JapanManufacturing2022 – ongoingToyota Motor
In-house AI Platform for factory staff to build their own machine learning models
Toyota factory staff built about 10,000 models of their own on an in-house AI platform in 2024
Problem
Toyota's plants had many manual, labor-intensive inspection and monitoring tasks. Building a machine learning model for each one normally needed AI expertise that most floor staff did not have.
Approach
A small team in Toyota's Production Digital Transformation department built an AI Platform with web applications for creating models and links to production line equipment. It runs on a hybrid setup of on-premises systems and Google Cloud. Floor employees train and deploy models themselves, for example to inspect finished parts, check adhesive application on back-door glass and detect abnormalities in injection molding machines at the Takaoka plant. The platform is used in all 10 car and unit manufacturing plants.
Results
Takeaway
Toyota put model building in the hands of the people who know the line, and the count of models grew each year. The hours-saved figure comes from Toyota's own staff on a Google Cloud blog.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer visionPredictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
A Toyota Production Digital Transformation staff member described the in-house AI Platform on Google Cloud's blog. He reported nearly 1,200 active users, model counts rising from 8,000 in 2023 to 10,000 in 2024, and a reduction of over 10,000 man-hours a year. He also described the hybrid architecture and the small team that built it.
Full details are in the original sources.