How to Put AI to Work in EDM

AI in EDM is a partner for smarter machining, not a replacement—embrace change or risk stagnation. Source (all images): StackTeck Systems Ltd.

I dismissed AI as a passing fad.

After experimenting with a language-based AI using general conversational prompts about EDM, I received underwhelming responses like “EDM requires little skill to be used effectively." However, my perspective changed after watching Trevor Goforth’s practical interpretation of AI in a YouTube video.

Featured Content

How is an Aluminum Mold Energy-Efficient? READ MORE

Your Guide to Smarter, Faster Mold Design READ MORE

MMT Chats READ MORE

His extensive experience as both a moldmaker and EDM machinist inspired a significant revelation for me. This technology may not be just a lot of hype; it is a tool that has the potential to elevate manufacturing, reshape perceptions of machining and even tackle systemic challenges like the skills gap.  I thought it was worth giving a language-based AI another look to discover its true potential in a machining environment.

6 Ways to Use AI in EDM

Ultimately, in EDM, what truly matters is machine hours and efficiency, not just the hours logged. I've come to understand that AI can serve as a skills multiplier; it allows me to push forward new ideas that otherwise would not be possible.

At work, I had a brand-new wire EDM equipped with advanced probing features, but I wanted to add batch pickup cycles to get more unattended machine hours. I was able to create a simple transposition tool in a few hours that accomplished my goal. Was it groundbreaking? No, but it worked and demonstrated AI’s potential as a powerful tool for problem-solving and bridging several of my knowledge gaps. For the first time, I saw real potential in the technology.

Was it groundbreaking? No, but it worked and demonstrated AI’s potential as a powerful tool for problem-solving and bridging several of my knowledge gaps. For the first time, I saw real potential in the technology.

Integrating AI has now become almost second nature. In addition to the multi-part transposition tool for wire EDM, here are a few other noteworthy projects that highlight its transformative role:

• 1 Machine Monitoring Dashboard Built in Hours, Not Months: With minimal programming knowledge, I used AI to develop a system that aggregates real-time machine data from MTConnect into a functional dashboard. I was trying to develop automated record keeping and I'm well on my way to making this a reality. AI is excellent for gamification, and since you are the architect, you can customize it to meet your specific needs. Understanding your past performance is essential for determining your future goals. 

AI-built machine dashboard in hours, not months—custom features, real-time data and gamification without costly code or enterprise pricing.  

• 2 NFC Crane Check Integration: We embedded NFC (Near Field Communication) tags for crane inspections, allowing AI to bridge the gap between hardware and software. This integration provides a faster method to record the checks into our proprietary ERP system.

• 3 Job Data “Spider”: Did you know that you can extract tables and data from websites and then format those numbers in programs like Excel or Google Sheets? This capability will be useful later for automated job time tracking and process planning. Think of it like a smart dashboard that weaves all the scattered EDM job data into one clear, traceable "web."

• 4 Wire Cutter: While designing an electronic wire cutter for an EDM, as an alternative to a lighter, AI assisted me with the complexities of Ohm’s Law and resistance. I discovered that having no resistance would cause the melted wire to curl and risk burning my fingers. Live and learn.

• 5 G-Code Optimization: Will it write all your programs? No, and I wouldn't trust it to do so. However, with the right parameters, it can help identify hidden variables and does a good job of cleaning up messy loops and advanced macro programming. Although it can be used as a sounding board, there are clear limitations, and it is important to remember that this is a tool to elevate your skills and not a human replacement.

• 6 Tool Setter Post-Processor: Years ago, I wrote a manual G-code template to integrate the tool setter data with the Fanuc controller on the CNC machine. More recently, I asked AI to perform the same task, and it replicated my original process fairly accurately. 

In EDM, where downtime can inspire creativity, AI serves as a partner in progress. This journey is not about pursuing complexity for its own sake, but about creating systems that allow us to work smarter, attract new talent and redefine what it means to be a modern machinist. The real question is: are shops willing to embrace this new way of thinking or will they continue to adhere to the status quo?