digital manufacturing

Digital manufacturing: An overview of an increasingly complex scenario

Share this story

Digitalisation is, in essence, simple. It means exactly what the word implies – the digitalisation of everything. 

But as simple as it sounds, digitalisation can be applied to any business in any sector, and that is what makes it difficult to provide a simple overview.

If we stick to one sector – manufacturing – that might make it more manageable.

But since manufacturing is already a hugely complex endeavour, the process of digitalisation – even in this one single sector – encompasses so many different disciplines that it’s difficult to think of it as being, essentially, simple.

But enough complaining about having to do some work for a change, let’s get on with the show. 

 

In this article, we will try and bring together as many of the facets of digitalisation in manufacturing as we can, in order to provide as complete an overview of the entire digital manufacturing landscape as possible.

We’ll probably fail miserably, or heroically, depending on your point of view, but it’s definitely worth a try since digital manufacturing is, actually, very interesting.

Fake it ’til you make it  

First, let’s list a few definitions of some of the jargon used in digital manufacturing.

But even before that, let’s deal with the difference between digitisation and digitalisation.

Basically, there isn’t any difference between the two.

The reason why digitalisation – with the al in the middle – became more commonly used is probably because when it was first discussed and debated, the people who talked about it mostly used that version rather than the shorter one, and it just caught on from there.

But if we want to be finicky, we could say that digitisation is the simple process of digitising something like, say, a component or part, perhaps on one isolated, unconnected computer, while digitalisation is the process of digitising entire operations and businesses – from individual components, through factories, to everything else, all of which are connected on a computer network of some sort.

However, it’s probably not necessary to distinguish between the two, since, as words, they both mean exactly the same thing.

It’s probably worth pointing out, though, that digitisation is probably more commonly used outside Germany, which seems to have taken a lead in articulating many of the trends people in the manufacturing sector talk about today – digitalisation and Industry 4.0, for example.

Industry 4.0 is sometimes used interchangeably with digitalisation, but strictly speaking, Industry 4.0 is a term that describes a historical phase – as in, the fourth industrial revolution, in which everything is being connected to the internet, which is something that hasn’t happened before, or wasn’t as widespread.

The logic is that the first industrial revolution was brought about by steam power; then, the second was about electrification; and the third was when computing technology was introduced, although it was mostly used in isolated work cells; the fourth – or Industry 4.0 – is the process of connecting machines to larger computer networks.

Industry 4.0 does not really refer to a specific technological process as such in the way digitalisation does – it refers to a time period in which certain things are happening.

A term like “Industry 4.0 standards” could be used to mean “up to date” with current industrial technologies, specifically connectivity.

Connectivity – as in, connecting industrial machines to sensors which are, in turn, connected to a computer network – is critical to digitalisation, and we’ll look at connectivity technologies later on, but let’s not overcomplicate things now. There’ll be plenty of time for confusion.

Digital jargon

Now that we have some of the broader, contextual definitions in place, let’s see what’s left of digitalisation to explain.

There are many different facets of digitalisation, with terminologies and jargon referring to each one, and we’ve listed as many of them as we could find below, along with a brief explanation of each.

Digital manufacturing jargon includes:

  • digitisation: as explained above, this is essentially the same as digitalisation, but in this article, we’ve decided to use it as a component of the broader process of digitalisation;
  • digital thread: often described as referring to a communication framework which connects all the different departments and individuals who work on a product or component; in the past, many of these departments and individuals may have worked in isolation or disconnected from each other; a digital thread could theoretically connect research and development to manufacturing, and manufacturing to logistics, and everything else to everyone in the business; it could even include the customer feedback, which may be used to the modify or improve the product or process in the future; brings to mind the old phrase, “How long is a piece of string?”; connecting everything to everyone is probably unnecessary and might be a bad idea when trade secrets are involved, so the critical questions to answer then becomes, “Who needs what information and when do they need it?”;
  • digital twin: while digital thread refers to the process of developing and manufacturing a product, digital twin refers to the product itself; obviously, we’re talking about a digital version of a complete product or its individual components; and it doesn’t have to be exclusive to discrete – or product – manufacturing, it can also apply to continuous manufacturing – such as oil, gas, chemicals and so on – particularly as new, composite modern materials are being developed; but mostly it’s a term used in discrete manufacturing, where a digital twin can include all the data relating to a material, component, and assembled product – its dimensions and geometries, how it looks, how it behaves, according to real-world physics and so on;
  • digital product definition: used interchangeably with model-based definition, digital product definition refers to the practice of including all the data about a product in one place; so, for example, if you have a virtual model of a smartphone on your computer, it would include all of the data relating to the smartphone as a product – from the dimensions and materials used in the enclosures to the processors and software used for the operation system, and everything in-between; it can be the same as the digital twin, but digital product definitions tend to include only high-level data, excluding data which is less relevant to the people dealing with it at that stage; so, for example, while the materials and components might all be named, the physics of how they behave under a variety of conditions might not be included; however, the level and depth of data available depends on each company’s requirements;
  • digital factory: increasingly these days, companies are designing robotic work cells and and even entire digital factories inside a computer, and getting them just as they want them, before starting actual, real-world physical construction and set-up; the level of detail that can be included in a digital factory designed inside a computer is quite amazing, with virtual humans and robots and all kinds of machinery and conveyors available to systems designers; it follows, then, that that digital factory inside your computer could be used to simulate the manufacturing process of that plant, and could be continually used – even after physical construction of the facility – to monitor and manage the actual facility, using sensors connected to the machinery and everything else, including the human workers if required;
  • digital manufacturing: by now, you’ve probably realised that there are some overlaps between the various terms, and digital manufacturing has some overlap with all of them for obvious reasons; but in simple terms, digital manufacturing refers to the utilisation of computer technology in every aspect of manufacturing, from development through to production; it’s probably possible to find manufacturers that still use pencil and paper to design and develop products which then go straight into manufacture – without any computers involved anywhere – but it’s probably not very common; but digital manufacturing does imply a much higher level of computer technology – possibly for everything, from one end to the other; however, obviously we are still talking about manufacturing of a physical product at the end of the day, rather than just virtual products, like those you might find in a computer game;
  • digital model: this could probably be said to be the same as the digital product definition term, although it could imply less detail; a digital model may not necessarily contain any data about physics, but then again, since digital models tend to be ready to use in additive or subtractive manufacturing processes, the digital model tends to contain data such as correct dimensions and geometries at least;
  • digital prototype: this is similar to the digital model, except that whereas a digital model tends to refer to a virtual object that is ready for manufacturing, a digital prototype is used more for internal discussion and development in preparation for finalising a digital model or a digital product definition; obviously, a digital prototype can be the equivalent of an initial sketch or as complex as the finished product itself – just depends on what stage it’s at; and, of course, it can be accessed by different departments – from research and development to sales and maketing, and manufacturing in-between – but that also depends on who the company thinks should be given access to the data;
  • digital automation: probably a less well-known or used term, but could be said to refer to the part of the manufacturing process which deals with automation technologies, such as robots; as manufacturers look for ways to increase automation in their processes, this aspect of digitalisation is likely to become more important; however, automation is probably inherent in every stage of planning – meaning, the whole digital manufacturing trend is aimed at achieving higher levels of automation, which usually result in higher productivity and greater efficiencies;
  • digital control: this is more involving than it sounds, since it tends to refer to the computer hardware used to control machinery and, increasingly, data; so, for example, you could have what are called “edge” devices – essentially small computing devices – which might have one single function, such as operate a piece of equipment; a digital control system could also take the form of a desktop computer or even a cluster of cloud computers, configured to act as one controller for an entire operation; the level of autonomy, or parameters of control, would depend obviously on how the digital control system is configured;
  • digital transformation: this is probably self-explanatory, and refers to the transition from traditional manufacturing processes to computerised, digital processes, whether it be swapping pencil and paper for a computer in the design department, or the attaching of sensors to previously isolated machines and equipment and connecting them to a network, which can then be monitored on computers both at a high level and at a minute, machine level, or “granular” level, as some might say.

There are probably many other glossaries worth looking at if you want to learn more about the phrases used in manufacturing these days, including one published by the US government under the title Glossary of Advanced Manufacturing Terms.

However, we thought we’d pick out the ones that used the word “digital” since it’s the word that connects everything in the digitalised world, and is specific to this article.

Terms of relativity  

If you do look at other glossaries, you will see that there are many other terms which may not use the word “digital” but which necessitate the digital aspect.

Perhaps you’ve heard of many or all of them, and know what they mean. But since we’re on a list-making binge today, we thought we’d mention some of them

Terms related to digitalisation and digital manufacturing

  • mass customisation: some call this “product of one”, meaning that, at some point in the future, digital manufacturing will become so responsive to customer requirements that it would be capable of making a one-off product for each and every one its many customers through being able to customise each product for each individual customer; at the moment, a manufacturer may offer a limited or broad range of colours or functions for one particular product, but in the era of mass customisation, there may be as many variations of that one product as there are customers for it; how this can be achieved – let alone whether it’s a good idea – is still being debated, but many people say additive manufacturing, or 3D printing, advanced robotics, and perhaps materials science hold the keys to mass customisation;
  • cyber-physical systems: this could, for example, refer to a factory which has been designed in a virtual environment – as outlined in the digital factory definition above – and takes account of all the humans and machines involved in the process; some people point to collaborative robots as a good example of a cyber-physical system because a human would be working closely with that machine; the word cyber may bring to mind cyber security, but on its own, “cyber” relates to computers; so, a cyber-physical system is one where a computerised, or digitalised, production process includes human participation; it could also be seen as a combination of computers and moving machines… without humans;
  • advanced manufacturing: this is a catch-all term referring to all of the new technologies being used in manufacturing now, as well as the ones being developed for the sector; most people consider 3D printing, or additive manufacturing, to be an example of an advanced manufacturing technology; 3D printing is a good fit for this term since not only is it already in use, the technology is still developing, and holds the promise of much more; as mentioned earlier, mass customisation would be possible through 3D printing; materials science has become particularly interesting of late because most 3D printers can only deal with plastics, really expensive ones used in industry can deal with metals, but what’s probably most appropriate is a new generation of composite materials; materials science too will see many advances through the use of virtual materials, with accurate physics, which can be mixed together and tested – all in a virtual simulation environment – before being produced and used in components or anything else;
  • smart factory: another term which may be considered to be similar or the same as digital factory; but perhaps whereas a digital factory, on balance, refers more to the properties of a factory, smart factory may be seen to be about the process, or the state of a factory; in other words, a digital factory is about the factory as a production facility, smart factory is about the factory as a production process; however, that’s just our view – the two terms are often used in similar contexts, and some even use smart factory interchangeably with Industry 4.0;
  • smart manufacturing: if you thought distinguishing between smart factory and digital factory was a somewhat convoluted process, try separating smart manufacturing from smart factory; but saying that smart manufacturing can refer to a chain of integrated factories whereas smart factory refers to just one facility might be a good start; but smart manufacturing is also a term that is used interchangeably with Industry 4.0 to mean the whole trend of connecting machines up to the computer network, digitalising everything and using advanced manufacturing technologies and processes;
  • Industry 4.0: probably enough said about this term, but just for the record, it’s a useful and interesting term; it articulates a standard without actually being an official standard in the technical sense; and rather than it being an unreachable aspiration, it simply refers to getting machines online and extracting data from them so you can see how they’re operating, which isn’t all that difficult, and by all accounts has a lot of commercial benefits.

Where were we? 

While many readers of this website will be experts in these things, and perhaps this article hasn’t enlightened them any further, but there are some of us who are new to the area and appreciate simple explanations of basic technologies and concepts.

Also, there are other areas which relate to digital manufacturing which might not be obvious to those of us who aren’t engineers, or involved in manufacturing, and it’s probably worth pointing some of them out.

  • materials science: as suggested in the 3D printing outline earlier, materials science holds the promise of an entirely new way of making products; it may not be quite as advanced as the magic microwave on the old Star Trek TV show, but current advances in the field certainly put us on the journey to that shop that sells those microwaves at the very least; we’ll try and look deeper into this subject, especially since a lot of readers visited the article we published recently about graphene;
  • data science: these days, it seems every other person is a data scientist, which is lucky for the manufacturing sector since a lot of new types of data are being generated by newly connected machines; and while some obvious applications of that data have been found, there are many that can probably only be discovered or unlocked by data scientists in combination with computer programmers who know how to develop artificial intelligence software, using big data as their raw materials;
  • additive manufacturing: it’s been implicitly explained earlier in this article, but additive manufacturing is used interchangeably with 3D printing; “additive” refers to the fact that, in 3D printing, material is added, layer by layer, to build an object, whereas the traditional process is “subtractive”, in that it starts with a piece of material and takes away bits of it to leave behind the shape and object that was required; this is something probably most readers know about;
  • software engineering: this is another familiar term, since it’s been a long time since Google and Apple took over the world; but it’s likely that software engineers will become much more important to the manufacturing sector because of digitalisation;
  • systems engineering: an interesting area which may be something that grows in importance as everything becomes simultaneously more complex and more interconnected; like a smartphone or any computer, which has so many hardware and software systems all working together, the digital manufacturing process also features many systems and sub-systems working together; this may increasingly require more systems engineers, not just for designing individual products and facilities, but the entire process, across many factories in many countries, especially as the internet of things connects everything up.

Delegating everything to the machines 

The origins of the phrase “digitalisation” are lost in the mists of time, but most likely, it was first articulated in Germany, where the term Industry 4.0 comes from.

The process of digitalisation, or computerisation, was already ongoing globally, so it wasn’t that the process was entirely new, it’s just that no-one was so idle that they had time dream up names for these things.

Is there are a point to giving them names? Maybe. It certainly clarifies the concept and enables the discussion of what can be a complex topic to be understood by a wider variety of people, many of whom might be directly involved in manufacturing, which would probably result in a better understanding of the products and services required, and a shared sense of direction and goals.

Ultimately, digitalisation has enabled, and is continuing to enable, the manufacturing sector to become more efficient and productive.

Through digitalisation it is possible to micro-manage a business; it is also possible to see a global, high-level picture of an operation so you can make better-informed decisions on a macro scale.

Some companies have gone quite far down that route, but there are many more which are on the way.

Within this framework of digitalisation, there are many tools already available and many more being developed.

Already in existence are applications in the following categories:

  • product lifecycle management;
  • enterprise resource planning;
  • manufacturing execution systems;
  • computer-aided design;
  • computer-aided engineering;
  • computer-aided manufacturing; and
  • many others.

And these are distinct from other software categories that are also already available, but which perhaps some manufacturing businesses may not have used as much in the past.

For example, network management software might become a standard component in a manufacturing company’s information technology requirements.

And given that the physical movement of an object in three-dimensional space in the real world generates a colossal amount of data, it follows that the amount of computer storage and processing power that the manufacturing sector will need will far exceed that of, say, the software development sector.

Even a massively-multiplayer online role-playing game played by millions of people over the course of, say, a year probably generates less data that one single, large, fully digital manufacturing company would in that same time period, especially if its customer or product usage data circles back to the research and development department, which might be using supercomputers or cloud clusters to simulate iterations of a large range of components and assemblies.

If digitalisation, or digital manufacturing, is to work, is to function properly, it probably will require faster, more reliable data networks as standard, which likely means some type of network monitoring software will be standard in most production processes, even if it’s one acquired by an individual manufacturer in the form of software-as-a-service, or a managed service.

Network management software generally sends alerts to let you know that something is amiss or needs your attention. You can choose to deal with it manually, or maybe set up some parameters within an automated process to which you can delegate some of your decision-making.

The next step is obvious: the machines take over the world.

With a variety of companies developing artificial intelligence systems, using machine learning and deep learning methods to train their computer models, it’s inevitable that we will find more examples of fully automated digital manufacturing processes.

And not just automated through a computer, but also managed in an optimal way through an AI digital manufacturing manager of sorts – perhaps using the parameter-based automation example outlined above in the network monitoring example.

Here, robotic process automation and business process automation applications – many of which integrate AI – also come to mind.

The Star Trek microwave may be a long way into the future, but it will be an interesting journey getting there.


Share this story