Part Number is a fundamental element of any identification mechanism in product development and manufacturing. The debates about intelligent and non-intelligent part numbers are unstoppable like a wildfire. Back in time when the most updated bill of materials was on the cork board in shop-floor, people memorized part numbers. I believe new data management technologies can provide more part number intelligence to help selects parts and components. The brutal reality of engineering and manufacturing – both intelligent and dumb numbers are needed.
Is there a better way? What can be a potential alternative to complicated part numbering schemas and endless debates about how to classify parts? I think we have a chance…
The power of cloud computing paradigm can open new ways to identify information. Yesterday news from Google about PlaNet development are fascinating. Google is on a heavy mission to recognize the location on the earth based on the picture. Navigate to the following MIT technology review article to read more – Google Unveils Neural Network with “Superhuman” Ability to Determine the Location of Almost Any Image.
Here’s a tricky task. Pick a photograph from the Web at random. Now try to work out where it was taken using only the image itself. If the image shows a famous building or landmark, such as the Eiffel Tower or Niagara Falls, the task is straightforward. But the job becomes significantly harder when the image lacks specific location cues or is taken indoors or shows a pet or food or some other detail.
Today, that changes thanks to the work of Tobias Weyand, a computer vision specialist at Google, and a couple of pals. These guys have trained a deep-learning machine to work out the location of almost any photo using only the pixels it contains.
You can play www.geoguessr.com. Give it a try—it’s a lot of fun and more tricky than it sounds.
What’s next in computing article by Chris Dixon speaks about maturing of multiple technology over the course of last decade and moving into product development phase that can take advantages of these technologies. You might find few interesting examples there. It brings several amazing examples of technological achievements that were available only in Hollywood movies.
My favorite example is real-time object classifier developed by TeraDeep.
TERADEEP develops Deep Learning Software and Hardware solutions to accelerate the runtime of Convolutional Neural Networks and Recurrent Neural Networks resulting in more scalable solutions for datacenter applications.
The following two examples can give clear connection between past science fiction and reality of technologies today.
It made me think about problem of identification turning into the problem of object recognition. You might think about it as a next step after VR technologies. But at the same time it has a potential to replace VRs with more advanced technologies.
I can see some signs of machine learning development happening with the work done by CAD vendors today. Here is one example I captured from DEVELOP3D article sponsored by Autodesk – Design & machine learning. It speaks about organizing graphical content.
Here is the passage explaining the problem design graph is supposed to solve.
Organizing design data is a significant challenge, according to Haley. “After all, design data is inherently graphical. It’s not text. And it’s hard to search for something that’s not text.”
So how should you organise it? Haley and his team at Autodesk are working on a solution. Over the past four years, they’ve used machine learning to create a system that can categorise design data without human assistance and build design taxonomies, so when you need a certain design, you can find it.
The product, called Design Graph, will become available to the public for the first time later this year as part of Autodesk’s A360 cloud-based design and collaboration platform.
What is my conclusion? For very long time, meta-data was the ultimate way to mark every digital or real time object for the future recognition. Part numbers in the design is just one example. There are many others. We are coming close to the opportunity to eliminate meta-data identification and shift to the era of product recognition. Just my thought…
Picture credit MIT article, Chris Dixon blog and DEVELOP3D
Image courtesy of jesadaphorn at FreeDigitalPhotos.net
Disclaimer: I’m co-founder and CEO of OpenBOM developing a digital network-based platform that manages product data and connects manufacturers and their supply chain networks. My opinion can be unintentionally biased.