Manufacturing / Business electronics and solutions
Panasonic Connect: Checking drawings against design specifications in the design and development department
- Company
- Panasonic Connect
- Country
- Japan
- Adoption stage
- In operation
- Source published
- Date basis
- The date the source was published. It can differ from the date adoption started.
- How the source was checked
- Read the full source text
The work problem
According to a press release Panasonic Connect published on February 19, 2026, checking whether the specifications written in multiple documents such as product drawings, part drawings and technical specifications match one another is essential to assuring product quality in manufacturing design and development. This work relied on manual work and visual checks by staff and took a great deal of time, and when human errors such as missed checks occurred they led to rework in later processes, product recalls and repeated inspections, carrying the risk of large economic losses and damage to the brand. Because the items to be checked were PDF drawings, unstructured data that is hard to automate with AI, existing solutions could not solve the problem.
Technology and data
Panasonic Connect developed in house an AI application called the Manufacturing AI Agent that supports checking work, using Snowflake's AI capability Cortex AI on Snowflake's data cloud platform. The agent automatically extracts text from multiple PDF drawings, automatically checks items such as material and finish between product drawings and part drawings or technical specifications, and shows the results in a list. Staff no longer compare several documents side by side and finish the work by confirming the results the AI has checked. The company deployed the agent to its design and development departments as a business AI of its in-house AI assistant service ConnectAI, and began by applying it to checking standards and checking exterior parts in its own business.
Results
According to the press release, checking work that took 50 to 340 minutes with visual checks was cut to 10 minutes, reducing working time by 80% to 97%. The company said that standardizing the work also reduced the variation in quality between staff. Panasonic Connect cites this as a concrete use case of Connect Corpus, its AI and data platform that brings company data together in one place, and plans to extend it to other checking work to raise productivity across design and development.
Limits and open questions
The press release does not disclose the checking accuracy, the share of results staff corrected, the number of drawings processed or the measurement period.
Sources
- パナソニック コネクト、図面／設計仕様の照合業務で独自開発のManufacturing AIエージェントの利用を開始news.panasonic.com, Accessed
Compiled from public sources. These are not results from ATF Works customers.