DENSO TEN Limited.
AI visual inspection to improve soldering quality and safety of electronic circuit boards.
- Use Case
- Manufacturing
Personnel in charge: (Left to right) Akihiro Yamada, Ryoichi Ishizuka, Jun Sugimoto
HACARUS took into account the characteristics of the data, they could provide a quick response to the entire process from data processing to AI model creation. Many of the AI models and techniques proposed by HACARUS had not been thought of by our company, which led to an improvement in the skills of our staff. We are grateful to HACARUS for their quick response to our requests and questions.
Background
Our department (ICT Business Promotion Department) is in charge of making effective use of the data held by each department within Hitachi Zosen. The goal is to contribute to high profitability and business transformation by improving productivity, improving operational efficiency, and upgrading products and services. We had a hard time to create highly accurate AI models, because some of the projects had lots of missing data. Until work in cooperation with HACARUS, we kept looking for good solutions.
Why HACARUS
Firstly, dealing with missing values was the key to this project, and HACARUS, with its extensive experience in analyzing missing data, was able to provide effective solutions. Secondly, the fact that HACARUS data scientists and our staff are able to work closely together is very useful in terms of improving the skills of our staff.
Service Content and Results
HACARUS developed the data science process of literature review, baseline model creation, and accuracy improvement, that taught us about lots of new approaches.
HACARUS followed up with our staff well, explaining not only the AI models to be created, but also the details of the algorithms, and how to manage the project. We got the expected effects, both in terms of AI modeling and skill development.