Knowledge and Skill Driven Manufacturing of High Value Added Products - Human Aspects

Knowledge and Skill Driven Manufacturing of High Value Added Products - Human Aspects J. Heilala 1, T. Määttä 2, J. Viitaniemi 2, S. Kivikunnas 3, M. Sallinen 3 1 VTT Technical Research Centre of Finland, P.O. Box 1000, FI-02044 VTT, Espoo, FINLAND 2 VTT Technical Research Centre of Finland, P.O. Box 1300, FI-33101 Tampere, FINLAND 3 VTT Technical Research Centre of Finland, P.O. Box 1100, FI-90571 Oulu, FINLAND Abstract Production systems are getting more complex. Introduction of new technologies, new products and production paradigms have an effect also on system design and how human operators collaborate with these complex intelligent systems. At the same time global competition, customer demand for personalised products, and shorter life cycles of products make profitable automation and information management on the factory floor demanding. The challenges are most important in labour intensive small series production, where old-fashioned ways of production are no longer competitive in the West. There is a need for a new Smart Manufacturing Paradigm to fulfil needs and create new ways of working. The authors are searching for methods to create smart manufacturing systems and enabling technologies in order to meet the need, focusing on the role of the human operator. This paper is based on the authors experience of industrial collaboration, a literature study, and ongoing feasibility studies at VTT. Keywords: Smart manufacturing, human technology interaction, ICT for manufacturing, 1. Introduction In Finland and other European countries there is a need to evolve from resource based manufacturing to knowledge and skill driven manufacturing. In Western countries, companies are not able to compete with labour costs; instead they should focus on customised high quality, high value added products. Working life is at a turning point as the amount of information and knowledge demanded and produced at work continues to grow. A new paradigm is called for: Smart Manufacturing with efficient use of ICT, ambient intelligence, ubiquitous computing, novel interfaces for Human-Technology Interaction, human centred automation, and affordable and flexible robot automation technology that meets the requirements of SMEs. The word manufacturing derives from two Latin words: manus, meaning hand, and faceo, meaning to make. Literally, to manufacture means to make by hand. However, time and a number of inventions and innovations have significantly transformed this meaning. Manufacturing is also defined as the processes and entities required to create, develop, support and deliver products. According to an old dictum, Knowledge is power. The solution lies in working smarter before working harder. On the other hand, new information and communication technologies offer new possibilities. Traditional workplace learning is no longer enough. Faster and more flexible methods to

collect processes data and convey information and know-how to the right people at the right time and in the right place are needed in various working environments. A streamlined organization in which every part has its place and where everybody is doing the right job at the right time may take some effort but the results are worthwhile. Tools and processes are needed to ensure a seamless and continuous flow of material and information within the factory and beyond. This paper presents a vision of Smart Manufacturing with human factors, and identifies research needs. 2. Future of Manufacturing Many of the recent international manufacturing roadmaps such as Manufacture Challenges for 2020 [1], IMS Vision 2020 [2], FutMan [3], discrete parts and process manufacturing, and Strategic Research Agendas (SRA) of the European Technology Platform; Manufuture 2003, [4], 2006 [5] and some of the European Commission s High Level Groups background papers show the challenges to manufacturing in Europe. The vision of the technology platform on future manufacturing technologies, MANUFUTURE SRA 2006 [5], calls for the European manufacturing industry to reinvent itself, shifting from cost-based global competition towards a dynamic creator of knowledge-based added value. A similar development is proposed also in the European Commission s report KTE 2005 [6]. The Network of Excellence on Intelligent Manufacturing Systems, IMS-NOE, is also pinpointing similar development needs [7]. IPROMS, The Network of Excellence for Innovative Production Machines and Systems project, has developed a comprehensive research roadmap on production organization and management. The roadmap has been divided into six topics: 1. Fit Manufacturing, 2. Virtual Enterprise, 3. Holonic Enterprise, 4. Individualised Manufacturing (Mass Customisation), 5. Integration of human and technical resources, 6. Manufacturing Knowledge Management. These topics are reflected in two themes: Costeffective and rapid reconfiguration of the Manufacturing System, and Integration of human and technical resources. In developing a conceptual structure for manufacturing systems the roadmap considers these two themes as Dynamic Systems. These were further examined in terms of agents (humans, organisations), structure (relationships) and dynamics (behaviour). [8] CE-NET The Concurrent Enterprising Network of Excellence has developed a vision of integrated Product-Service-Organisation configurations. The CE-NET roadmap divides Collaborative Enterprise as a domain down into five complementary areas: Human Aspects, Business Models, Integrated Product-Services-Organisation Development, Information & Communication Technology, and Policy & Regulation. [9] A Smart Assembly initiative has been proposed by OOONEIDA (www.oooneida.info) in the USA. OOONEIDA is a non-profit organisation that aims to increase applied research, technology innovation and training efforts for more flexible, open integration and reconfiguration of embedded intelligence for industrial automation systems. Key elements of the proposed Smart Assembly are [10]: Empowered, Knowledgeable People: Multidisciplined, highly skilled workforce empowered to make the best overall decisions. Collaboration: People and automation collaboratively working in a safe, shared environment for all tasks. Reconfigurable: Modular, plug and play system components easily reconfigured and reprogrammed to accommodate new product, equipment, and software variations and to implement corrections. Model and Data Driven: Modelling and simulation tools enabling all designs and design changes to be virtually evaluated, optimised, and validated before being propagated to the physical plant. Capable of Learning: Self-integrating and adaptive assembly systems that prevent repeated mistakes and avoid new ones. According to the authors own experiences of industrial collaboration, there is a need for continuing research (Fig 1). This includes the creation of a smart manufacturing paradigm, focusing on human centred automation of physical and cognitive tasks and the use of modern ICT for manufacturing. The aim is to provide support to human operators to enhance learning, shaping, intensifying, and coping with complexity and

Human operator is the key resource Skills, Context Knowledge Human Operator KNOWMAN System Intelligence Learning and optimization to variants and generations (applic. knowledge, skills, rules) Knowledge Exchange (across individual operators, lessons learned from execution ) Human - System Interaction at Factories Product Info Process Info Raw materials Adaptive Task Device interface to smart machine Fig 1. Manufacturing is facing changes, new paradigm and new ways of working are needed Feedback, monitoring High value added, quality products adverse events in their work on the factory floor. The ultimate goal is to increase the productivity of the human operator. 3. ICT for Manufacturing One consequence of the deep penetration of ICT (Information and Communication Technology) into daily life is migration from the conventional factory floor to intelligent manufacturing environments built around the AmI (Ambient Intelligence) paradigm. That is, workplaces with emphasis on greater userfriendliness, more efficient service support, userempowerment, and support for human interaction. Industrial AmI solutions are still rare, even if research efforts are being carried out. EU level networking AMI@Work [11] and related research projects are developing, and Industrial AmI. Ami@Work has this vision of new working environments: "Next Generation Collaborative Mobile Virtualised Working Environments focuses on workers interacting with their environments and collaborating with each other, having access to all the (also virtualised) resources (including also assisting robotics) required to carry out their tasks and enhancing their capabilities. Among these resources, the key is the knowledge of the co-workers to complement in dynamic groups the needed competences and skills to carry out the task in an efficient way leading to an increase in productivity, and generating innovative and creative solutions." In a manufacturing environment where workforces are surrounded by a collection of reconfigurable production components (physical agents) that include mechatronics, control and intelligence (intelligent sensors and data processing units, autonomous, self-tuning and self-repair machines, intuitive multi-modal human machine interfaces etc.), the challenge is to develop production automation and control systems with autonomy and intelligence capabilities for cooperative/collaborative work, agile and fast adaptation to the environment changes, robustness against the occurrence of disturbances, and easier integration of manufacturing resources and legacy systems. Today's high production process performance requirements, combined with their increasing complexity, represent a great challenge for staff members at all levels (from the worker to the plant manager) for controlling the production process in such a way that customer orders will be fulfilled perfectly. 4. Human Centred Automation Manufacturing using robotics and future robot technologies has been presented by the EUROP technology platform in the EUROP SRA [12]. The main challenge of industrial robotics seems to be the creation of a robot capable of co-operation with humans. There are also visions that the robot would have skills to learn tasks. When robots and humans are working co-operatively, they should also share their skills and experiences. These kinds of properties go well beyond current learning systems. Co-operation is based on advanced sensor technologies. Current technology enables real-time sensing and processing. The range and resolution of the sensors have improved in the last few years. However, there are still several safety issues to be

resolved before fully operable co-operation is possible. Current Integrated Projects (IP) within the EU in the field of robotics and human co-operation are SMERobot [13] and PISA-IP [14]. These projects will give new information on how the co-operation can be carried out. The SMERobot project focuses more on the physical structure of the robot arm, and the PISA-IP project more on natural interaction and operation within the same workspace. One conclusion from the topics of these projects is that programming of the robot is expected to be more online programming than off-line in manufacturing. The reasons for this are customer-oriented production, the fact that every product can be unique, and human flexibility, which is more flexible than conventional off-line programming. There is a need for a new type of human robot interaction on the factory floor for teaching and programming the robots. 5. Human Factors in Smart Manufacturing Users main tasks in (semi-)automated, e.g. robot-based manufacturing systems, are 1) configuring and teaching the system; 2) monitoring and fine-tuning the system; 3) charging and unloading; 4) dealing with unplanned events; and 5) communication with other team personnel. The general experience is that the more advanced, complex and smarter the system is the more critical is the role of human operator [15], [16], [17]. Smart systems are vulnerable to failure and, therefore, they are even more dependent on human intervention than less intelligent ones. Studies show that task characteristics and psychological mechanisms are responsible for human error in complex sociotechnical systems. Also existing in complex systems are loosely designed features that can easily lead users to unpredictable situations suddenly requiring more thinking than doing. This may overstress users if they continuously have to balance between very opposite types of actions: quick reactive doing and time-consuming creative thinking cf. efficiencythoroughness trade-off [18]. Some general design criteria in developing smart manufacturing systems are 1) user participation and active involvement; 2) cooperation and teamwork between users and smart artefacts; 3) sense of control over the process and smart environment; 4) safety and accident prevention; and 5) usability of the human-system interface [19]. These systems should be sensitive to the skills of the user and his/her preferences, they should provide users the opportunity to shape their own working behaviour, and they should unite different task components of the work and encourage communication and cooperation between operators [19]. Smart tools and artefacts should support users normal behaviour, but more important is also managing the unpredictable situation, e.g. in the case of human error, the tools should be robust enough while being smart. They should help people to manage complex situations, to make good decisions, and to act in a reasonable way. A key element in flexible agile manufacturing is the ability to easily configure and program the systems concurrently, taking into account the relevant human factors and considering the change of systems and human activities in order to provide the sufficient affordance. Several approaches have been developed to teach robots to perform complex actions. Some promising new approaches are based on learning by demonstration and learning by instruction paradigms, in which users first demonstrate the to-be-learned task or procedure (see e.g. [20]). Important issues in this approach are 1) understanding the procedure and the human intention; 2) learning and representing the procedure; and 3) mapping and executing the procedure with the system [21]. Knowledge of human-robot interaction plays a key role in the development of these systems. 6. Vision of Smart Manufacturing VISION: "Human operator as the key resource collaborating with a smart manufacturing environment (see also section 3, AMI@Work) Imagine your work being completely different: Instant access to information, knowledge sharing, an intelligent assistant being able to find the relevant piece of information you need, whenever and wherever you need it. Imagine that your personal assistant is always at work with you, without disturbing you, and you can use this assistant whenever you want and forget him when you don t need him. This will allow workers at different levels, when relevant, to: - Improve availability of information, communication and knowledge sharing, - Speed up learning of human operators, - Enhance quality of products, zero defects, - Improve teaching of automated device, - Shorten production standstills, speed of delivery. Future manufacturing enterprises will be able to

capture individual expertise and experience for efficient reuse and draw on a rich, openly accessible shared base of scientific, business, and process knowledge to make informed, accurate decisions and to ensure the right people get the right information at the right time to do their jobs (Knowledge-based Enterprises). Improved understanding and shared knowledge of the scientific foundations for material and process properties and interactions will support optimised process design and total understanding of complex transformations and interactions at the micro and macro levels. Smart manufacturing systems should support team play between users and machine agents. Since users should have a sense of control over ongoing activities and system behaviours, the smart machines should be sensitive to the constraints and pressures under which the work is conducted. In the practicecentred approach, both human and machine agents are seen as mutually independent and their interaction is seen as dynamic and continuous [22]. Activity theory provides a promising general framework for analysing the practice-centred design of smart environments [23]. Activity theory typically analyses human beings in their natural environment and takes into account complex properties of the environment with which human beings are interacting. Activity theory also emphasises the fact that human activity is mediated by different types of external and internal tools and that the use of these tools has an effect on the way people act and develop their skills. Manufacturing companies making high quality value added products, especially SMEs, are able to serve customers more efficiently. Competition obviously is not based on cheap labour. Manufacturing enterprises have to progress towards being: - Customer-responsive enterprises - Totally connected - Reconfigurable - Efficient - Based on knowledge and technology innovation - Environmentally sustainable. Selected European and other international roadmaps are indicating the same, and Finnish industry needs to evolve from resource-based to knowledge-based skill-driven manufacturing. Labour cost is not a question when making high quality customised products with small lot-size and short delivery time. 6.1 Enabling technologies and features The smart manufacturing paradigm is based on using advanced ICT, information sharing, transparency, traceability, integration in both vertical and horizontal directions, synchronisation of manufacturing operations and asynchronous events using optimisation, simulation and related technologies. The aim is increased adaptability, agility and cost efficiency of production. This is based on concurrent approaches taking into account also human centred automation design, human technology interaction (HTI), in factories and workshops, use of ICT for decision making and task support for human operators to carry out complex tasks, as well as human collaboration with smart automated equipment. Use of human intelligence involves task sharing with human/automation and novel user interfaces (Augmented reality AR/Virtual reality VR, multimodal, haptics, etc.). The complexity of such systems poses a signifcant challenge to their usability. It is very difficult to develop distributed intelligence that can flexibly support the activities of its users. One prerequisite is that the system is able to adapt its behaviour according to the user, to the user s activities and to the physical and social context [24]. The smart manufacturing paradigm supports formative design approaches in re-configuration, adaptation to changes, agility and utilisation of machine learning capabilities, from resource based manufacturing to knowledge and skill based smart manufacturing. Recognised features in knowledge-driven manufacturing are: - Context-sensitive information/instructions to operators/workers; information availability, wide communication and feedback, use of design knowledge; - Knowledge and experience capture, knowledge/experience sharing, context awareness, on-the-job learning; - Science based management by workers; model and data driven decision making; - Data traceability and transparency; - Adaptability to product changes, to variations in productions, fast ramp-up, exceptional state management; - High quality, first product correct. Similarly, recognised features of human-centred automation are:

- User friendly, appropriate allocation of functions between users and automation; cognitive robotics, interaction; - Sense of control of complex processes; support for complex interplay between human and machine agents/actors; - New operating and maintaining concepts for smart environment; context-awareness, operative state management, task and decision support; - Multimodal interfaces, and accompanying AR, VR technologies, affordance table-based solution. 7. Conclusions Experts at VTT are currently defining the Smart Manufacturing Paradigm, with special focus on human technology interaction, ICT for manufacturing and human centred automation. Open research questions remain: How do we make the transition from resource-based manufacturing to knowledge and skill-driven manufacturing? How do we create a smart manufacturing environment, how do we operate it, and what are the enabling future technologies keeping the human operator as the key resource? The final goal is that in the future with the Smart Manufacturing Paradigm, using advanced ICT based applications and knowledge-intensive human technology interactions, industry will be able to keep the manufacturing of high value added products in Europe. Development has started. Acknowledgements The authors wish to acknowledge the financial support received from VTT. The development is part of the VTT Thema Complex System Design project KNOWMAN. VTT is a partner of the EC-FP6 I*PROMS Network of Excellence. References [1] Visionary Manufacturing Challenges for 2020 National Academy Press, 1998, ISBN 0-309-06182-2. [2] Summary Report on IMS Vision 2020 Forum - Refreshing the Vision of IMS", Beckmann Center, Irvine, California, USA, February 2000, (ims.kitech.re.kr/bbs/v2020.pdf). [3] The Future of Manufacturing in Europe 2015-2020 The Challenge for Sustainability, March 2003, EUR 20705 EN. [4] MANUFUTURE - A vision for 2020 Report of the High-Level Group November 2004, ISBN 92-894- 8322-9. [5] MANUFUTURE - Strategic Research Agenda. Manufuture Technology Platform, September 2006. www.manufuture.org [6] KTE. 2005. Manufacturing. Background paper for the European Commission s High Level Group on Key Technologies for Europe, 7/2005. [7] Taisch, M. & Thoben, K.-D. (eds.). 2005. Advanced Manufacturing. An ICT & Systems Perspective. IMS, Network of excellence on intelligent manufacturing systems. 328 p. [8] I*PROMS. 2006. Deliverable D7.5: Research roadmap covering all research areas, The Network of Excellence for Innovative Manufacturing, 43 p. [9] CE-NET. 2004. A Roadmap towards the Collaborative Enterprise CE Vision 2010. [10] http://www.oooneida.info/sector_analysis_smartasse mbly_project.html, accessed 16.4.2007 [11] www.amiatwork.eu, accessed 16.4.2007 [12] EUROP - Strategic Research Agenda. Technology Platform on Robotics. www.robotics-platform.eu.com [13] www.smerobots.org, accessed 16.4.2007 [14] www.pisa-ip.org, accessed 16.4.2007 [15] Bainbridge, L. 1983. Ironies of automation. Automatica 19, 775-779. [16] Rasmussen, J. 1986. Information processing and human-machine interaction: An approach to cognitive engineering. New York: North-Holland. 215 p. [17] Vicente, K. J., & Rasmussen, J. 1992. Ecological interface design: Theoretical foundations. IEEE Transactions on Systems, Man, and Cybernetics, 22, 4, p. 589-606. [18] Hollnagel, E. 2004. Barriers and Accident Prevention: Or How to Improve Safety by Understanding the Nature of Accidents Rather Than Finding Their Causes. England: Ashgate [19] Karwowski, W., Warnecke, H.J., Hueser, M., & Salvendy, G. (1997). Human factors in manufacturing. In G. Salvendy (Ed.) Handbook of Human factors and Ergonomics, pp. 1865-1925. New York: Wiley. [20] Billard, A. & Siegwart, R. 2004. Robot learning from demonstration. Robotics and Autonomous Systems 47, 65-67. [21] Dillman, R. 2004. Teaching and learning of robot tasks via observation of human performance. Robotics and Autonomous Systems 47, 109-116. [22] Sarter, N.B., Woods, D.D. & Billings, C.E. (1997). Automation surprises. In G. Salvendy (Ed.) Handbook of Human factors and Ergonomics, pp. 1927-1943. New York: Wiley. [23] Engeström, Y. 2001. Expansive learning at work: toward an activity theoretical econceptualization. Journal of Education and Work 14 (1). [24] Norros, L., Kaasinen, E., Plomp, J. & Rämä, P. 2003. Human-Technology Interaction Research and Design. VTT Roadmap. VTT Industrial Systems, Espoo. 118 p. (VTT Tiedotteita - Research Notes: 2220)