The sensors are getting smaller and more efficient. The last BGR blog featured remarkable new sensor chip pulls power out of the air so it never needs to be charged. The Telegraph article – Marc Andreessen: ‘In 20 years, every physical item will have a chip implanted in it’ few days ago brings an interesting perspective on the future of connectivity and computing. According to Andreessen – “The end state is obvious – every light, every doorknob will be connected to the internet.” While it is obviously interesting to read predictions for 20 years ahead, my special attention was caught by the investment and interesting claims made by IoT company funded by Andreessen – Samara:
His [Samara’s CEO] big idea: installing cheap sensors, and uploading and analysing data to the cloud makes Samara 1/10th of the cost of existing industrial sensors (complex systems made by huge incumbents like Intel), and deployable in under 10 minutes.
Which made me think how it will impact PLM platforms. I recall one of my first articles about IoT on Beyond PLM back into 2009 – PLM and Internet of Things. Here is a quote back from that article:
So, you can ask me “where is PLM in this story?”. Here is my take. All physical objects surrounding us designed and manufacturing with a certain level of PLM system involvement. Different manufacturing companies have various levels of PLM deployment, but most of them have CAD to create digital models, they track data about what they were engineering and manufacturing. Since we take more and more environmental responsibilities, we need to track what we designed in real life. This is where I see “internet of things” are finally connecting. I think RFID technologies is only beginning in whole story of connected virtual and physical devices.
Fast forward in 2015. We can see many changes. PLM companies are getting excited about IoT- related business. Some of them are saying “IoT is PLM”, which makes me smile a bit and recall that according to Gartner’s hype cycle, 2015 is a peak of inflated IoT expectations.
In my recent article – What’s wrong with PLM on Things?, I’ve been sharing my thoughts about potential danger of siloed PLM-IoT deployments. The problem is that most of companies are building “PLM of Things”, which has a lot of similarity with old “single source of truth” PLM model. Now, this model is extended into the domain of capturing information from sensors and data analysis. Cloud technologies (especially in data crunching and analytic) are enabling new level of functionality and can make a better sense for the data capturing from devices. Think about REST API, but for devices. It can change a trajectory of PLM IoT development to become an “internet” and not IoT in a box.
Chip in every physical item can make companies rethink PLM platform architecture. The whole purpose of PLM systems today is to be a data management repository serving manufacturing companies during the design, engineering and production phase. But, it is going to change. The fundamental purpose of PLM systems will be to provide information about how systems is built and how it can behave in a real world. Think about design and engineering as a “development” (initial) phases of product lifecycle.
What is my conclusion? IoT will expand the horizon of lifecycle into the realm of physical objects. But at the same time, it will change a traditional PLM vision. Originally, PLM was designed as a “silo” to manage design and engineering activities. The result (design and engineering data) was thrown over the wall of manufacturing and disappeared in manufacturing, service and support solutions. Future PLM architectures will change the nature of PLM platforms – it will be global to cover all stages of product lifecycle. The design and engineering phase will be just a beginning stage that will create an initial information about product and its behavior. PLM platforms will serve as a source of product information and behavior. The cost of future PLM platforms will be a critical element to support a lifecycle processes for every design in a world. PLM platforms are expensive and slow to adjust to modern manufacturing environments. This is a note for PLM architects and futurists. Just my thoughts…
Picture credit BGR blog