Industry 4.0: Is this “fourth industrial revolution” just another buzzword boardrooms use while the machine shop is knee deep in oil & swarf? Meanwhile, nobody's bothered to do a proper inventory in the stockroom, so we’ve run out of machine screws...again. Or is industry 4.0, as some sources have you believe, the dawn of a new age representing a paradigm shift in not only how things are manufactured but how they interact with the user?
Since it means a lot of things to many different people, we’d like to share our view of Industry 4.0 - and what it means for additive manufacturing (AM).
According to a definition from the Industry 4.0 working group in Germany:
“Industry 4.0 will involve the technical integration of cyber-physical systems into manufacturing and logistics and the use of Internet of Things and Services in industrial processes. This will have implications for the value creation, business models, downstream services and work organization.”
Cyber-physical…what?!? If you ask us, this bit of jargon is worse than the Industry 4.0 buzzword. But bear with us as we try to pick it apart. It basically comes down to this: as advanced as our modern manufacturing facilities might be with lean manufacturing, quality assurance, and CNC machinery - most machines still have the behavior equivalent of a very obedient, small child.
Instructions can be given to a machine and it will follow them, to the letter. Even if this means smashing a tool right into a billet. Holes will be drilled and components assembled - even if they’re the wrong ones or in the wrong place.
By inventing machines to automate processes that remove undesirable human characteristics such as fatigue, boredom, strength, accuracy, forgetfulness and sickness - we’ve lost some of a human’s ability to cope with the unexpected, catch mistakes and generally see the bigger picture. Arguably most manufacturing plants are too big for any team of humans to keep track of this big picture and control all logistics anyway.
This is where the ‘cyber’ in cyber-physical systems comes in. The falling cost and subsequent improvements in overall size and capability of sensors and programmable logic controllers have unlocked a vast amount of data. Everything from stocking/inventory levels, the temperature in the factory, or throughput per hour can be tracked and analyzed.
It is argued by Industry 4.0 proponents that we now have all of the building blocks to make ‘smart factories.’
Interestingly enough, smart factories don't necessarily mean factories with artificial intelligence that makes products before you even need them. It just means ‘not a totally dumb factory,’ as even simple decision-making and error-catching have big impacts on overall manufacturing process.
Industry 4.0 can be much more than pure manufacturing. It represents a cultural shift in the way we do business, design, manufacture, and interact with one another. Early industrial revolutions took humans and made them machines. Mass manufacturing and commerce forced us to do mindless repetitive jobs, a single assembly operation on a production line, or endless data entry.
Industry 4.0 concepts can give machines that little bit of intuition they lack, enabling them to do mindless repetitive jobs while still having the human oversight to cope with the unexpected. Humans are now beginning to have the freedom to work the way they work best - or want to live. Mass connectivity has enabled flexible working hours and home offices. Mass automation is forcing us to become more educated as machines finally take mindless jobs off our hands. However, a cultural shift will be required to encourage us to become more educated so we can handle these new occupations. Furthermore the software tools that carry out work need to become more intelligent to support the burdens of new jobs that require higher levels of thought and attention.
An important role will also be played by the paradigm shift in human-technology and human-environment interaction brought about by Industry 4.0.
These possibilities can change manufacturing in profound ways. Our analysis of the resulting technology trends motivates a framework that captures activities for additive manufacturing’s move into mass manufacturing.
NIST (National Institute for Standards & Technology) has been working to standardize this for many years now, and there have been developments to the ASME Y14.41 & ISO 16792 standards - meaning Model-Based Enterprise (MBE) Architecture is becoming more widely adopted. But trying to import PMI data from your favorite CAD package to CNC machine is never as straight forward as it should be and adoption rates are still quite low. Industry 4.0 is pushing for a true digital thread, in its approach from Analog to digital conversion.
Digital Design Verification & Simulation
The design rules for additive manufacturing are far different from traditional manufacturing, and must break away from the traditional process loop of: guess, design, build, and test then redesign. Design validation and simulation for AM must be realized for industry 4.0 to reach its full potential. Digital design verification and simulation aims to test designs while you work. All of the PMI data available, coupled with expected environmental conditions the part endures in its lifetime, enables software to guide design decisions during the process.
Design verification techniques must assess a part design for functionality, manufacturability, and conformance to industry-specific quality management systems, and simulation results which allow for AM design freedom fully realized within the confines of validation strategies.
Process modelling allows AM machines to reach their automation potential by storing past manufacturing data. Part files will adapt to account for inherent inaccuracies in the manufacturing process to produce higher fidelity parts. Eventually this process could be used to digitally certify designs and parts with minimal physical testing by simulating and accounting for the unquantifiable unknown, instead of enduring long and costly physical testing processes.
AM has the potential to be the most efficient method of converting data from the cyber to physical worlds. To complete this coupling, sensors, controls & connectivity (Internet of Things) must feed relevant information to connect both worlds. This data loop, its proper analysis, and interpretation are key principles of Industry 4.0.
This is sometimes referred to as 'Deep Learning' - using a neural network to account for sources of ‘unknown’ variables, and past patterns to statistically conceive the correct value. The goal is to solve problems in a similar manner to the human brain, albeit more predictably and repeatedly. For example, a 3D printer which pre-emptively corrects mistakes based on the results of previously similar but not identical jobs.
Authors:
Dave Hayden: Sr. Engineering Consultant, Stratasys Consulting
Dave has a master’s degree in mechanical engineering from the University of Sheffield. He works as a consultant providing technical insight into the broad range of consulting projects we undertake. As a CAD specialist, he is also involved in the redesign and manufacture of AM parts for our clients and enjoys projects that incorporate both electronics and mechanical systems.
Kunal Mehta: Principal Director, Stratasys Consulting
Kunal’s work has spanned both B2B and B2C as well as across multiple industries, driving strategic and operational transformation for Fortune 100 clients. He spent eleven years with Accenture Strategy, a global strategic consultancy, prior to joining Stratasys. He holds a Bachelors of Biomedical Engineering from the University of Minnesota and a Masters of Business Administration from Melbourne Business School.
