autoregion international - Edition 1/2020

32 Self-organisation as a control principle in modular automotive production By Dr.-Ing. Olaf Sauer, Automation business unit/Deputy Institute Director Fraunhofer Institute of Optronics, System Technologies and Image Exploitation (IOSB), Karlsruhe One application of Industry 4.0 is “self-organising production”. Even the first documents relating to Industry 4.0 contain the vision that “by their ad-hoc networking capability as well as the inclusion of a digital product description, intelligent products are able to autonomously control their paths through the production pro- cess” [1] . Initial applications can already be seen in several factories, for example where automated guided vehicles (AGVs) transport workpieces to the next free assembly station. This idea of self-controlled production was already pertinent by the turn of the millen- nium, triggered by the then software agent technology [2, 3] . However, this production control technology has not become estab- lished for various reasons. Computer per- formance has been insufficient and there has been too little confidence in self-or- ganising units in factories. In body production, painting and final as- sembly, the automotive industry has long utilised the pearl chain control principle [4] , right up to the so-called just-in-sequence (JIS) delivery of components, which are as- signed to a particular “pearl”, i.e. a specific vehicle. However, this principle is reach- ing its limits due to the increasing variety of vehicle types, versions and derivatives. The space available for the provision of components on the assembly line, syn- chronisation of average cycle times, the fluctuating work content of specific cycles and the need for a high level of adaptability demand newcontrol principles. In any case, the linear path and strict adherence to the pearl chain between body shell production and assembly is usually interrupted in the paint shop, where color groups are formed. The sequence must then be restored in a body storage that allows selectable access after painting. The 2008/09 financial crisis in particular made it clear that the production technol- ogy installed and control principles used by automakers can result in problems in cov- ering fixed costs when sales fluctuate. This is primarily due to the legacy equipment currently in use, which is designed for spe- cific production series, engine versions or assemblies and is too inflexible to be used for other purposes in extreme cases, i.e. when capacity is underutilised. Automotive production will have to meet this challenge in the coming years. For future automotive production, au- tomated systems will therefore have to become more flexible, e.g. through equip- ment which can be used more universally and/or modular production lines that allow scalable, universal equipment to be quickly assembled and configured for new produc- tion and assembly tasks without great en- gineering effort. • Quality, e.g. ppm requirements • Delivery time, throughput time • Manufacturing cost, TCO for shipped vehicle • Other KPIs, such as OEE, OPEX, CAPEX, etc. • Optimization in terms of searching for a global optimum • Competitiveness based on manufacturing excellence and data based services; ‘the winner takes it all’-paradigm • Customer orientation • Sustainability • Resilience and agility • Uncertainty and reactiveness • Network, coopetion • Economies of scale • Good static ‘white box’ models for entire vehicle manufacturing down to single processes • Automation technology • Skilled workforce, their experience, knowledge about causalities • Clear manufacturing strategy • Computing power: Cyber Physical Systems, Edge, Cloud, Quantum computing, real-time communication and machine learning • Augmented people and assisting systems • Partners, market places, e.g. for technology data or capacities Drivers of the past Future drivers Enablers of the past Future enablers Fig. 1: Digitalisation of value creation: drivers and enabling technologies for future automotive manu- facturing Dr. Olaf Sauer Images/Graphics: © IOSB Fig. 2: Possible business models for future auto- motive manufacturing (perspective for 2030) [6] low Individualization high low value creation high „Integration- champion“ „Flex Champion“ „Plug & play- champion“ „Efficiency- champion“ S H A R E