New Engineering Data Flows | Automation.com

New Engineering Data Flows

March 142016
New Engineering Data Flows

By Peter Thorne, Director, Cambashi

New connectivity enables new data flows.  New data flows trigger new ways of working.  The changes that happened in production years (and in some cases, decades) ago can now happen in product development, driven by the new connectivity of Industry 4 and Internet of Things initiatives.  In this article, I will explore Industry 4.0 and Internet of Things (IoT) from the point of view of the design engineering process.

Life changing

“It changed my life.”  This was how a process engineer described a new remote access system that allows him to login to production lines in the factory from home.  Now, he volunteers for the night shift – extra pay, “…and if I can’t handle it online, I can guide the onsite team to fix problems before they stop the line.”

Remote monitoring and control is hardly new. Back in the 1960s, NASA was connecting to things in extremely remote locations.  Go back a bit further, French engineers set up instruments and monitored snow conditions on Mont Blanc back in 1874. Tesla (the scientist not the car company) demonstrated remote control of boats in 1898.  Commercial jet manufacturers have been offering in-flight monitoring of engines since about the year 2000, and a year after that, medical equipment manufacturers made it possible for surgeons to operate from a distance.

New capabilities

The improving price, availability, and reliability of technologies makes new connectivity relevant beyond this set of exotic and niche examples. The cost of an Internet connection is a whole lot less than NASA had to invest to connect to its ‘Things,’ and you get many-to-many not just one-to-one (or a few) connectivity.  So the business case is easier.  There are parallel changes in related technologies. Sensor technology has moved on; look at the sensors on mobile phones and other consumer devices – GPS, motion, orientation, sound, video, temperature; pressure, all at consumer prices.  Software technology has moved on; code is still important, but you can program and simulate an anti-skid braking system controller by connecting function-blocks on a diagram.  Cloud technology makes it easy to pay-as-you-go to run your software on someone else’s servers.

Figure 1: An in-service smart, connected product links a range of capabilities

The result is that just about every product with an on-off switch, and quite a few others as well, can be re-imagined as smart, connected products. So, as a design engineer, the block diagram in figure 1 is the new scope of your product.  The product you can touch is in the two boxes at the left hand side.  The full capability of the product is distributed across all six boxes.  If you need to enable your product for automatic ‘configuration to connect’, then yes, you may need to work with your IT colleagues in the corporate data center to devise a way of extracting the as-sold specification from the business systems, and translating this into a set of start-up commands for the product.

New data flows

So what is happening to data flows? It’s over 40 years since, in Germany, Pahl and Beitz defined procedures (and dataflows) for systematic design.  I wonder if Germany is again going to be a source of ideas on how to apply the new technologies. For example, this PDF document proposes design principles for Industry 4.0.  You have to get past page 10 to see them, but most of the points (interoperability, virtualization, decentralization, real-time capability, service orientation, modularity) have been very successfully applied to the development of cloud-based IT services.

In Industry 4.0, it’s not just computers; real equipment and whole factories will be treated in this way.  A new machine will be based on a common platform, it will integrate standard material handling subsystems, and will be differentiated by the way the software in the common platform uses the subsystems, and connects to the outside world.  A new factory will be part of a connected network of customers and suppliers, and algorithms will optimize the network performance, not just the individual factory.

The business case that appears to be IOT’s low hanging fruit is maintenance and service. This can be as simple as remote monitoring, but may involve new instrumentation on a product, together with predictive analytics software running on a connected server.  If the analytics can indeed predict failure, then the business case will identify lower service costs and higher customer satisfaction.

Impact on design engineers

But what about the design engineer?  How will a service-based business initiative, leading to development of a smart, connected product, impact the product development process?

The answer is…in a big way, especially once there are smart connected products in service in customers’ hands.  Figure 2 lists some of the organizations whose people and systems may need access to the product.  When products are developed into smart, connected products, design engineers must fight to make sure they get access to the stream of sensor data from the products.

Figure 2: Many people and systems may have reason to access a connected product

Of course the customer service team may say that engineering should be restricted to read-only access (and they may be right).  But if in-service product performance is available, design engineers must use it.  Just watch the data stream, you’ll build up a picture of how the product is used.  But you can do much more, for example, check, calibrate and improve, and calibrate simulation models.   If there is a problem, imagine being able to run an engineering review meeting using real loads, deflections, temperatures, and vibrations to animate and discuss a dynamic CAD model, and compare it to your simulation model.

New opportunities

Engineering software and service providers are alive to the new opportunities to provide tools for engineering teams to use this way. For example, PTC has placed IoT center-stage, and can demonstrate new information flows in many use-cases, including a product development use-case with a ‘wow’ factor, showing live sensor readings superimposed on live video of the product, and a dynamic CAD model, as well as a basic dashboard.  Another example, IBM presents 'continuous engineering' as part of the IoT opportunity.  There are specialist application examples as well. Noise and vibration specialist Vibrant offers an 'Experimental FEA' option. Siemens offers Transfer Path Analysis which enables use of direct measurements and operational data in analysis.  Wolfram’s electric kettle example includes the use of real data to calibrate a system model.

Product development teams will have to work to make sure they can get what they need.  The reason is that the business case for most IoT/Industry 4.0 initiatives will be based on service and lead-time improvements, not just better engineering.  So the investment will be structured around, for example, predictive analytics to enable lower servicing costs.  This project will be the opportunity to design the sensors, embedded software and back-end software so that the data can also be a new source of information and inspiration for design engineers.

A new frontier

This is a new frontier for product development, and another example that the silos are changing.  Of course it is a complicated environment - for reliable market sizing for just a subset of the tools used in IoT, namely systems engineering and embedded software development tools, Cambashi had to include 579 product groups from 425 providers.

The potential is immense. No need to wait for the fault report from field service, because live (and recorded) performance data from real in-service products is available at the desktop, in the meeting room, and via a tablet or smartphone.  But this use of real data won’t happen by magic.  The sensor readings are coming from an as-built, as-maintained product.  Is the CAD model as-built, as-maintained?  Does the simulation model match? If this matters in your environment, then you may have to find ways to use your smart connected product development project to handle these issues as well.

Even if it doesn’t make the night shift attractive, this new data-flow – the feedback of real data from in-service products - could change your design engineering life.

Author

Peter Thorne is the Director for industry analyst firm Cambashi (www.cambashi.com). His research focuses on the business needs of engineering and manufacturing organizations, and the way these needs are addressed by information and communications technology.  Peter has applied information technology to engineering and manufacturing enterprises for more than 30 years, holding development, marketing and management positions with both user and vendor organizations. Immediately prior to joining Cambashi, he spent seven years as head of the UK arm of a global IT vendor's Engineering Systems Business Unit, which grew from a small R&D group to a multi-million pound profit center under his leadership. He holds a Masters degree in Natural Sciences and Computer Science from Cambridge University, is a Chartered Engineer, and a member of the British Computer Society.

Related Articles

MORE ARTICLES

VIEW ALL

RELATED