This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.

The robot isn’t the problem ...

12 May 2015

Attendees:  (L-R) John Rainer, Keith Thornhill, Dennis Allison, Alan Hunt, Nigel Dawson, Grant Collier, Mike Wilson, Rich Walker and John Rowley
Attendees: (L-R) John Rainer, Keith Thornhill, Dennis Allison, Alan Hunt, Nigel Dawson, Grant Collier, Mike Wilson, Rich Walker and John Rowley

On 18 March 2015, nine specialists working in the food and beverage robotics and automation industry gathered to discuss the latest issues and topics that are affecting the industry, moderated by Helen Bahia, Editor of Food Processing.

On 18 March, at Festo’s Applied Automation Centre in Northampton, the leading companies in UK robotics and automation were represented at Food Processing’s Robotic and Automation roundtable event. Joining us on the day was: 

Dennis Allison, Managing Director at Pacepacker Services
Grant Collier, Head of Marketing at PPMA
Nigel Dawson, Product Marketing Manager at Festo
Alan Hunt, Sales & Marketing at KUKA Robotics UK
John Rainer, Regional Sales Manager at FANUC Robotics UK
John Rowley, Sales Manager Food & Beverage Industry at Mitsubishi Electric
Keith Thornhill, Business Manager Food & Beverage at Siemens
Rich Walker, Managing Director, Shadow Robots
Mike Wilson, General Industry Sales Manager Robotics at ABB

When asked what type of processes are suitable for robots and where should the food and drink industry be looking to automate, there was a lot of discussion around the end of the line with packaging and palletising. “Historically, the food and drink robots have in the main been at the end of the line,” says Rowley. “But recent developments in technology in terms of grippers and vision systems are opening up new opportunities in the food and drink area.”

Palletising and packaging are easy processes to automate because the products are consistent, and involves handling packets rather than food products themselves. “But there have been successful applications, in the meat industry for example with slicing and stacking,” says Rainer. “And there are far more processes that are suitable for robotics but there’s got to be a willingness to put in a high level of investment. While people are reluctant to invest, it’s easier to work with the end of line process.”

But according to Wilson, the investment needs to be focused on the grippers rather than the robot: “It’s not the robot that’s the challenge. When you’re handling a product without packaging that varies in size and shape, then it’s the gripping technology that has to be developed. When you’re holding products that are wet, or sticky, or difficult to pick up, it’s the grippers that become the biggest challenge.”

And of course there are challenges with hygiene requirements and wash-down, meaning that robots used during the process stage will cost more than palletising automation, especially while labour costs are still low. With food companies working on short-term contracts, committing to a long-term, significant investment can be a struggle. And yet if they don’t invest to improve efficiencies, they run the risk of going out of business.

Allison believes the most important step is to work with prospective customers to meet their needs. “We concentrate on secondary packaging rather than picking up raw product but what we find is that the presentation of the product is really key. When you move from a manual process to an automated one, the idea is that you lose the ability to manipulate, which humans are well suited for. You have to change the way the customer understands that process to ensure that it doesn’t hold them back from moving forward with automation. You also have to look at value for money and looking more long-term. We as an industry might not be able to offer a quick ROI every time, but that doesn’t mean it’s not worth investing. Once a company has invested in automation, they never want to go back.”

Return on Investment
The issue of ROI is one that the robotics industry has so far failed to overcome as an argument against automation in the food industry. It’s a tricky subject, as each production line is different, each solution is different and the cost of the robot is merely the tip of the iceberg. 

A project that initially looks at a simple robot could end up being five or six times the cost of a robot by the time the customer has looked at conveying and guards and integration, for example. And a project that was originally priced at £40,000 can now run to hundreds of thousands of pounds. 

“It’s convincing the customer to move away from looking at cost and towards investment instead,” Allison says. And that includes employee engagement, which every manufacturer and integrator agreed on. Most staff won’t have to be retrained to use the new technology, which is a big concern for end users. The existing staff should be able to operate the technology after training and support.

“Perhaps rather than focusing on what the cost of robotic installations is, customers should look at what the cost will be if they don’t invest,” suggests Rainer. “If you’re looking to invest in robotics, there has to be a justification for that investment. And there are so many intangibles to consider that perhaps aren’t, like employee injuries or the cost of continuing labour.”

And one of the big questions always raised is whether companies can get their ROI under a year. As a general ballpark figure, the cost of a robot over 12 months can be repaid if a company runs three shifts a day. But in the food industry, where wash-down and hygiene usually prevents a production line from running three shifts, payback is more likely to be within two to three years. “Companies are missing the bigger picture, which is that if a robot lasts for ten years, then after the first two years the robot is paid for and running on pure profit,” says Wilson. “And the other thing to consider is leasing, which provides a payback on day one. The problem is that most companies don’t want to lease, but there’s no reason why they shouldn’t.”

When you take into consideration that of a £100,000 installation, only around 30% might be tangible asset with the rest of the cost taken up in software, installation and labour, for example, it seems that leasing should be a viable option at least. All the manufactures and integrators agree, but how do they convince industry to take up a lease? As yet, no one seems to know. But it’s something to consider for the future in an industry where short-term contracts cause companies to shy away from large investments.

“I think we’ve got to be better at putting our business case forward,” says Thornhill. “We have to provide customers with irrefutable evidence that makes the decision easy for them. We also have to present it correctly to remove all concern over risk.”

Wilson agrees: “I think one of the challenges we face is that we deal with production and the engineering team, which is of course necessary but we don’t often get to speak to the finance department, who are responsible for signing off on expenses. Speaking to them and helping them to understand the bigger picture would make the process easier.”

The process
Automating any section of the production line can be complex, and it’s always specific to the end user. But perhaps the longest step of the process is getting to the point of approval. Many manufacturers are wary to take that final step, and robotics and automation companies find themselves advising food and drink customers. There is, perhaps, a lack of experience in the food industry that isn’t found for example in the automotive industry, who know what they want and they can hand over a spec for their needs. The food industry isn’t there yet, more often than not asking for help with an inefficient process and leaving it to the robotics companies to create a spec and quote at the same time. 

“Simulation is going to be a big driver,” says Thornhill. “It can be a visual aid for companies and let them see the technology in situ in the factory. It starts to take that risk away for the customer.”

Proof of concept is a big step in getting that all-important approval. Some customers require proof of application, which can require a significant level of investment from the robotics company to set up. But with the food industry bursting with products that don’t fit one mould, or who’s products are easily breakable or requiring a certain level of precision, many companies are not comfortable moving forward without seeing their product being handled by the technology. 

And those variables are what differentiate the food industry from many others. “Even with delicate items like fruit, you can bruise it without even realising because it doesn’t show up for another two days,” says Rowley. “Sometimes you have to wait up to a week after demonstrations to make sure.”

“So you’ve got 100,000 new customers, and every single one of them is a new sale, and they have to be educated,” Walker says in summation. “Robotics is possible, automation can be done, but even if you can convince the customer to go ahead, the next customer won’t believe you because they believe their world is different to anyone else’s. And the cycle repeats itself.”

Which perhaps leads to one of the biggest problems in the food industry; it’s silence. The food and drink industry is notoriously close-mouthed about their practices. There is a lack of collective sharing of knowledge in the food industry that the automotive industry, for example, had thirty odd years ago that developed their knowledge cooperatively. 

There was much talk of the biggest barrier to automation coming from the supermarkets themselves. The supermarkets have come under increased pressure recently for their business activities and one result of this is that the average production engineer doesn’t have time to look to the future and long-term because they’re too busy getting out what they need to make today. 

Rainer pointed out that one difference between the food industry and the automotive industry is that cars are sold directly to consumers, while in the food industry, products are sold through the supermarkets on the whole, and whether their focus should be on those supermarkets rather than the food manufacturers.

“Aldi and Lidl are doing well because they work within a completely different supply chain concept with long-term contracts and their suppliers can look at their own business needs in the long-term,” says Thornhill. “Maybe to get efficiency into the supply chain, the industry needs to look towards companies like Lidl and Aldi and work on longer-term contracts.”

Hunt agrees: “If a supermarket is only willing to give a six month contract, how can companies afford to invest in expensive equipment when there is a risk of no contract in six months? There’s no guarantee or stability.”

Aldi and Lidl operate on a smaller number of SKUs, long-term contracts and good profits, and their suppliers have more stability than those in the UK. The UK supermarkets hold more choice than those on the continent, and questions are raised as to whether consumers need as much choice as they’re given.

“It’s the Apple strategy,” says Walker. “A very small number of SKUs and products, investing cash into their own supply chains and encouraging their supply chain to invest in themselves.” As an example, Apple sells less than 50 products in total. Samsung, on the other hand, currently sells over 140 smartphones, which doesn’t even begin to take into account their tablets, computers, televisions and assorted other products.

The market is also notoriously reactionary, with orders coming in first thing in the morning to be processed and out of the door before the end of the day. And it’s difficult for a robot or automation to be as flexible as humans can be under those circumstances. 

“The German manufacturing industry is 13 times the size of the UK manufacturing industry because they have a much longer-term view on manufacturing, automation and investment,” says Dawson. “The food industry as a whole seems to want to automate and invest in robots but the market won’t allow it because of their short-termism.”

When the issue of ease of use for engineers to programme and manage robots was raised, the general response was that it shouldn’t be difficult for anyone already running a production line. 

“In our experience, a typical basic training programme is five days,” says Hunt. “And after that, the engineer should be able to programme and manage their robotic system. Almost anyone can be trained, there’s no minimum requirement other than a working understanding of the production line.”

Rather than thinking about managing a robot, the group advocated considering it as part of the production line. If the robot is integrated seamlessly, then operating the technology has to be simple because the person running the line isn’t just managing that robot. They’re running the entire production line, so they’re managing what’s coming on and off of the line. In essence, they’re running a process, not a robot.

“With the time it takes to get a project from concept to commission, companies have plenty to time to prepare their staff for the culture change of running a robot, as well as arrange for training and preparation time,” says Dawson. “If you spend time engaging your employees on the investment in robotics and spend time training them, they should be thoroughly prepared.”

Talk turned to whether it’s a terminology problem. Is the term ‘robot’ too intimidating?

“It’s materials handling, when you look at it,” says Thornhill. “It’s still a production line. The term robot suggests something futuristic and difficult to understand, when really it’s just materials handling. All you’re really doing is picking something up and placing it elsewhere.”

Walker points out that a robot is only called a robot until it works. “Once it works, it’s a printer, or a self-driving car,” he says. “When it has a market, the word robot is dropped. Machine tools are a classic piece of robotics, with motion, planning, control, sensing and mechanical design, but it’s not sold as a robot, even though it’s programmed from the ground up.”

Another barrier raised by the group is middle management, who are good at dealing with people but don’t understand automation. They have no use for it or view it as a potential threat. 

Another topic that is often raised is how easy is it to integrate robotics into a production line. There is universal agreement that using an open network is key to integration, and the earlier it’s taken into consideration when planning integration, the better it is. “Increased use of Ethernet-based networks have assisted dramatically,” says Dawson. “Most companies will manufacture technology to sit on almost every network, so as long as end users have an open network, it should be fairly easy.”

Collaborative robotics
The unanimous answer to what the next big thing in robotics would be was collaborative robots. 

“Safe, energy efficient robots that work with people, not working instead of people,” clarifies Dawson.

“It’s the ability to have humans and robots on the same line without the need for guarding,” adds Wilson. “Using sensors, making robots lightweight, making them compliant. And those machines are on their way now through the development line, so that type of technology is getting closer. The food industry will also need to make the robots hygienic and food safe which will take a little longer. They may also be too expensive for the food industry at first.”

The cost of new developments in robotics is an important factor for the industry. Unlike the automotive industry, which produces an expensive, high value product and therefore can justify the cost, the food industry produces low value, cheaper products and therefore the latest thing in robotics might not be the most cost-effective solution.

Of course, the classic conundrum in food production is sandwich assembly. And while no company is close to being able to perform the whole process with robotics, the challenge is that robotics can’t be used anywhere in the process because of the need for guarding and the inflexibility of the technology. “If you can put a collaborative robot in there that can perform some of the operations where applicable, with people working around the robot to perform the rest of the assembly, there’s a better chance of getting automation into sandwich assembly,” says Wilson. “But unfortunately, it won’t be any time soon.”

Looking to other industries
The automotive industry is hailed as a leader in robotics, and for good reason. But the food industry is very different in terms of product, supply chain, longevity and manufacturing process. So is there another industry that food and drink companies can look to?

“Medlab is an interesting growth area in general for the UK,” says Dawson. “It grew by over 60% in robotics last year, and they have similar issues to the food industry such as hygiene requirements. And while their products are relatively uniform, it’s still a massive growth area and the food industry could learn a lot from another industry that has been a late adopter of robotics.”

“And there are still things to be learned from the automotive industry,” insists Wilson. “It’s not just about the technology, but also the culture and path taken by the industry. The car plant is very different to the food factory, but what they’ve learned about applying technology is equally applicable to food and the process they have to go through.”

As the group debates whether the food industry is as slow to automate as is perceived, talk turns back towards the supermarkets and large food manufacturing companies. “Thirty years ago, how many Mars bars, for example, were on the market though?” asks Dawson. “It was one. Now they make fun-size, king-size, the small Celebration size, and then there are all their different varieties with dark chocolate and hazelnut and so forth. Then of course they have to be packaged differently depending on whether they’re going to the corner shop or the supermarket. Mars doesn’t have more production lines to cope with these variations; they’ve just become more flexible. They’re automating so that they can change batches without losing downtime. They’re driving automation themselves.” But not all food companies benefit from the brand recognition of Mars, for example. 

The automation industry itself is doing well, as can be seen by the number of companies signed up already this year to exhibit at the PPMA show in September/October. Whether it’s niche markets like filling, where there are three big companies who cover most of the market, or towards end of line packaging, where there are hundreds of companies competing over business for more varied products. It’s a successful industry and it’s doing well. “There’s perhaps just a lack of visibility for companies wanting to automate,” says Dawson.

Awareness and apprenticeships
In the UK, there is one degree that focuses on food engineering, launched last year by Sheffield Hallam and developed by the Food and Drink Federation (FDF) and the National Skills Academy for Food and Drink. In its first year, 15 students joined the course, which was by no means full. While it’s thought that those graduating from the course will be snapped up quickly by the industry, the lack of promotion by the government and the lack of numbers on the course bring up the issue of why the food industry isn’t perceived to be a desirable industry for students.

Thornhill pointed out that while Sheffield Hallam had funding of £6.9 million to create the first centre for excellence for food engineering, there is a £90 million project to create a Glass Innovation Centre at the Sheffield Business Park Phase 2 site. The funding for the Glass Innovation Centre includes £5 million from Government. As a comparison of industries, the UK glass manufacturing industry employs 6,500 people across the UK and is valued at £1.3 billion for the UK economy. The food and drink industry has over 400,000 workers and has a turnover of £95.4 billion.

And Dawson points out that the Manufacturing Technology Centre in Coventry rarely covers food manufacturing in its projects, while Wilson calls attention to the fact that food manufacturing doesn’t sit naturally anywhere in the Government, as agriculture is covered by DEFRA and the Department for Business Innovation and Skills covers automotive and aerospace engineering, but food manufacturing sits somewhere between the two, despite food manufacturing being the largest manufacturing sector in the UK. It’s also one of the most stable, as the country will always need to feed itself.

The general opinion across the group is that children don’t grow up wanting to be food engineers or to go into the food industry at the production level. And more often than not, food engineering is a career they ‘fell’ into. 

As talk steered towards apprenticeships, the group agreed that there was a lack of apprenticeships in the UK, with demand far exceeding supply. Festo, who put a lot of emphasis on apprenticeships and training, take on two apprentices each year who are also placed within an OEM and food producer during their apprenticeships and vice versa to help them understand the pressures and challenges of other businesses. The company also sponsor their apprentices through their HNC.

“Within the food industry, there is a lack of food manufacturing engineers to train them,” says Wilson.

“And maintenance engineers are very specific, they lack the experience and knowledge to understand the entire production process or the pressures on production,” says Dawson. “There’s an absence of broad spectrum training where engineers can spend time on the shop floor but also in production, food technology and chemical engineering and across the process side of the business. Again, food companies need to look ahead five, ten years at what their engineers will be dealing with.”

The industry’s willingness to invest in apprenticeships is improving, according to the group. From apprenticeships being almost non-existent, companies are now starting up their apprenticeship programmes, but the challenge appears to be where to place them, what their role should be and what do companies give back to the apprentices. There is also great discrepancy between companies in the industry. Big multinational companies like Nestlé and Unilever have good apprenticeship and graduate schemes in place, while smaller companies take on apprentices when business is going well. It’s also one of the first schemes to be dropped if cuts need to be made. 

Print this page | E-mail this page


Article image Artificial intelligence in the food industry

Artificial intelligence (AI) has been heralded as the next best thing since sliced bread. But what might it really mean for the food industry and what are the implications? Stephanie Duvault-Alexandre explains. Full Story...

Article image Reduce, reuse, recover

Taking simple steps to reduce water consumption or access wastewater treatment technology can help change the way this valuable resource in managed, says Simon EmmsFull Story...

A recipe for continuous improvement success

Added value: the best way to deliver ROI

What role does refrigeration play in the supply chain?