Breakthrough: Using AI for Lean Data Mining

About The Webinar
Many of you have been exploring the capabilities of ChatGPT and AI tools. One of the areas of high interest are in its use for “data mining” or data analysis. I wanted to see how well ChatGPT and Claude (a ChatGPT competitor) would be able to analyze our standard line design and simulation model data, and provide useful feedback. To be honest, I was blown away! Here are the topics we’ll be exploring in detail:

Analysis of Static Line Design Data
​​​​​​
I provided a one-sheet summary of basic line design data including products, volumes, processes, takt time and resource calculation and process times. The level of understanding was surprising, and I will report on what we were able to learn.

Detailed Process Flow Data
The AI system was then challenged further with detailed station-by-station and product-by-product data, including descriptions of the process flow. This data came directly and without modification from our Lean Design Simulator data file format. With this more detailed level of data we were able to drill quite a bit deeper with our questions. We’ll give you a full report.

Buffers and Sequencing Rules
One of the biggest challenges of a Mixed Model Line Design is the setting of optimum buffer (In Process Kanban) sizes, and in the development of sequencing rules. We ask AI to tackle this, which it did with confidence (which we then tested).

Running Simulation Models With Simulation Software
Finally, I ask the AI system to actually “run” the simulation model, since we had provided all of the necessary data. It then proceeded happily to do so, and we’ll report on the AI simulation results versus running the model with real simulation software.

Be Prepared to be Amazed!
We’re only just scratching the surface of what is possible here, but attend the webinar to get a taste of what it’s like working with an AI Lean Sensei!