secure way to use confidential maintenance documentation and IoT device data. Accuracy and safety were critical for troubleshooting guidance.

What was the impact of the AI solution?

The initial pilot had 100 users for 3 months. The assistant is now live in 11 countries with around 1,500 active users. KONE expects around 6,000 users in the coming months and eventual availability to all 40,000 technicians. The solution helps technicians get elevators back up and running more quickly and reduces time onsite.

Source

KONE Uses Amazon Bedrock to Build a Secure Technician Troubleshooting Assistant

Use case typeKnowledge managementUpdated Jun 13, 2026

KONE, a global elevator and escalator company, built a generative AI assistant on AWS to help field technicians troubleshoot issues faster using confidential documentation, maintenance reports, and IoT data. Technicians access the assistant through a mobile app called Technician Assistant as a first step before escalating to technical help desks. The solution emphasizes security, privacy, least-privilege access, data minimization, and encryption with AWS Key Management Service.

Organization: KONE
Industry: Manufacturing
Location: Finland
Published: May 2026

Reported outcomes

40,000 technicians

total technicians targetedOther quantified impact

100 userspilot users3 monthspilot duration1,500 usersactive users6,000 usersplanned users

Strategic outcomes

Speed & agilityHelped technicians troubleshoot issues fasterCustomer experience & trustReduced time onsite for techniciansMarket & geographic expansionRolled out in 11 countriesScale & capacityPlanned availability for all technicians

Primary read

Use case focus

Showing 3 of 3

1Technician support
2Field service assistant
3Knowledge management

Technicians faced long waits for technical help desk support.
KONE needed a secure way to use confidential maintenance documentation and IoT device data.
Accuracy and safety were critical for troubleshooting guidance.

KONE built a generative AI technician assistant on Amazon Bedrock.
The assistant is delivered through a mobile app called Technician Assistant.
The solution uses a pipeline covering preprocessing, retrieval, evaluation, and text generation.
KONE applies least-privilege access, data minimization, anonymization, and AWS Key Management Service for encryption.
KONE worked with the AWS Generative AI Innovation Center and AWS prototyping team to build and refine the solution.

Technologies

Amazon Bedrock AWS Key Management Service

The initial pilot had 100 users for 3 months.
The assistant is now live in 11 countries with around 1,500 active users.
KONE expects around 6,000 users in the coming months and eventual availability to all 40,000 technicians.
The solution helps technicians get elevators back up and running more quickly and reduces time onsite.

Architecture

KONE built a secure generative AI troubleshooting assistant on Amazon Bedrock and delivered it through the Technician Assistant mobile app. The architecture includes preprocessing, retrieval, evaluation, and text-generation stages, and uses data-minimization, anonymization, least-privilege access, and AWS Key Management Service encryption to protect confidential maintenance and IoT data.

Implementation partners1

AWS Generative AI Innovation Center

Sources & evidence1

Groundedness: 5/5Type: Customer StoryPublished: May 27, 2026Publisher: AWSEvidence: PrimaryConfidence: High

Primary source

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Customer evidence