Gunnebo 2020: The Automation
An interview with Andy Puhl, Bennie Bus and Hans Salemink
The past decade, Lean manufacturing accounted for a gradual increase in efficiency. The ongoing quest to offer customers more flexibility, a more uniform and even higher quality, and a competitive price – it requires automation. So roll up your sleeves (for a lengthy insight into a daring project).

Gunnebo 2020 is an automation project. It’s aim is to increase the level of automation of the mass production of ATM-safes on the assembly line of Gunnebo Doetinchem. The project commenced in 2014 and consists of 3 stages. Phase 1 has been completed in 2016 and comprised the automation of the punching of steel metal sheets (a part of the ‘preliminary’ work), in essence by the purchase and utilisation of 3 Amada Punching machines. Phase 2, due to be completed in 2018, involves the automation of the final sections of the assembly line. Major elements of this phase are an overhead conveyor for safe transport, and automated paint-spray and cleaning booths. Finally, Phase 3 aims to automate the ‘bending’ process of steel metal sheets, and is due to start in 2018.


The Gunnebo 2020 project is a prolongation of a trend towards automation which originates in the commercial success of a particular kind of ATM safe developed in Doetinchem. From around 2006 onwards, a fast increase in demand for this type of safe had lead Gunnebo to abandon the traditional job production process (by which individual teams produce customised vaults and other safety products such as cell-doors). A ‘Single Piece Flow’ production style was introduced (think Henry Ford, or Charlie Chaplin – an assembly line consisting of stations, each at which roughly the same operation is performed on a ‘stream’ of ATM-safes). In 2010, a major customer of Gunnebo, NCR, initiated a ‘Supplier Production Improvement’ program, to further professionalise the assembly line according to the principles of ‘Lean’. Well known from its implementation by Toyota, the Lean system aims to eliminate any operations that do not efficiently add to the value of a product. By ‘slimming down’, an optimal flow of products through the assembly line is created. The implementation of Lean at Gunnebo was a success: productivity grew steadily, and employment was sustained, due to the increased turnover.

By 2014, it became clear that the substantial gains in productivity by implementing Lean were obtained. During that year, the first steps in automating the punching process were taken – a task which was comparatively easy, given that efficient machines for this process were readily available. Andy Puhl and his management team decided that automation was, indeed, the best way forward to guarantee the company’s continuity. The envisioned automation required substantial investments and approval of the Gunnebo Board of Directors, and therefore, a more formal and detailed policy plan was drafted: Gunnebo 2020.

In this article, three men will elaborate on the Gunnebo 2002 project: Andy Puhl, Bennie Bus (Production Manager) and Hans Salemink (Production Engineer).

Any Puhl: “Gunnebo 2020 – that is the corollary of the introduction of Lean in 2010, which was implemented with the help of external consultants. We acquired the necessary skills ourselves in the following years, and became increasingly convinced of the merits of Lean. We continued these projects, yet – the major steps advancing had been taken. Any improvements which were left to us, were, with all due respect – more or less incremental. So in 2014, we asked ourselves how we could achieve a significant step forward. We concluded that this was only possible by means of a substantial investment in automation. “

Bennie Bus: “From 2010 onwards, we implemented several Lean projects related to process improvement. These enabled us to pick the ‘Low hanging Fruits’ – relatively straightforward measures, yielding significant financial benefits. We definitely will continue with implementing Lean measures, to reach the ‘Higher hanging Fruits’ – yet, we became aware that our progress was loosing pace. That lead us to consider the amount of manual labour which is involved in our manufacturing. An amount which makes our products less competitive compared to low-wage economies. In our view, there was only one effective solution: to automate wherever possible. “

“This involves hefty investments. As a subsidiary of a larger corporation, we can only proceed after the endorsement by the Gunnebo Board, of each of the stages of the Gunnebo 2020 plan. This required us to come up with a well devised and considered proposal, allowing the Board to make a fair assessment.”

Phase I (2014 – 2016): Amada

Visitors to the ‘Preliminary’ section of the assembly line at Gunnebo become acutely aware of a relentless cadence of loud and clear, brisk strikes, generated by the Amada punching machines. The 3 machines are situated in front of an extended depository which allows a trolley system to automatically collect and store the processed steel metal sheets, awaiting further processing. The system reads the jobs which the operators from Process Control have cued, and directs the trolley to feed the required batches of large metal sheets into one of the three punching machines. It is a complex system, which is monitored and directed from within the Process Control room. The level of automation ensures that at any given time, the exact status and physical location of a job can be viewed on screen.

Bennie Bus: “All our ATM-safes are composed of steel metal sheets, which are relatively thin – merely 2 or 3 millimetres. They are assembled into an ‘inner box’, onto which concrete reinforcements are welded. The inner box is then enveloped by a separately assembled ‘outer box’, after which the intermediate space is filled with concrete. These steel metal sheets, the building blocks of safes, can be produced, be shaped, in a number of ways. We have chosen to automate this process with ‘punching’ machines of the Amada brand. These use terrific force and speed to punch out shapes and holes in large metal sheets, by means of ‘stamps’. A technique which, at the time of our implementation, was regarded as the more reliable, in terms of uptime, when compared to the use of lasers. The overall system we have implemented also ensures the automated storage of processed batches, which allows its continuous and unattended operation even during off hours and during weekends.”

“It is a vast improvement, compared to how we previously manufactured these sheets – which involved a lot of manual labour, and an extensive routine of moving sheets around, during the subsequent operations which were needed. We had to manually load and unload the former punching machines. Moreover, every punched sheet had to be extracted from the original source-sheet by removing a large amount of ‘micro-joints’ – again, by hand. All in all, it was labour intense, and physically demanding.”

“The automation has resulted in an immensely increased manageability. With our software, we can instantly monitor which batches have completed, and which are still in cue. We are no longer hampered by a large amount of paper forms – which had to be completed, printed and archived. All we currently need is an operator to digitally enter a production plan, which the system then processes. The amount of manual labour has been reduced enormously.“

Hans Salemink: “As a production engineer, I’m not involved in the decision which brand, or which type of machine is to be bought. Only after the actual purchase, my work begins – “We’ve bought this and that, and we need to have it just there. But it won’t fit, so you’ll have to rebuild a little.” Well, I take care of these things – I get the contractors to erect a new part of the building, make sure the machines are installed correctly, facilitate the waste flows. The whole nine yards. It has to all come together. Which isn’t always easy. For example, when we had ordered the magnetic conveyor belt to transport the steel residues from the Amada punching machines to a collection point, we were promised it to be modular. Which it turned out to be anything but – just one, lengthy stretch. The Amada machines were already in place, and by no means could we manage to get the belt underneath. It took us quite some time before we figured a solution.”

Bennie Bus: “When we were considering the optimal scheduling of the Gunnebo 2020 project, we concluded that the automation of the sheet processing could be accomplished the soonest. A matter of purchasing machinery, which is well-established in manufacturing – even though we had a complex implementation in mind. The required construction work in our factory was reasonably straightforward. We transferred our existing machinery to another location, some 20 kilometres from here, so that we could carry on manufacturing during the setup and testing of the new machinery. A major undertaking, naturally, but all in all, it proved to be relatively easy to continue operations during the transition stage. “

Hans Salemink: “Wim Lobeek from QHSE is the project manager; my role is operational. I compose the technical schemes, and the precise alignment of new sections with the current shop floor lay-out. But I’m not an inventor, probing new innovations during long walks, so to speak. My main source of information is the production floor itself – the people who’s day-to-day job is manufacturing, who have a wealth of accumulated experience. My goal is to make their jobs easier, more pleasant. I use their input to design the new sections of the assembly. Bennie Bus regularly schedules meetings to discuss these design proposals. With Shift Leaders, the production workers, the Engineering Service. In my experience, most people will only show an active involvement in projects if you can actually illustrate your intentions with drawings – with something tangible. When they can visualise, they get ideas – “Why not do it this way”. I also consult frequently with Ronald Buiting, head of Engineering Service. He’s really helpful when it comes to foreseeing a wide range of practical issues, and advises frequently on maintaining the proper sequence of operations. That’s an invaluable aid.”

Phase 2 (2017 – 2018): From Finishing Assembly Line to Expedition.

The operations which take place at the stations of the Finishing Assembly Line include the merging of the safe bodies and doors, the cleaning of concrete residues, and several other operations. Every safe travels from station to station on a trolley, guided by rails. At the Painting Line, each safe is again cleaned (degreased) and dried. The safe surface is perfected by filling with putty, sanding and finally polishing, before a spray paint is applied. After drying, the safe locks are attached at the doors (Lock Assembly line). Then, each safe is subjected to a quality inspection, and finally packed for shipping at Expedition.

Thus, during production, a ‘train’ of safes meanders through these sections. At some stations, the operations performed may seem demanding, to any visitor unaccustomed to a production environment. For example, degreasing of the safes is done in the compound of a small booth, where an operator first sprays on a chemical agent, before using a pressure jet to degrease the safes. There are no health risks (an adequate oxygen mask is used), but the works does seem monotonous. Likewise, the spray paint is applied manually. Although these tasks may seem unappealing, the operators at these stations regard them as ordinary, and, when asked, have no particular desire to replace them by automated processes. They take pride in their work. Nonetheless, Phase 2 of the Gunnebo 2020 serves a twofold purpose: to increase efficiency and product quality, and to improve working conditions.


Bennie Bus: “Currently, the safes move between stations on rails, and, in general, they are pushed forward manually. This system will be replace by an overhead conveyor – the safes will hang above the floor, as opposed to resting on a trolley. The locomotion is mechanically controlled, but not enforced by a rigid timing, so to speak. Safes will only continue along their route after, at each station, an operator has given clearance by pressing a button.”

“The degreasing process, as part of the Painting line, will be done by robots. Initial drying after degreasing will be automated, but the required additional drying of threaded holes and other inner parts will continue to be done manually. Likewise, the filling with putty, sanding and polishing is such an intricate operation, that automation is virtually impossible. Then again, the painting process will be fully robotised.”

A section of the design of the new assembly line lay-out. In the centre, the conveyor belt, which precedes the overhead conveyor (top of the illustration). The numbers correspond with the separate operator stations within the assembly line.

“A noteworthy upcoming change in our process will be that the safe bodies and doors will be joined at a far later stage of the assembly. Presently, they are merged at the Finishing Assembly line, before operations as cleaning, filling and spray painting. With the new lay-out, safe bodies and doors will remain separate until after they are spray-painted. Why? The doors of a safe are very heavy. It would be an overly complicated job to design a safe and efficient automated system for the opening and closing of doors during the many stages of the assembly. When you work on them separately, it is far easier to standardise operations.”

Hans Salemink: “To me, there’s no doubt we will benefit tremendously from this project. The first part of the Finishing Assembly line will have a conveyor belt, which converts into an overhead rail. That will run automatically. At each station, you can uncouple a safe before working on it. It’s not a continuous, or ‘pushing’ system – only after you’re ready with your task, the safe will continue along the line. An improvement both ergonomically, and of the ‘flow’ of products along the line. No more accumulations of safes, which leave entire sections inactive. We simply can’t afford these gridlocks. So yes, I think we will profit substantially – it’s more coherent, more sensible, and far more workable.”

Tread carefully, but don’t linger.

Automation does not end (nor start) with the purchase of machinery. The specialised manufacturing of safes at Gunnebo often requires customisation of the commercially available machines and technology. And during the implementation, the continuity of production, in volume and in quality, is an absolute priority. Therefore, the roadmap for phase 2 consists of a modular approach. For each section to be automated, the current and automated processes will run parallel, to allow for testing and fine-tuning before the final commissioning. The entire project phase is counselled by a consultancy firm specialised in process automation, and contemporary developments in technology.

Andy Puhl: “Inevitably, there will be minor setbacks during the implementation of this project. And with hindsight, we may eventually wonder if some of the choices we made could have been better. Nevertheless, automation, and the use of robots – it is tackled at countless factories all over the world. The technology exists. When the decision is made to go forward, thén you focus on getting the best robots, the proper tools, the excellent software.”

Bennie Bus: “This phase constitutes a large investment, financially, it involves extensive construction engineering, and a significant change to our process. Above all, it will have to be implemented whilst production continues. We can’t just shut down our production for 4 or 5 weeks. That is why we execute it in stages. With the first stage – the cleaning of residues of concrete from our safes – we foresee some complications. We have performed multiple tests at companies who operate such a cleaning facility. These turned out to be unsuccessful. Concrete, compared to grease, is far more difficult to rinse with a chemical agent, unless you use very acid or alkaline agents. Apart from their environmental footprint, they also may infiltrate into the safe through joints and seams, which could result in less bonding of the paint. Therefore we have planned a precaution. We will expand the Painting section to make room for the new cleaning facility, which will operate concurrently with our current facility. This allows us to configure the process until it works as needed.”

“We operate 3 paint spray booths. Two of these will be automated. We foresee no problems for this, as the use of robots for painting is widespread in manufacturing. And if problems do occur, we can fall back to the third booth, which will continue to be operated manually, and utilised for custom products. Regarding the overhead conveyor system, for the internal transport of safes – we will start with a partial installation, in order to test it. As for the overhead rail carrier itself, we expect this to operate smoothly, as there are – again – thousands of applications worldwide. However, the pendants which will carry the safe bodies and doors, may pose a challenge. We manufacture a wide range of safes, varying in weight and dimensions, but the pendants need to remain stable during movement at all times. For this, a custom solution has been devised, which we want to test extensively.”

Hans Salemink: “As you would expect with such a large-scale project, at times I do feel a little stress. Will it progress as planned? Is everything thought of? But it is an altogether wonderful task. It involves dealing with people from a wide range of disciplines. Contractors, suppliers, engineers. When I start work in the morning, I can’t foretell what I’ll be doing in the afternoon – there’s so much diversity. And at our company, there’s a high level of flexibility, of possibilities to achieve what has to be done. If there is a way to improve ergonomics, even if it is accompanied by a significant price-tag, we will implement it. This sense of possibilities – it becomes apparent if you look at the number of colleagues who eventually return, after an employment elsewhere. “It’s good to be back”. It’s one of the advantages of being part of a larger company. Small companies, they are limited – there’s rarely any cash for investments. So yes – I think we’re in a luxurious position here. It’s good to be aware of that, now and then. There’s a generous budget for professional training. If you want it – you name it.”

Bennie Bus: “We began this second phase similar to the first – with the consultation of suppliers of the systems that we had in mind. The Amada punching machinery, which we purchased during the first phase, was part of a complete solution its supplier could offer, based on the evaluation of our requirements. But with this phase, we soon realised that the interwoven nature of these subsystems required a bird’s eye view. You can’t just purchase an overhead conveyor system and ‘drop it’ into a production process which has been shaped over so many years. We therefore are consulted by an external firm with a broad expertise in process automation. They also bring an insight into the evolving modern technologies. That’s helpful, as these technologies are changing rapidly. Since we started this phase, some 1.5 years ago, new developments already account for slight alterations in our plans for purchasing – which will soon be finalised.”

“In our region, there is a saying on how to best move forward. It translates roughly to ‘Lightning Smart’. That’s in line with the current ‘Smart Industry’ trend – for example, we’re participating in a regional ‘Smart welding’ project. And it fits our mentality – to get things done. We want to progress rapidly, to keep up with the ever faster changing world. To establish a secure and strong position. But we can’t always enforce a high pace. There’s the municipality, and several regulators – whose wheels may turn a little slower. And if you want to be thorough, you will have to allow time for a project – time to adapt to changes, to setback and opportunities you didn’t anticipate at the start. But most of all, we need to absolutely ensure the continuity of manufacturing – in quantity and in quality. These are the vital pillars of our customer relations.”


One of the objectives of Gunnebo 2020 is to reduce the cost of manual labour, for each safe which is manufactured. Sustainment of employment thus depends on an increasing demand for ATM-safes. This was what happened after 2010, when the deployment of Lean meant a decrease in the relative amount of manual labour, but employment levels remained steady. It is however not a given fact that this mechanism will apply again. Gunnebo in Doetinchem employs many temporary workers, and can therefore respond flexibly to any fluctuations in workload or demand. Nonetheless, the automation will definitely bring changes for the permanent workers as well. With the automation, many tasks will change, and the overall demand for higher skilled workers will increase. The start of the gradual commissioning of the automation is no longer a distant notion, and soon, a partial reallocation of tasks and responsibilities will begin to take shape. Employees that anticipate on these changes, will, to a larger extend, be able to actively participate in this process.

Andy Puhl: “Our task, as management, is to inform our employees about the goals, and the progress of the Gunnebo 2020 project. Not once, but continuously. To allow people to adapt, and to become familiar with upcoming changes. By initiating a higher level of automation, we offer our employees a chance to be become skilled in the application of new technologies. Automation is not a goal in itself. It serves the overall aim of the continuity of employment. Which can be achieved by offering quality products to our customers, for a competitive price – and thus ensuring our profitability. Profitability is what legitimises our existence as a company. It requires flexibility, and the willingness to commit to the changes that are necessary – the willingness to embrace new technologies. To say: “Will we deploy robots? I want to operate them.”

Hans Salemink: “Our assembly line – it will become quite different. Most of our workers are aware of this, but some – although they do know a change is imminent, they’re not necessarily aware what the implications will be. When we were working on the first phase, with the Amada machines, we asked around: “Do you want an extra training? Welding?” And some replied indifferently. Those who did took on new tasks, were more aware that – well, so to speak – there wouldn’t be enough for everybody. Which, so far, has been avoided – with all the changes we have implemented, only very few have left. But the tasks will change. Definitely. Jobs will be different. Our cleaning section – where you work in a booth for a full day. Personally, I think it’s best to change that, and we will. But these workers are accustomed to it, and do not disapprove – even enjoy it. There are, surprisingly, a fair amount of workers who do not wish to change their tasks. “I’m okay with this, it’s what I do well”. So – it takes time, and patience, to bring everyone aboard when you want to do things differently.”

Andy Puhl: “We’ve been using welding robots for many years now. Already some thirty years ago, Martens Brandkasten was considered a front-runner in the industry – being among the first users. There are many operators that like to operate them. Yet – operators who are also eager to maximise their performance – they are scarce. With a determination to improve efficiency. And I think that people with this kind of proactivity also benefit themselves, because it will embed them firmly into our company, and so provide them with a high level of personal continuity.”

“With this project, we are currently implementing profound changes in the way we work. In essence, our employees are provided with the privileged opportunity to develop their expertise in contemporary technology. There has been a consistent, and I think adequate level of communication on the changes we will implement. This allows for all our employees to consider if these changes will effect them personally, and if there is a need to address certain preferences or questions. This is a collective project, and we, as management, are always willing to answer any questions that arise.”


    • The Factory.

      The Factory.

      Each factory can be considered a system – of machines, materials, energy, production workers, transport, safety – and much more. Which of …
    • From Eastern-Europe with love

      From Eastern-Europe with love

      Imagine, working abroad, in another country. The distance between that country and your 'home' being nearly a thousand miles. Peter Jancek is …
    • The Engineering of a Safe

      The Engineering of a Safe

      Jos Leusink has been designing safes for 25 years, causing headaches with those who long for easy cash. But there's no such …