Partner: Pathfindr

Advances in data processing and technology offer opportunities for big production gains through upgrading factories. By adopting technologies like AI robots, augmented reality, and increased data processing factories can be turned into smart factories.

The Advanced Manufacturing Research Centre at the University of Sheffield is working to develop and trial technologies that could be used in factories of the future, working with companies in the advanced manufacturing space to bring forward the smart factory revolution.

There are also companies, such as Pathfindr, working in data collection to make identifying the hidden factory easier for manufacturing companies.

Hidden factories

In the late 1970’s Dr Armand Feigenbaum was writing about his concept, Total Quality Control, and one key element of the concept was the hidden factory.  The hidden factory is the part of organisations that doesn’t show up on balance sheets but leads to hidden costs and areas of inefficiency. Costs caused by things like production errors or supply chain delays.

Dr Feigenbaum estimated that companies could be losing anywhere between 20-40% of their production capabilities to the hidden factory, but with the improvements in data processing power, using technology to identify and rectify the hidden factory is more possible than ever.

Industry 4.0

The terms, Industry 4.0, Smart Factories, and Industrial Internet of Things, often get used interchangeably despite some key differences. Firstly the Industrial Internet of Things or IIoT, refers to an interconnected network of devices or sensors in a factory. IIoT is the data collection element of a Smart Factory. 

Industry 4.0 or Fourth Industrial Revolution is a more generalised view of how technology advances, like the IIoT, will make modern manufacturing more efficient.

Finally Smart Factories are simply a manufacturing facility that embraces the above concepts and modern advanced technologies to make their factory more efficient.

Real time data

The underpinning of a smart factory is real time data. Pathfindr is a technology business that uses sensors in a factory or across an entire logistics chain to extract information and present the data on a user software platform.  Their sensors give real time information on the location of an item, within a warehouse or as it travels around the world, they also give information on the environmental conditions an item is experiencing as well as any micro movements it might face.

Important tools can be tracked down quickly reducing wasted time and product quality and condition can be tracked so any issues can be addressed much earlier in the production process.

Companies from across advanced manufacturing use the sensors across their supply chain. A steel producer can track the condition of the steel as it moves around production facilities. A petroleum company is able to track their containers as they travel on ships around the world and keep exact data on the amount of product in the container. Other industries like car manufacturing that rely on just in time supply chains, can track all their shipments and know the condition of what’s being shipped.

Andrew Scheer the marketing director for Pathfindr said about working with car manufacturers, “We’ve been able to provide them with the management information by using our technology to really understand how to maximise the benefit of packaging to ensure that actually those parts arrive from one side of the world to the other in the best possible condition. And really, we’ve been able to show and uncover a level of detail that they’ve not even known existed previously, which has really made a massive difference in terms of how they manage their logistic chain.”

Advanced Manufacturing Research Centre

The AMRC at the University of Sheffield tests and develops new technologies that could be applied in advanced manufacturing. Alex Godbehere is a Technical Fellow for Smart Factories at the AMRC and he explains why the AMRC is necessary for the development of manufacturing technologies.

“What you find is that research tends to be relatively cheap to prove out and develop in a kind of a lab-based environment. But it can be quite expensive to kind of de-risk and mature and typically, academia can’t afford to do that and industries aren’t really prepared to accept it at that level of maturity. So there’s historically a bit of a disconnect. So, we act as an intermediary between the university and our industrial partners. And so, what we do is we effectively make the mistakes for them in a kind of a safe but representative environment.”

The AMRC has built what they call Factory 2050, which was originally designed to be a reconfigurable manufacturing facility, the idea being that anyone could rent time in the factory which was capable of building anything.  Now Factory 2050 instead focuses on digital autonomous technologies and proving their viability in commercial settings.

Augmented reality

One technology that Alex Godbehere and the AMRC think will have a big impact in the manufacturing sector is augmented reality.

“There are a number of uses for it. One of the ones that I think is really interesting is the use of personalised work instructions. So typically, when we’re manufacturing a part, we will have a set of work instructions, which will say, you know, insert bolt into this hole tightened to this talk. Historically, they will be on paper printed out more recently, they’ll be on some sort of computer screen or terminal. But kind of the real cutting edge of that is displaying them on a piece of augmented reality equipment.” Alex Godbehere said.

Augmented reality could be used to help train employees and help to improve their productivity, but beyond that it could be used to verify and validate work which will reduce errors and waste and begin the shrinking of the hidden factory.

As Alex Godbehere explains, “So if you have to insert a bracket, for example, the cameras on the device that it uses for tracking could also potentially be used to say, Okay, we validate that that component is in the right position, for example, all automatically.”

Autonomous guided vehicles

Another technology being trialled at the AMRC are Autonomous Guided Vehicles or AGVs. AGVs can be programmed to move components around the factory, however the largest benefits of AGVs will only be realised if operating within a high level smart factory and can respond to the large scale data processing and help fix issues that arise quickly.

“If, for example, we’re producing a component and we are when we have a number of individual assemblies at different stages of completeness. And they’re moving down a production line, and one of the machines goes down, it breaks. If we have a factory that is able to detect that and react, it could, for example, instruct the AGVs to route the components through a different route, a different process a different way around the factory, or without any kind of human intervention.” Said Alex Godbehere.

An increasing use of AGVs within manufacturing leads to fears over whether there will be human roles in the process. A professor of management from the University of Southern California once said the factory of the future will have only two employees, a man and a dog. The man is there to feed the dog and the dog is there to stop the man from touching the equipment.

However Alex Godbehere doesn’t share this view and thinks that like previous fear over technology taking over jobs, it won’t come true and other jobs will come up.

“It’s also created a lot of high skilled and highly paid jobs as well. I think, fundamentally, the jobs need to be fit for purpose. Humans are really quite clever and are much more suited to non-repetitive tasks than robots. So in the future, there’s going to be a huge demand for cloud things like collaborative working between humans and robots, but also humans to kind of use the data, say the data. So data tends to be refined, right becomes knowledge. And humans can then act on that knowledge. So to summarise that there will be humans in the loop. And they will be making the key decisions.”

Smart factory implementation

A common concern that both Alex at AMRC and Andrew Scheer from Pathfindr come across is the cost and time of turning an existing factory into a smart factory, however that isn’t necessarily the case.

As Alex Godbehere explains, “So one of the common misconceptions that we tend to come across with smart factories is that they must be deployed on, to kind of, Greenfield sites. So you need to start from scratch, you need to deploy a factory with cutting edge kind of networking and infrastructure. But any factory really can at least slightly become a smart factory.”

The AMRC released a free publicly available blueprint they call Factory Plus. Factory Plus shows how an existing factory can take steps to implement technologies and elements of smart factories to improve production capacity. The AMRC wants the blueprint to raise awareness for manufacturing technologies and show that the transition is easier than many expect.

Pathfindr have also encountered some hesitancy around implementing smart technologies. “I think there are massive concerns in terms of adoption, and their journey of digital transformation. And I think these principally revolve around historic problems, or historic challenges with implementation of this type of project, which have often suffered extensively from mission creep, and cost creep.” says Andrew Scheer.

However, Pathfindr’s technology involves very limited installation. The sensors and detectors they use are equipped with their own power source so don’t need to be connected to the facility and use Bluetooth or GPS depending on the technology. This means all that is required is attaching the tracking sensors to the assets that need to be tracked and then setting up the user software platform.

While Pathfindr’s technologies don’t turn an ordinary factory into a smart factory on their own, it is an easy first step to increasing data collection and revealing areas of the hidden factory that can be improved.

Andrew Scheer thinks Pathfindr is an important first step towards smart factories. “It becomes an easy gateway into the benefits of what ultimately our digital transformation, industry 4.0 smart, smart factory, and whilst it’s not a complete solution, it does provide some really key building blocks with which to build that type of solution. So for us, we believe we can help 99.9% of organisations improve what they’re doing quickly and cost effectively, because of the nature of our technology.”

Pathfindr challenge

Pathfindr believes their technology has useful applications across many industries. However, discovering use cases for the technology is much easier for companies themselves.

For this reason Pathfindr are launching a competition for companies to put forward exciting and innovative applications for the technology and the best use case will be awarded 12 months of free access to Pathfindr technology.

Reaching industry 4.0

Technologies are still being developed and tested and smart factories and the technology they use will continue to evolve. The process of becoming a smart factory isn’t as simple as implementing a certain technology.

Advanced manufacturing has been quick to adopt modern technologies but being a smart factory and achieving the fourth industrial revolution is a process.

Alex Godbehere explains “Manufacturing is typically quite far behind the curve when it comes to adopting these digital technologies. And if you kind of take that and couple it with the fact that as organisations grow, typically, so to their digital requirements, and what we tend to see is not always right, but the general trend is organic development of digital architectures. They become, I guess, almost Frankenstein, like, so a new technology comes along that solves the problem, so we’ll bolt it on. And then another technology comes along, that solves the problem. So we’ll bolt it on.”

While certain technologies will only be implemented further in the future that doesn’t mean the process of becoming a smart factory shouldn’t start right now, and data collection and processing is a way to identify areas of the hidden factory and beginning to look at which technologies that could be implemented would have the biggest impact on reducing the hidden factory.

Andrew Scheer thinks that taking the more manageable steps in the direction of Industry 4.0 is the best way to be ready for more technological advancement in manufacturing, “Whether [Industry] 4.0 will still look like [Industry] 4.0 in 5 to 10 years, I don’t know. There’s still a lot of work to be done across a whole host of organisations in order to get anywhere close to that. Which is why yes, you can set yourself a massive objective of Industry 4.0. But actually taking smaller, much more manageable steps in that direction will also deliver a lot of benefit. And that’s, I suppose, the message that we are pushing and pushing hard.”

Alex Godbehere says that the adoption of these technologies has been accelerating fast, “[Companies] at the top of the supply chain, are seeing the benefits of applying this kind of technology outside of the factory. So across the supply chain, which is, I would say, probably accelerating the adoption of that within maybe some smaller companies within that supply chain. But I would say confidently that there are manufacturing companies or organisations which are firmly in the Fourth Industrial Revolution and have been for quite a long time.”

This article is based on episode #168 Industry 4.0: Finding the Hidden Factory, click here to listen