Industrie 4.0 and how it will transform the automotive industry
Prof. Dr.-Ing. Thomas Bauernhansl
Industrie 4.0 is the trending "buzz terminology" in the manufacturing sector - with BMW and Daimler among many prominent proponents - and encompasses the digitalisation of industry value chains. just-auto's Calum MacRae caught up with Prof. Dr.-Ing. Thomas Bauernhansl of the Fraunhofer IPA, who is one of the world's foremost experts on Industrie 4.0, to find out what the implications are likely to be for the automotive sector.
just-auto: Firstly, Professor Bauernhansl can you tell us a little bit about the Fraunhofer IPA and what your day job looks like?
Prof. Dr.-Ing. Thomas Bauernhansl: With nearly 1,000 employees, Fraunhofer IPA is one of the largest institutes in the Fraunhofer-Gesellschaft. It has an annual budget of over EUR60 million, with more than one third of it coming from industrial projects. Our 14 departments are supplemented by six business units: Automotive, Machinery and Equipment Industry, Electronics and Microsystems, Energy, Medical Engineering and Biotechnology and Process Industry. This structure enables us to help our practice partners improve their market position as well as support their market entry into new application fields. Our strategic cornerstones are sustainable projects with high industry participation. Mass sustainability aims at minimising the consumption of resources while maximising the standard of living. In flagship projects, such as the Ultra-efficiency Factory, Fast Storage BW, the Center for Lightweight Production Technology and the Center for Smart Materials, we are putting this concept into practice together with our partners from industry, university research and government. Mass personalisation unites the advantages of economies of scale and scope. In ARENA2036, the research campus for smart manufacturing in automotive industries and in Campus Personalized Production, we are working on ways to manufacture personalised products in batch sizes of one at the same price as mass-produced products.
j-a: All this is enabled to some degree by Industrie 4.0, referred to by some as the fourth industrial revolution - how do you see it?
TB: It certainly is a paradigm shift that affects all of us. I think that digitisation of society and especially production driven by the Internet of Things and Services truly is a revolution comparable to the invention of steam power. But like any revolution it will take time to be fully realised; in this case a timeframe of the next 10-15 years.
j-a: We see in the automotive industry - indeed in any industry - the ethos of continuous improvement. Do you think having such a distinct staging post as Industrie 4.0 runs counter to this ethos?
TB: On the one hand, in the framework of Industrie 4.0 there will be continuous improvement. But there will also be disruptive changes. The changes come unexpected, occur quite often and will be comprehensive. The digital transformation of value creation, especially in the automotive industry, has two aspects. It is used as a tool to improve existing processes, thus addressing the area of continuous improvement. Then again there also are disruptions. New possibilities arise, such as organising mobility (using external resources to provide mobility services). In the automotive industry some technical developments are already under way. On the one hand mobility as a service in the sense of the digital transformation, everything becomes a service (XaaS). On the other hand we see the comprehensive electrification of vehicles, which leads to a further technical revolution in the technical construction of vehicles and also in the value-added processes. The third aspect is superimposing on the first two: the autonomisation of vehicles with autonomous driving. This leads to a completely new - and especially emotional - perception of vehicles.
With continuous improvement, you can optimize what you already know, but the future looks quite different under Industrie 4.0.
j-a: Complexity management is seen as one of the biggest challenges for the automotive industry. Can you explain how Industrie 4.0 will help?
TB: With the three megatrends discussed already, which are associated with technological changes, but also with changes in business models, IT architectures, etc., a very high level of complexity arises.
The whole thing has to work globally, but should be adapted to local and personal needs. If we try to deal with this complexity with the existing value-added systems or bring them into the market as a critical success factor (to meet the market requirements), then we die of complexity. We will lose to complexity and profitability will be reduced to zero or worse.
Therefore, it is important to develop completely new means of dealing with complexity, and Industrie 4.0 and digitisation helps. It helps because a system that is complex always reacts to an increase in autonomy and decentralisation. If I want to be flexible, I need to decentralise and increase the autonomy of the system. And these are the basic principles of Industrie 4.0. We network everything and thus bring decentralised responsibility into the work system and thus increase the autonomy of these work systems.
At the same time, the value added changes away from the classical pipeline orientation to a platform orientation. We have very non-hierarchical cooperation structures, which are also decentralised. The technological enabler is comprehensive networking in real-time, the internet of things and services and people.
j-a: How do you recommend that companies implement 4.0? Holistically or on a project-by-project basis?
TB: There are basically two ways to implement Industrie 4.0. First, by outsourcing a company unit, like Trumpf in Ditzingen did with the Axoom platform in Karlsruhe. Second, within the core business.
But first we must distinguish between bottom-up and top-down. It is very important that you develop a clear vision from the top as a guide and then turn it into a rough roadmap of how you want to get there. To do this, one has to deal intensively with the topic. Business leadership must fully understand and have been convinced that it is the correct direction to go in. Then, in a second, bottom-up step, you have to ensure that the workforce can contribute to this process through user cases, qualification measures, etc. The third question is the implementation. When it comes to disruptive business models, it makes sense for many companies to not house this business in their existing structure, but to distinguish the two worlds. One is the new disruptive theme, the other is the evolutionary digital enhancement of the existing business model. The disruptor is taken outside of the company's structure either as a start-up or through acquisition.
There is no patented recipe for change management in a company to carry out a digital transformation. It can make great sense to move an organisation, which drives Industrie 4.0 in the enterprise, outside of the parent organisations. They can be protected against the classic controlling instruments, which can over-analyse and stymie. The new organisations (start-ups) also need some protection from the people, who may not be able to change, but whom may hold leadership positions and thus exercise power. Protection also has to be placed in front of employees who are afraid of losing their jobs and therefore do not want to push the development forward. In Germany, there are massive fears that the digital transformation is changing roles and that jobs will be lost.
j-a: The security vulnerabilities of the Internet of Things was believed to be a factor in the successful cyber attacks centred on the US (the Distributed Denial of Service attack on 21 October), how can companies manage this risk?
TB: In Europe, especially in Germany, we have very high security standards. This is why we recommend SME's to instal and use German digital production platforms. At my institute, the Fraunhofer IPA, the Virtual Fort Knox was built-up and is now commercially online. But there is no such thing as absolute safety. There never was and probably will never be.
There are already enough technologies to protect companies. But there will never be 100% protection. The main safety gaps that we see are in the people who are not sufficiently sensitised to the vulnerabilities. In this context, it is useful to outsource certain capabilities - like cloud technologies - because they can be better handled by service providers. We need to rethink and develop a security culture within companies and to sensitise our employees and to ask ourselves what we can do by ourselves and where we need cooperation with communities that are subject to uniform rules to ensure the security of the cooperation space.
j-a: Do you see your work as being in competition with the Industrial Internet Consortium (the Massachusetts-based organisation founded by AT&T, Cisco, General Electric, Intel, and IBM)?
TB: On the contrary. There is a strong need to cooperate, even for competitors. The digital transformation not only changes competition between companies, but also between economies. Economies that export a lot, such as Germany, need to import knowledge to keep a balance. For us, this means that a digital transformation of companies, which in any case are largely global, cannot be realised through a national initiative, but only through international cooperation and competition, coopetition if you like. This is already difficult on the corporate level and certainly more on the economic level, but we have to learn to work like this, to act like that, otherwise we cannot lever the potential opportunities.
j-a: Can you explain how a company could perform a cost-benefit analysis on Industrie 4.0 implementation at a factory versus continuing with 3.0 practices?
TB: Especially indirect areas will be fully automated within Industrie 4.0 (e.g. order and distribution, quality management, marketing, sales etc.) so it is obvious that there is a huge saving potential. But we do not recommend that companies calculate and compare everything down to the last detail, because they'll usually lack the means to assess properly. It is much more important to determine at a qualitative level what are the developments that will come anyway and cannot be avoided. Then the question is not whether one does this, but when and in which form.
From there it is a good idea to carry out the projects which have the highest relevance and involve a small implementation risk, but that involve as many employees as possible. That should be done in the form of a first proof of concept, then use cases, then business cases. Then one will be able to understand the mechanisms and knowledge to assess properly will have been built up. On this basis, you can decide how to plan and start to layout a detailed technology roadmap.
j-a: How do you see 4.0 changing automotive production? Will it work equally well through the tiers and at the OEMs?
TB: ARENA2036 - Active Research Environment for the Next Generation of Automobiles and 2036 because it will be the 150th birthday of the automobile - is our research factory for the production of the future, and also our notion for automotive production in general. It set up a revolutionary hybrid concept for an adaptive production of future smart automobiles using innovative robots as the base. Whereas today the manufacturing of automobiles is timed and synchronised on the assembly line, tomorrow there will be decoupled, fully flexible and highly integrated production systems. In such a hall, assembly stations are not interlinked and can – in addition to just assembling – carry out other tasks such as coating. Therefore, we use the term process module for defined manufacturing and assembly operations. Since many of these process modules work side by side all necessary technologies for vehicle production can be provided. The carbody is put on its wheels and equipped with the respective control and communication technology. Without needing any handling technology or main computer, such a rolling carbody moves to the required stations and gives the impulse for the respective build further along. Thus we gain a decentralised, very robust system which can quickly adapt to alterations.
Thus, the vehicle drives autonomously as a cyber-physical system (CPS) through the production modules in the assembly hall. The modules work in different cycles. This way, individualised manufacturing steps with scalable tasks can be integrated into the assembly. The vision of the fenceless factory will be a reality soon, because the current man-machine-collaboration is being pushed intensely.
We have our research campus ARENA2036 to answer this question and see a revolutionary production system in the future. As an industry we've already maximised Fordism and Taylorism, reached the end of measurement and takt-time and we replace it now. We need a highly flexible value-added system with process modules, which are connected and supplied by flexible logistics systems. These flexible process modules can be operated by the OEM as well as the supplier on a common value-added surface. We use the results and connect them with the possibilities of the internet of things and services and thus come to a very flexible and changeable production system. This manufactures no longer a single car model but several models and corresponding derivatives, depending on customer requirements, in different numbers and equipment variants. And the whole thing is highly economic. The first results we have achieved here show that we can achieve a 20-25% area reduction, reduce personnel costs by 25%, and reduce planning effort by more than 40%.
j-a: For many companies PLM and ERP implementation has not brought the benefits advertised, how is 4.0 going to be different?
TB: The nature of the systems is quite different. PLM and ERP are highly integrated systems without a service-oriented architecture and take root and movement data for themselves and then make it accessible only to the internal functionalities of the system. In the future, we will separate these levels. We separate the application from the data and we pack everything into a service-oriented architecture. The result is that I can very flexibly use different functions of a system from different manufacturers. I go into an app store when I can decide what support I would like to have in which process. This can be a support today, i.e. coming from supplier A and supplier B tomorrow. And that gives me a high degree of flexibility and ensures that the competition will continue to exist even after the implementation of a system and with this effect I will achieve a higher benefit at a lower cost.
j-a: Industrie 4.0 is a component of the German government's 2014 High-Tech Strategy: Innovations for Germany. Do you see it giving Germany a continued competitive advantage and what lessons can be learnt by other countries around their implementation of industrial policy?
TB: Industrie 4.0 is a big hit in Germany because we have managed to give a name to an industrial transformation that is highly relevant to industrial competitiveness which is very clear and memorable. This leads to a high level of political interest, as well as from businesses, and gives this topic a helpful tailwind. At the same time, it has become a brand for industrial change and the industrial strength of Germany, worldwide. And this helps us in the marketing of the corresponding technologies.
I can only recommend that other countries orient themselves similarly. You do not have to call everything Industrie 4.0, but it makes sense to give the baby a name!
j-a: You talk about 10 guidelines for the value-added system of the future. Within them you talk about 4.0 no longer having separated production and logistics functions. Can you elaborate?
TB: Today, the production and assembly system is planned independently of the logistics system. As a result, only an area of 10-15% is used for the actual value-added activities (in the assembly process, for example). A great deal of space around the logistics is needed for supplying the system with the right material at the right time. Because of the necessary horizontal as well as vertical flexibility (i.e. volatile markets bring constantly changing requirements, which is reflected in increased variants) I cannot realise the required flexibility and changeability in the strongly planned and timed production and assembly system that we have presently, combined with a logistics system that controls the planning and consumption.
If today's logistics systems tried to match production systems the complexity would explode and it becomes very uneconomical. That is why it makes sense to separate from this production system and to integrate and plan production and assembly in the future. Then, as explained above, there are process modules which have both a value-adding part and a logistic part. This means that material is also stored, prepared and installed directly in the process module. If so, the material does not travel to the product any longer, but the product comes - in parts at least - to the material which is stored in these process modules, prepared, partly also manufactured and assembled. This results in a tremendous increase in flexibility and thus an enormous cost reduction.
j-a: Your guideline 7 - shift process complexity to where it can be handled most efficiently - seems difficult to understand, but is it just the customer-pull of lean thinking optimised by our new digital environment?
TB: This is about the fact that I am more and more able to pass on complexity-driven content to other protagonists in the value-added system. More and more I can involve the customer and let them do things. The customer is already today in a position to configure his own product and does not need any sales support. He is able to order it himself. More and more, he will also be able to influence the design through digital platforms, he can make the service on the product in parts by himself, if he is upgraded via the platform. There, the customer can get the knowledge he needs to run his product optimally.
Through the internet of things and services, the so-called "access economy" comes about. I have access to my customers, know very much about him (passive access) because he is surrounded with sensors that watch him. I know where he is, when he uses my product and how he uses it. However, I can also use this access to the customer to integrate it into my value creation and to make things happen - I pass complexity-driven content to the customer (as I said it has been developed, ordered, planned for production).
At the same time, however, I also have access to resources, my own and external resources, also in the sense of knowledge, to communities that develop things, for example, open source. Now I have the possibility to distribute the contents in my whole value creation system so that I end up at the lowest possible complexity costs.
I concentrate on what I can do better than others and orchestrate it via the Internet of Things and Service so that a completely new business model is created which is highly productive and the whole thing is at a very low cost.
j-a: How can Industrie 4.0 handle the in-house complexity that OEMs have to deal with because of different market requirements - e.g. chassis settings; emission calibrations; powertrains; trim levels etc?
TB: Here one needs a product modularisation. The vehicle architectures are changing very much, think of the BMW i3 with its Life and Drive module, or a new module/platform structure of vehicles with the goal that the corresponding modules can be produced in corresponding process module landscapes. There is not only a modular product module, but also a process module which can be flexibly networked with each other, and then filled with corresponding know-how and added value both in the development of the product and the process modules. This means I decentralise and bind the customer by giving him the opportunity to bring both the module combination and personalised elements that he himself creates. This makes it possible to completely redefine the principle of complexity distribution in the automotive industry.
j-a: Will changeable digital production require the participation of all members of a processes value chain? For example, in vehicle manufacture, if there are several suppliers still at an Industry 3.0 level is that compatible with an Industrie 4.0 OEM?
TB: Gradually, all participants in the value-added system, which are now in the supply pyramid, will be covered by the transformation. First of all, the ones that are very close to the OEM, or the ones that draw a great advantage from these principles. The platform-based economy brings together supply and demand almost non-hierarchically because it is not a classical pyramid anymore but a platform and therefore the access from the OEE to the 3rd tier supplier is given in real time but also vice versa. Such a platform also makes performance very quickly transparent and companies that do not generate any additional benefits through this networking become the commodity suppliers. And here the mechanisms used in the second and third industrial revolutions are coming to bear, namely the economy of scale effects, because companies have to produce in huge quantities to be profitable when they are the commodity supplier. We will see corresponding consolidation effects among those who cannot generate additional digital returns in the sense of customer benefits.
j-a: There's been a pilot 4.0 project at Bosch that's delivered a 30-50% increase in the performance of value creating tasks - can you talk more about this project and what has been involved?
TB: We were not involved in this Bosch project, although we do work with Bosch and have carried out similar projects with other companies. We ourselves did similar projects with other companies. As the results of the project you mention were published I can say that it managed to make the inventory transparent along the entire value-added chain. Real-time inventory information has been brought together with real-time demand information. As a result, they were able to reduce the buffer stock, reduce inventory by 30%, increase the overall productivity by 10% and, at the same time, reduce the necessary milk runs by 10%. The 30-50% are the expectations of different experts from different companies. They believe that through the digital transformation, a total performance increase of 30-50% is possible at the end. And the projects that we have carried out so far, which have had a value-added system changing aspect, show that this is to be achieved, in some cases even exceeded (60-65%).
j-a: Are today's automotive players all prepared for Industrie 4.0 and the seeming switch from all that's been invested in Six Sigma and lean production?
TB: Many automotive manufacturers are prepared because they have long dealt with the topic of digital transformation, e.g. with the digital factory, manufacturing execution systems (MES), networking of production plants, etc. In the area of process management and control, too, a lot has already been achieved. Now comes a very big step because we go into a whole new form of cooperation, because we need a whole new IT architecture and at the same time we need to change the business models. So it is very important to have this good foundation, that is, process standards and process capability. But there is also a lot more to do for automotive industries and they should get started as soon as possible.