My recent article “What is wrong with PLM of Things? led me to have discussions online and offline about integration of IoT and Product Lifecycle Management applications. In general, I think, the topic is still hugely misunderstood and the level of confusion among technological vendors, manufacturing companies and average consumer is very high. While for average consumer, Internet of Things can be represented by the ability to connect home sensors to their mobile phone, manufacturing companies and software vendors can see it differently.
Control Design article IoT is just a connection to move the data by Dave Perkon brings a perspective about IoT from recent Autodesk University 2015. According the Carl Bass, Autodesk CEO, IoT will enable three main things: (1) software to control and change actual product behavior; (2) analysis to optimize product maintenance; (3) turning sensors and data coming from products into the design feedback.
I can see how software can control products. The examples of delivery of new product features through the software upgrades became almost normal thing for many manufacturers. To control products in real operation is probably tricky, but possible. Predictive maintenance is also something we can imagine. Companies can collect information from sensors and use some KPI to plan maintenance accordingly. Think about it as a glorified solution to collect mileage information from the car and scheduling maintenance appointment in car dealership.
Navigate to Autodesk SeeControl solutions page. You can see several solutions there – Machine monitoring and Failure prediction, Spare Parts Management, Maintenance Workflow, Warranty and SLA management. In my view, all these solutions are using data about product captured from sensors combined with service-related product information. Very interesting solutions, but to me it is still a “silo”.
At the same time, to turn sensors information into design feedback sounds magical to me. Here is a passage from the article that speaks about that.
“If we are gathering sensor input, we can feed it back into the design process,” said Bass. “Version two will then use real data to drive its design. The design won’t be perfect the first time, but the data defines the lessons learned, and they are applied in an improved design. The sensors and the data will be a part of the design process.”
How do you get the data? Well, Bryan Kester, head of IoT at Autodesk and formerly CEO of SeeControl, announced Autodesk SeeControl, an IoT cloud service platform company allowing users to collect, analyze and use data from remote products. It looks like an easy-to-use way to optimize existing products and equipment by capturing the information or, better yet, intelligence required.
So, how sensors data can be turned back in the design process? A typical design process used tools that are usually not connected to actual product used by customers. How do you think it will be possible? I can think about some analytical tools that can provide an information about product performance, but it sounds “too marketing” to me.
In my view, manufacturing should close the loop between PLM and IoT by developing a mechanism to provide measurable feedback to engineers on their design activity. I want to capture Carl Bass’ quote – “the design won’t be perfect the first time”. So, it is not about design, but about every change engineer will do into already existing and released product. Which is ECO management in PLM. To create a tool that can make analysis of product performance connected to ECOs related to the new product release we can give a feedback to engineers about quality of product improvements. Sounds like an “agile development” approach or A/B testing? Maybe, but I can see some practical sense and real opportunity to measure engineering activity.
The idea itself isn’t new. I found PLM portal blog with report from PTC LiveWorx 2014 conference. The following picture of PTC strategy – Closed Loop Lifecycle Management is a good marketing strategy visualization of gathering sensors’ information to show how efficient last ECO and related product changes were in the field. If this solution is available, I’d love to see that.
What is my conclusion? Maintenance and spare part management are great solutions to leverage sensors’ data. But these are early low hanging fruits for IoT vendors. The real opportunity is to close loop between PLM and IoT and bring measurable impact on every product change and ECO back to engineering office using sensors and analysis of product performance data. That could be a real deal. Until that time, IoT solutions will be glorified sensor data connectors with some data management and analytic. Just my thoughts…