It’s a common fallacy to assume we have a handle on the future based on past or present circumstances. If 2020 has collectively taught us anything, the bright shiny lesson is likely that we should never be so hubristic to assume we can know what challenges the future holds.
A second lesson, which is informed by the first, is that we can be prepared for the unexpected even if we don’t know the particulars of what’s around the corner.
The concept of Industry 4.0 (or sometimes referred to as the 4th Industrial Revolution or “Factory of the Future”) is the methodology by which manufacturers are surviving 2020 and fitting themselves for whatever 2021, 2022, 2023, etc. has in store for them.
One answer to achieving a nimble production process that adapts smartly to new and even earth-shattering changes (this is the dream of Industry 4.0) lies within quality inspection, validation, and monitoring. The collection and transfer of quality data has the power to unlock new pathways to production efficiency, throughput acceleration, process innovations, root cause analysis, product performance failure avoidance, and zero defect manufacturing.
In short, optimized data equals lower costs and higher yields.
A good way of thinking about the transition between each industry epoch is by looking where data exists in each industrial iteration.
We are currently in an era of metadata and the promises of Industry 4.0 are a shift away from a document mindset. What this entails is not eschewing documentation per se but not thinking about data in the flat ways that paper and even PDFs lead to. With all this information flying around, not being able to discern the most important data nor understanding how to interpret it, causes 70% of manufacturing companies to not implement Industry 4.0 successfully.
To learn more about looking at your adhesion process holistically, collecting predictive data and putting all the pieces together, download our eBook: The Manufacturer’s Roadmap to Eliminate Adhesion Issues in Production
A dynamic approach to data collection, sharing and analysis is what is going to bring manufacturers through.
What this looks like for Quality Engineers is the accumulation of data that pertains to the success of their products. Anything that can be measured is fair game; temperature, strength, topology, and surface quality, just to name a few.
What we mean by surface quality is the preparedness of a material surface for bonding, coating, sealing, printing, painting, or just that the surface is clean of all organic contaminants that could hinder adhesion and performance.
Surface quality, in this sense, is a chemical state that is extremely difficult to measure in production. Most equipment, that is sensitive to the molecular changes that affect surface quality, is very limited in its ability to be used “near to action” - a crucial aspect of Industry 4.0.
If you can’t measure quality on real parts, in real time, then you are going to be left behind. But it’s these types of data problems that manufacturers must overcome to usher in their own futures.
Digital assets aren’t necessarily any more complicated than they sound when taken at face value. Any content that can be stored and shared electronically are digital assets. This could be a time stamp, a video, images or a contact angle measurement used to test the cleanliness of surfaces during an adhesion or cleaning process.
The thing that brings Industry 3.0 into the bright light of Industry 4.0 is how these digital assets are collected and utilized.
Many manufacturers across every industry are heeding the call by automating their production lines. This can be an extremely important step in developing a smart manufacturing framework. Automation breaks open opportunities to adopt new and emerging technologies that are fueling advanced manufacturing processes.
New, disruptive, and unconventional technologies are a big part of creating the kind of high rate, zero defect production manufacturers are looking for. What’s critical to remember is this: the adaptive, flexible and responsive nature of technology and the data they provide will be what revolutionizes processes.
Robots alone will not save us. Automating a flawed process oftentimes merely speeds up the production of poor quality parts. Manufacturing processes that include automated data analysis allow actions to be taken faster and as far upstream as possible to prevent major disruptions. It’s not just riveting, surface treatment, adhesive application and curing, or other traditionally manual processes that we should be automating.
Data about part quality that is instantaneously collected at every Critical Control Point (CCP) - from the supplier to inventoried parts to finishing procedures - is the best tool to build a holistic map of what is actually happening to material surfaces, and therefore product quality, in an actual production environment.
A key characteristic of manufacturing processes that achieve an Industry 4.0 level of production is data collection that occurs where the action is. This is sometimes referred to as a “near to action” approach because it takes the esoteric ideas of cloud computing, artificial intelligence, virtual and augmented reality, the ever-buzzy Internet of Things and it brings them all down to earth.
These advanced tools are only as cool as they purport to be if they produce results.
One major benefit of the Industry 4.0 ethos is that it actually brings high-end analytical power within reach of mid-sized and smaller companies. Digitization of data can make it less expensive to gather and share so it ostensibly levels the playing field.
Although, there are many companies out there that would have you believe that in order to be able to legitimately claim an Industry 4.0 facility you must purchase every shiny gizmo and do-dad available. It simply isn’t so.
This kind of marketing can make it stressful for Quality Engineers who hear from their executive team that they want to implement an Industry 4.0 model because it makes the move seem like a budget drain with only theoretical gains.
The truth is, Industry 4.0 is more about bringing the decentralized, intrapreneurial, cooperative spirit of the internet age into the manufacturing sphere. It’s about empowering each level to innovate and act quickly with minimal to no extra cost. The principle idea is to harness the power of predictive analytics so trends toward failures can be headed off or changes (either big or small) can be adapted to with precision and as little guesswork as possible. When you have visibility into the minutiae of your process you can see where adjustments can be made to accommodate new situations and you can anticipate how those adjustments will affect all interrelated processes.
The best way to understand this is to start with a current problem or an issue that you want to avoid. Look for the gaps in your current process.
For instance, an automotive manufacturer needs to reduce the weight of a new design by a certain percentage so they construct the frame of a lightweight composite instead of steel. To further cut down on weight, they adopt an adhesive bonding process to remove metal fasteners from the design. They know that plasma treatments will help this adhesion process be more successful, and they are trying to implement more Industry 4.0 tactics, so they install automated atmospheric pressure plasma equipment throughout their facility. To follow through on the Industry 4.0 mindset, however, they need to be gathering data on all these new processes. They need to ask questions about the surface quality of the composite materials and they need data to know that the plasma treatments are truly creating a bondable surface every single time.
With a smart sensor attached to the automated system they can immediately assess the quality of the composite surface before and after treatment, send that data all the way across the world or to the manufacturing execution system to make automatic adjustments or flag trends. Now the automotive manufacturer is building with a more secure system that has a level of self-regulation that it didn’t have before.
Inevitably, equipment like plasma treatment systems will drift out of spec by dint of just using the machinery. Sensors that not only reveal whether drift has taken place but can precisely quantify how much, protect manufacturers’ investments in new equipment and legacy machines they want to keep around. Equipment can be more connected and assured to always be running at 100% efficiency when checks are built into the process that prove you’re getting the most out of your equipment.
Being strategic by attacking these issues using an Industry 4.0 approach will blossom out to other areas of production.
It’s well known that manufacturing processes are interdependent and that even a miniscule change at one end can have massive repercussions at the other. A 1 degree change in the rutter and suddenly you’re whole ship is a thousand miles off course.
That’s why data solutions to manufacturing issues need to directly inform engineers what is happening in their process and at scale. A common language from end-to-end means better communication from Research and Development teams in their testing and design labs to technicians overseeing the actual procedures to remote employees and management who may rarely step foot on the factory floor.
Each CCP needs to have correlative data gathered from its particular point in the process that directly references and compensates for every other relevant step. And this all needs to be done in real time.
Seems like a big ask, but if you break down the CCPs into categories based on how the data can be used, it reveals the quilted pattern that pulls the whole process together.
The baseline data gathered at the earliest stages of development can be considered preventative analytics. These are specifications and requirements for the end product and each step along the way to achieve flawless output. This lays the groundwork for all analytics gathered later on.
Next are predictive analytics that give clues and insight into how prepared operations are to meet real world circumstances. This can be a preliminary inspection before an established abrasion step or a cleanliness measurement prior to storing a part. This is the data that makes these processes more nimble than ever. Anticipating change means it doesn’t rock the boat when it occurs.
Validation analytics are data gathered to prove the efficacy of a process. For instance, when a fluoropolymer has been laser etched and a simple contact angle measurement verifies that the part is now chemically reactive enough to be adhered to, this is a perfect example of validation analytics at work. This data is often gathered at the most active points in the process although it can also be gleaned during post-production monitoring. These can be in the form of coating hydrophobicity verification or uniformity measurements.
Increasing precision, efficiency, technology adoption, and expanding the idea of what manufacturing can be requires creativity but is not as difficult or far off as it may seem.
Shifting mindsets about how to lean into the magnificent connectivity that already exists in most manufacturing processes will allow companies to mine those processes for nascent data just crying out to be developed and put to good use.
It’s all right there and Industry 4.0 is a fancy term for looking where manufacturers haven’t looked before to improve their processes.
To learn more about looking at your adhesion process holistically, collecting predictive data and putting all the pieces together, download our eBook that outlines the biggest dilemma for manufacturers. “The Manufacturer’s Roadmap to Eliminate Adhesion Issues in Production” eBook draws a line between all your CCPs and guides you through how to implement smart sensors that give you the data you need.