30,000 pieces, give or take, depending on the model; that’s the average number of parts in a car from the smallest nut and bolt to the frame. Multiplied by the billion-plus cars that exist in the world and the manufacturing process to get all those parts seems endless. Then consider every other industry that needs parts manufactured, from your phone to kitchen appliances to airplanes, the manufacturing scale is massive.
Today, many of the pieces are digitally manufactured, the machine being programmed to proper specifications to make parts at scale. However, only geometry of the machining process is considered in this process. Problems still exist around poor quality, lengthy iterations, and wasted consumption, all resulting in more money, energy, and CO2e emissions spent by machining companies. Startup Productive Machines has developed a solution: a platform to help machining companies produce parts cheaper, faster, and at a higher quality.
The company’s solution is a web-based platform that provides a replica or digital twin of the machining process considering the physics of the machining processes. The customer can easily plug in their parameters such as machine tool, cutting tool, tool path, and workpiece material. The platform then runs through simulations of the process, recognizing potential problems before they happen in real production. By identifying the ideal cutting conditions for the process, the customer can optimize the production to create a higher quality product faster with less cost and waste.
Using MATLAB and the machine learning tools, Productive Machines created an AI algorithm at the basis of the digital twin platform. The team recognizes the challenges that startups face and counts on MATLAB toolboxes and embedded functions to help them get their work done quickly and efficiently.
Hear more from Productive Machines CEO Erdem Ozturk and CTO Huseyin Celikag in this month’s startup short!
To leave a comment, please click here to sign in to your MathWorks Account or create a new one.