The Fourth Industrial Revolution has had a tremendous impact on business. While fears exist that this revolution could exacerbate the segregation of the job market, thus increasing wealth and income inequality, what’s to come is also seen as the solution to the skilled trades shortage made critical because of the retirement of the baby boomer generation. Meanwhile, the interconnectedness of processes and businesses are raising serious concerns about cybersecurity.
While the debate about the pros and cons could go on until the next industrial revolution, one factor is generally accepted: Amazing advancements are making manufacturing plants more efficient and agile.
Three of these advancements were discussed at VMA’s 2018 Valve Industry Knowledge Forum in April: robotics, additive manufacturing and the power of data to drive it all.
The following articles came from three presenters at the forum who gave VALVE Magazine their thoughts on the power of Industry 4.0.
The Transformation to Smart
By Mohamed Abuali
Across the world, manufacturing is regarded as an essential and uniquely powerful economic force. The unprecedented ability to connect, capture and analyze plant floor data drives the engine that is the Industrial Internet of Things (IIoT), which offers a new way of thinking.
Data-driven shop floor management is already creating greater efficiencies in every corner of the manufacturing ecosystem. It produces better supply chain management at the front door, more precise and prescient maintenance practices on the line and greater control over all the raw materials that go into the finished product—including energy and personnel.
Today’s modern manufacturers are more efficient and effective than ever. As the demands on manufacturing change to single-lot orders and new business models emerge, even more efficiencies are required to make good on promises. What smart data and better analytics provide to this picture is more effective use of resources without spending any more money on either people or materials.
Yet it is worth noting that “big data” in manufacturing is not really about size. While the idea of better data management on the floor is part of the “big,” the data has been there all along. Thanks to the advent of great data analytics tools and less expensive sensors, what has changed most recently is that more data is available easier to more people in exactly the way they need it for better decision-making.
Data analytics gets the entire manufacturing plant—from the corner office to the front line—up to speed in a hurry. The information reaches workers with the speed they need to make not just fast decisions, but the right decisions quickly. Big data, then, is really about smart data. It begins with information that is delivered to the people who have the skill to make better choices about how to turn that data into action and bottom-line results.
IIoT’s primary selling point is more connectivity, interoperability and intelligent analytics of data to drive actionable metrics such as “overall equipment effectiveness” (OEE). This supports the probability that throwing more money at a problem might not be as effective as analyzing what information is needed to get a true idea of the current state of affairs. Such level of insight through connectivity to the shop floor not only makes the business more transparent and minimizes costly impact, it also allows for modeling of data so that manufacturing leaders can predict the cost and the operational results of a decision before putting it on the plant floor.
When humans make a connection, they use a straight-line process—a handshake, a phone call, an email. Usually, this is easily understood and provides immediate feedback. Connectivity in manufacturing is similar in outcome in that information is exchanged about how the plant is operating based on the communication received from the assets. To get to that point of truth, however, requires many other smaller moments of connectivity, all happening in fractions of seconds. For connectivity to work on the plant floor, an agreement needs to be in place at each point of communication.
In today’s manufacturing, such communications don’t take place in a straight line, but in an intricate matrix. Data analytics have made gathering and contextualizing those communications possible in real time. Moreover, that matrix is getting larger, involving more parts of the organization, more technology, and more room for interpretation. It’s called interoperability, and to be a truly interoperable ecosystem in a plant involves three parts: humans, machines and processes.
Creating a truly interoperable system requires a referee of sorts—a way to tie all of these disparate parts together into a network. That’s how IIoT really works: It begins with a straight-line communications process, but the power of real connectivity in manufacturing has the potential to do much more.
The success of this effort will require better training, better equipment, better software and a more progressive view of how a plant should operate. When completed, the process can produce more satisfied workers, better morale, lower costs, reduced downtime and higher profits. Nothing has changed in the process except the way data is managed. Yet at the same time, IIoT can change everything.
One example of a company using smart data is valve manufacturer Richards Industries. Using shop-floor management software, Richards achieved a 40% increase in productivity improvements in less than six months, and it continues to create opportunities for radical optimization for more overall global competitiveness. Shop floor workers detect errors during production, and any deviation from the target is corrected to ensure a more efficient and leaner production process.
The interactions of the real and virtual worlds represent a critical aspect of the manufacturing and production process. Whether you call it Industry 4.0, the Fourth Industrial Revolution or the Industrial Internet of Things, technology advances have paved the way for machines not just to produce, but to communicate with other machines. Virtual technology is the backbone of such flexible production founded on the principles of lean management and the interconnection of ‘things.’ This leads the way to a truly transparent manufacturing environment—in other words, the smart factory.
One of the technologies integral to many smart factories is robotics. While people think of industrial robots simply as equipment that accomplishes repetitive tasks, robotics today can do much more than this.
Robots as Solutions to Manufacturing Challenges
By John Tuohy
Plant managers and CEOs wake up every day facing critical challenges posed by today’s manufacturing climate. The modernization of the plant floor combined with continuous efforts to improve efficiencies and control costs is a heavy burden that every industrial business faces. Everyone is wondering who has the answers and what solutions are available to both improve quality and strengthen the bottom line.
In a recent survey of CEOs from top advanced manufacturing companies, the number one concern respondents shared was finding qualified workers who have the required skill sets or experience. The aging workforce is an ongoing concern, and as baby boomers retire, the millennials are not back-filling the technical positions, which has created a serious skills gap that continues to widen. By 2025, there will be an estimated requirement for 3.5 million manufacturing jobs, which is about 2 million short of what labor will be available. Most companies agree the talent shortage influences daily businesses, and we’re all asking ourselves: How can we overcome this skills gap?
Meanwhile, other challenges manufacturers face are the rising costs of labor, raw materials, work in progress, how we can better control inventory and how we can assure continuous improvement in our methods. Already facing labor shortages, we must now overpay just to retain employees that may not even be the right fit. Some people believe we are stuck with workers who may lack proper training or don’t have their heads in the game. I believe the best way to get unstuck and to overcome labor shortages and meet production goals is to automate.
The standard productivity/efficiency of manually loaded machine tools is 60% or less. With robots, the same machine can achieve efficiency rates that reach 80-95%.
Robotic cells are designed to run continuously. They can perform value-added functions such as pre- and post-process operations including deburring, inspections, marking parts, assembling or packaging. This affords a plant the ability to shift employees from basic, mundane or dangerous tasks to the more engaging tasks within process management. A robot will always load faster for better “chip to chip” time (the period between when a part is finished being machined and the spindle is idle to when the next part is loaded and the spindle is cutting, also known as spindle utilization optimization).
The following examples emphasize these benefits:
A plant manager was recently asked by his executive team to explain the company’s production efficiencies and what percentage of that efficiency machine tools provided. The manager, who was an educated, experienced individual with a deep knowledge of his product and how it’s made, responded that his machines on the shop floor were 70-90% efficient.
Without any way to verify the response, management was left scratching their heads. They didn’t want to invest in more buildings, people or machines without verification. Management decided to collect and evaluate data from their machine tools using machine monitoring software. The eventual goal was to be able to make more chips. The types of information the software could collect and they could evaluate were limitless: They could study torque and harmonics to extend tool life, amperage draw for motor life and temperatures to assure accuracy. To focus on increasing operational efficiency, they concentrated on actual cutting time, loading and unloading time, and secondary processes performed on a part. The results of the investigation were shocking to the whole team.
The plant manager had based his original estimates on machine availability. However, while machines may have been available, when they were loaded up or set up for the next part, the operation was manual. In reality, when the data collected by the software took that into account, it found the machines were operating at only 40-50% and as low as 15% in some cases.
The plant manager was surprised his efficiency numbers were so low and contacted a consulting firm to review his processes. Having zero experience with robots, the plant manager felt disconnected with what should happen in a connected world. He learned that not only could he mine data from his machine tool but robots as well. As he began to realize how much data will drive a business, he implemented robots for machine tending combined with monitoring software for his machine tool. After implementing all this, the company achieved machine utilization of 80-90%, a 40% increase in efficiency.
In another case, a manufacturer was at capacity with an expensive machine tool and needed additional throughput. Management initially considered buying a second machine, but the cost was prohibitive. To free up existing machine time, a robot was added to perform the gauging between loading and unloading cycles. By adding the robot, the manufacturer reduced cycle time from 16 to 12 minutes, opening up 25% more machine tool capacity and allowing more parts to be produced.
These two examples show that, when faced with manufacturing hurdles, analyzing the right data and evaluating it in the right way can lead to options that may not seem obvious at first, including robotics.
JOHN TUOHY is senior district account manager, Fanuc America. Reach him at john.tuohy@ fanucamerica.com
While 3D printing or additive manufacturing (AM), as it is generally known in industrial applications, has been around for 30 years, its value for manufacturing valves and components was not fully explored until recently. While AM has its challenges, its value is quickly becoming obvious to those who are beginning to see the many ways AM can be incorporated into the industry.
Design Freedom through Additive Manufacturing
By Stephen Anderson
The most important potential advantage of using AM in valve manufacturing is design freedom; it is possible to print any flow path through a valve that an end-user wants. That user can specify the upper and lower limits, which are fed into the software. What’s needed is then printed with those specs without having to have new equipment for just that one model or product. What’s more, it’s possible to do that for every single customer with basically any degree of variation.
One example is a product produced at Bray. (See more about this project at www.valvemagazine.com>web only>“Metal Additive Manufacturing in the Valve Industry.”)
In the case referenced, the ball valve had a hole with a shape that had a certain flow rate. However, Bray’s customer wanted to know if a different flow rate could be achieved. Rather than creating entirely new tooling and machinery that met the requested flow rate, the specifications were fed into AM to print just that valve with a different shaped hole.
This process is so versatile, it is possible to create 500 or more models or products with just slight variations. The design changes are made with a CAD design tool, then sent to the software, then on to the printer. This capability opens up the potential for manufacturers to do many more bespoke applications for different customer requirements without adding manufacturing costs to the process.
Another benefit of AM is that legacy parts can be reverse-engineered when no drawings or models are available. Also, in the case of a breakdown of a piece of expensive equipment, when the end user needs that equipment quickly, the user does not necessarily need to break into the production line. Just the required part can be printed, then shipped within hours, even to remote locations.
Still, there are drawbacks to AM. The cost of materials plays a role in the process as does the need to allow a cool-down period before removing a part from the printer. The part also has to be removed before the unused powder in the machines can be accessed, which can add time to the process.
One of the most significant drawbacks, though, is that finishing a part that has been 3D-printed can be time-consuming. In this case, robotics may play a significant part in the AM process. The factories of the future, then, should be envisioned with both AM machines and integrated robots.
Still, the vision is exciting. Imagine a situation in which parts would be recovered from the printer automatically by robotic centers that measure the parts. If those measurements turn out to be not quite right, the parts could be passed on by the robot to another robot or directly put into the machine tool where they could be finished with subtractive processing.
These robots would be passing the component from machining station to inspection station to machining station automatically. What’s more, all that data coming off the machine tool or the measuring machine or the robot itself would be aggregated into the product management system. There, smart software can make decisions about quality based on gathered data.
This makes it possible to have a full manufacturing line with massive efficiency of scale. We could effectively remove some uncertainty from the equation because humans bring individual variation to processes. Robots can be more consistent, which could translate into better quality.
It certainly won’t replace our workforce, but the possibilities for efficiency are far-reaching.
As valve manufacturing embraces Industry 4.0, it seems inevitable that robotics, additive manufacturing and intelligent use of data will become the norm rather than the exception in the process. Concerns about cybersecurity and potential job losses will have to be addressed, but successful implementation of this new industrial revolution has the potential to exponentially advance the quality of the products and the profitability of manufacturers and end users alike.