Hello and a welcome to the Process Automation podcast. A podcast from ABB that shines a light on their process automation business area and the work they're doing around the world. I'm Fran Scott, maker, presenter and all around engineering geek. And across this series we've been exploring the invisible
force of automation. So that's the incredible processes happening under the surface that enable our day- to- day lives from the heating in our homes to the water in our taps, ABBs technologies are working behind the scenes, orchestrating industrial processes, machinery, and systems. Today we are looking at the
future of process automation. For more than a century, automation systems have been absolutely central to empowering industries that provide these basic building blocks of our everyday lives. From energy, power, water, metals, minerals, chemicals, and transportation, and enabling them to scale to meet the needs of a growing population.
For more than 40 years, ABB has been leading the way when it comes to something called distributed control systems AKA DCS. Now these are at the heart of the
largest and most critical operations on our planet. Now, if you recall the first episode of this series, I spoke to Peter Turk, the president of ABB's process automation business area, and we discussed the key role that process automation is playing in making the world safer, more efficient, and
increasingly sustainable. Now, Peter introduced us to many of the industries that are benefiting from ABB's work in process automation, and over this series we have put those under the microscope. Industries like shipping and water to mining and continuous
emissions monitoring. Today we are going to dive deeper into the driving force behind process automation and find out exactly how distributed control systems are transforming the future of so many industries. To do this, we'll be speaking to Mike Williams from Modern Automation Consultant Services based in Boston about
the workforce of tomorrow. But first we'll hear from Bernhard Eschermann, chief Technology Officer for ABB's process automation business area about the development and trends that ABB predict will impact automation in the future. I started by asking him where exactly it is that we see process automation.
If you look at the industries that we serve, like energy, power, water, metals, chemicals, farmer, and so on, typically they operate around the clock either with a continuous production flow or running a sequence of production batches. So for example, if you open the water tap at home, you just expect water to flow all the time and without
interruption. And while all of these commodities are central to our everyday lives, they're actually provided with a relatively small number of people running such facilities around the clock. And the automation systems thus provide a large and growing well population with a reliable and cost- effective supply of these
basic needs. And they optimize productivity, ensure product quality, reduce risks to people and the environment and optimize resource and energy consumption.
Got you, just a few things there that they do, but let's focus in on the distributed control system. So DCSs, so can you tell me a bit more about how they work?
Obviously, it's a very abstract subject, so I always try to compare that with something that people know. And the thing that people usually know best is their own body, and the automation system is like the nervous system and the brain of the human body. So it collects sensory information from all parts of the body and triggers certain reactions.
Industrial plants are obviously much bigger than human body and thus the nervous system in the brain are distributed over this large expands, which is why the automation system for such industries is known as distributed control system or in
short DCS. And if you look at these distributed control systems, they can be found pretty much anywhere in industry where critical processes need to be monitored or controlled, and the sensors that they rely on our instruments measuring properties like temperature, pressure or flow, and the muscles, if you compare that with the human body that they use to control the plant or industrial equipment like motors, valves, pumps,
stuff like that. And in a larger industrial plant, you've got tens of thousands of such sensors and actuators. Now in between you've got communication networks and industrial grade computers, and they actually decide on what needs to be done.
As an example, to ensure the safety of a process, you might measure the pressure in the tank and the computer would reduce the flow into the tank and increase the flow out of the tank by communicating appropriate signals to the equipment connected with the pipes going in and out of the tank.
And if the pressure becomes too high, obviously that might be dangerous, and thus this whole thing has to be done with an extremely high reliability of the computer and the communication system, which is why multiple redundant components might be used. They monitor each other or are able to jump in if needed and thus provide for uninterrupted safe
operation. You also have all of these different levels in industrial control system from keeping the pressure in the reaction vessel safe through controlling the overall reaction in the vessel to control the overall production in a part of the
plant to coordinate multiple plants. For example, you might need the steam output from one plant to feed production in another plant, and ultimately you have to decide what specific product variant you want to produce anyway, some level you
move then from automation to human decision making. At this interface between automation and humans, it's critical to provide the needed information to human people, to the plant operators and other staff through computer work stations or so- called human machine interfaces, and which are typically located in a so- called controlled room. So it has to enable a person
to deal with the huge complexity of a plant. And another function needed in the DCS is to allow for engineering changes as raw materials or production processes change over time or certain production equipment is replaced. And by the way, the control rooms do not necessarily have to be in the plant like on an offshore gas platform in rough seas. Instead, you might have the operators sit in a cozy office somewhere on that.
In terms of as industries are obviously developing, how does a distributed control system support them? I suppose as they're going through this fourth industrial revolution and what I see as this up and coming fifth industrial revolution, what is the role of the DCS in supporting these changes?
If you look at these process plants, they live quite a long time and you don't tear down and rebuild a chemical plant every five years. While on the other hand, you actually have a couple of Windows versions during the same timeframe. At the same time, market requirements influencing what exactly should be produced are changing
more and more quickly. And so customers need to renew technology to stay competitive, but with this 24/ 7 production, they also want to keep producing continuously while technology might be updated, even if it's only for the newest cybersecurity patches. And ABB has always provided new capabilities to meet these changing needs of the time while preserving the investments
that customers have already made in their plants. That's one of the reasons I guess that ABB has built and maintained the leadership position in DCS over the last 40 years and has been a number one choice according to market analysis for more than 20 years. If you look in the future, one of the newer development central to the fourth industrial revolution is a much higher level of
flexibility. For example, you want to produce personalized medication that is not the standard medication for everybody, but just for one person. We've also been working on the use of artificial intelligence and machine learning to augment human's capabilities for decision making that allow them to focus on the higher level decisions that I described. Another development that we've been
supporting is the digitalization of industries. With the control system having all of these inputs from the sensors, there are huge amounts of data available on the operation of a plant. And an obvious move is to bring the data out of the control system to analyze it based on this to optimize the overall production or the energy efficiency
or whatever of the production. And in order to combine the ability to adapt to these technologies quickly while having a very reliable production that should not be interrupted, our concept is to have a very stable and robust inner core of the system that can be extended however, by additional capabilities without creating risks on the basic operations and their safety.
With all the changes that we're seeing in the industry and technology, we are already seeing I suppose a new generation of digital industrial workforce so people, so could you tell me how this digital workforce are influencing the developments of future distributed control systems? So what are they seeing that's happening now and how are they, I suppose, swaying the developments of the future?
Yeah. If I look at my children, and that's probably true for a lot of this generation Y or the generation Z employees, they are digital natives that are used to interact with technology without ever reading a manual or going through a training. And on the other hand, they don't have the 20 to 30 year experience that the baby boomers have accumulated that have been operating these plants over the last 20,
30 years. And thus, the automation system has to provide support and interaction patterns that this generation is used to as an example, rather than people scanning through long alarm lists, the system has to identify critical situations by finding anomalies autonomously, by looking at the data stream, dig out automatically what mitigated such a situation in the past, and maybe offer the operator to also simulate the result of
an action before actually doing something. And this is exactly where AI and machine learning in today's technology can help.
There is no doubt that we are seeing machine learning and AI integrated into our lives, and as we look to the future, we can see that how we work is also changing. I spoke to Mike Williams from Modern Automation Consultant Services to ask where process automation and distributed control systems come into this?
It's very important when you look at autonomous operations, essentially taking yourself on a journey from manual to automated and ultimately fully autonomous solutions, you must look at that as an investment and future proofing means making sure that that investment is spent wisely. The definition of future proofing is the ability to expand and adapt to the business changes
without wholesale rip and replacement of those legacy systems. This is very important from an economic standpoint. It permits the end user or customer to implement new or upgrade old functionality incrementally, no big bangs here, investment incrementally to achieve
business requirements, which do change over time. Future proofing allows the business to lower the business risk and increase the return on investment without the fear of becoming obsolete or non- competitive.
So what you're saying is the DCS in a way laser a digital foundation that can then be added to bit by bit as new technology comes on the scene?
That's absolutely correct. And fundamentally, the attributes of a future- proof system address the issues of horizontal and vertical integration, specifically horizontally the ability to scale or add to shop floor devices over time, and also vertical integration or extensibility upward from sensor to the boardroom from a communications application interaction
capability. This interaction may involve not only the shop floor, but it may involve design, engineering, supply chain, financial, and even research and development.
So what you're saying is there's this holistic approach, all levels of technology and all levels of people as well I suppose because along with the changes in industry and technology, we are already seeing the workforce adapting into, I suppose, the next generation of digital industrial workforce. So how are the developments in the technology changing the industrial workforce so changing the people themselves, I suppose?
You've probably heard of something called the Industrial Revolution 4.0 or Industry 4. 0. We've entered a new age of digital transformation where computers are fundamentally being substituted for actual human beings. This is because of the need or requirement for
higher computational and communications capabilities. And the question that you're asking is how do we navigate from what is a human- centric environment forward into a computer centric or autonomous plant operations environment using the new technologies? Quite honestly, the result of digital transformation simply is that physical tasks are being
replaced with logical execution executed by remote devices. So simply put the answer is traditional operators are now going to become knowledge workers, and these knowledge workers will be using computers augmenting their human cognitive abilities to solve very complex
problems which they could not solve on their own. And these complex problems are situations like abnormal equipment operation or suboptimal situations where the equipment is not performing as built well as it could. With these new digital transformation tools and techniques. Human beings who are still in the decision making loop can make these decisions quicker with better accuracy
and less variability. Essentially, you're translating less hands into more heads. So the need for human beings is accentuated and the span of control of the human operator is going to increase resulting in less human error and more importantly, less human beings in the hazardous environments which exists in the process industries, thus improving the quality of life while increasing in productivity, safety, and quality.
And I suppose the next stage of automation is the introduction of machine learning and artificial intelligence. So how will these become part of these systems?
That's a very good question and is not well understood in the marketplace today. I like to use a maturity matrix called the level of autonomy, which starts with fundamentally totally manual operations and evolves as DCS control systems have evolved through time over a five distinctive capabilities or levels
of autonomy from zero to five. When you're talking about technologies like artificial intelligence and machine learning, what you're trying to accomplish is even more sophisticated tasks than simple loop control and understand the interdependence of equipment not only with the human being, but machine to machine, using advanced mathematics and statistics to predict the future, not just the current
situation, but predict the future and be able to optimize the processes to achieve the maximum result. An example here might be quite useful. You've heard about autonomous vehicles, autonomous automobiles, multiple sensing devices are mounted on the car and continuous calculations are made to ensure that the vehicle operates
safely in an optimum manner. The same example exists in the process industries, and the objective here is to use artificial intelligence and machine learning to perform similar duties as an example, to extend the meantime to failure of a piece of equipment, thus avoiding unplanned downtime, which disrupts the customer fulfillment process, minimize the maintenance costs, and even improve
the quality of the product that's being manufactured. Now, I'd like to remind you at this particular point in time, one should always remember that a sound level three autonomous DCS foundation is required to enable the advanced capabilities of artificial intelligence and machine learning. I have a saying that I like to
tell my clients. Actionable intelligence has zero value unless it is acted upon in the correct context, in a timely manner. This is the role, the action is the role of the DCS, and the intelligence is provided by artificial intelligence and machine learning capabilities.
So if we looked at this I suppose zoomed out, what are the key benefits of all of this?
The benefits of artificial intelligence and machine learning are in the process industry, specifically related to process safety, product quality, the ability to improve the productivity without eliminating the human being, elevating the human consciousness and being able to achieve, most importantly, our sustainability goals, which is our right to
operate or permit to operate. So one of the emerging issues out there, benefits of higher levels of automation and artificial intelligence, is to achieve our sustainability goals to protect our planet. Let me give you an example. Process plants are huge consumers of energy and water resources required to sustain human life. Consuming these resources in a responsible manner
is the objective of companies today. More than 80% of the C- suite executive officers see sustainability is a prime objective of them remaining in business and being profitable. To achieve this automation and an advanced autonomous solutions like AI and machine learning are required to meet these business objectives.
Such an important part of it, is sustainability not just from a planet friendly point of view, but from a continuous production point of view as well with things like the pandemic that showed us that automating your processes meant that they could continue in a way that if they were manual, they just wouldn't be able to.
Right, the process industries exist to improve the standard of living and those things that we enjoy as part of our livelihoods. And to achieve this in an economical way and meet the objectives of sustainability, we believe in the process industries that automation or the in industrial re revolution 4.0 is absolutely essential. The digital transformation is absolutely essential to achieve mankind's goals.
I absolutely couldn't agree more. And it's not just the digital workforce that process automation is transforming. This technology can also be used for so many industries. So it's back to Bernhard for a final word on how a ABB's technology is transforming industries from energy right through to the world of fashion.
Obviously, two of the most important topics to make human life more sustainable are circularity to avoid waste creation and reducing carbon emissions by increasing energy efficiency and the use of renewable energies. Renew Cell is focused on the first of these aspects. It's a fast growing Swedish company specializing in textile to textile recycling. So they close the loop
when garments are worn out or no longer wanted. Some are obviously sold secondhand or used as hand me downs, but the vast majority today ends up in landfills or are incinerated. And what Renew Cell is doing is that they are changing that with a new technology by dissolving the used cotton and other natural fibers into a new
biodegradable raw material, which is called Renew Cell pulp. And that can be turned again into textile fiber and thus goes into this textile production cycle again. They are using a number of ABB technologies among those ECS that enables the production to be as resource efficient as possible with less material consumption and reduced waste.
Gosh, that sounds fascinating. And that is the very definition of a circular economy right there, isn't it?
Yeah. And obviously while you circulate garments in this case, obviously the process uses energy. And so this is exactly where you also have to look at the second aspect that this whole process becomes as energy efficient as possible and hopefully uses renewable energy to be driven.
Yeah, and exactly switching the conversation now onto energy usage, because energy is the largest consumer of natural resources and when it comes to industrial plants, I suppose shifting to electricity and so having that potential of using renewables I suppose is part of the solution but could you tell me briefly what role the DCSs would play in enabling this clean energy transition?
Yeah, again, I try to use an example from everyday life. Let's assume you sit at home and want to mainly use electricity from the solar panels that you have on your roof, and then you've got certain electricity loads like right now your computer that probably don't want to switch off. Then you've got other loads like the refrigerator, you can switch that off for a certain amount of
time, but probably not too long. And you've got other things like your sauna that you might decide, okay, I don't want to operate at this point in time because there's no sun right now. And all of these decisions are now decisions that you take, but you've got hundreds of similar decisions to take in an industrial plant that
uses both conventional and renewable energy resources. And at the same time, obviously, the plant needs to produce something to create a payback for operating plant. Another example that you can actually look at is this topic of carbon capture and storage, trying to get CO2 out of the air and moving the carbon out of that in a safe
place where it can be stored forever. This is a technologically very complex process that needs lots of equipment that again, needs to be controlled in the best possible fashion and as energy efficient as possible. So again, it's a typical application for a distributed control system.
So when it comes to the future, Bernhard, what excites you, I suppose, when it comes to DCSs and also process automation in general?
Fran, if I think back 20 years ago, computer and communication technology created a lot of constraints that we had to deal with and that we had to build the system around. And now most of these bottlenecks are gone.
So we've got all of these great new technologies at our disposal, powerful computers, powerful communication systems, AI, machine learning, augmented reality, all of these things that means that instead of focusing on the constraints of the technology, we can really focus on creating value from this technology now.
What a pleasure it was to speak to both Mike and Bernhard about these distributed control systems and what an insight they gave me and hopefully you about how these systems are at the heart of process automation and these systems that we all take for granted and take advantage of, and also how these DCSs can future proof these processes, so they're there for this growing population. But unfortunately,
that is it for this episode. Of course, a massive thank you to Mike Williams and Bernhard Eschermann for their brilliant insight and expertise. I'm Fran Scott, and the Process Automation podcast is a Fresh Air Production for ABB. Follow or subscribe now for free wherever you get your podcasts so you never miss an episode.