Preface Control of Uncertainty in Mechanical Engineering

Transcription

1 Preface Control of Uncertainty in Mechanical Engineering During its lifetime, any mechanical engineering product will go through various phases in the product development, production and usage. Phases in development extend from the initial idea, product conception and design to the release of the finished product for serial production. These phases can be described as serial and/or parallel virtual process chains. Process chains become physical and real when the product is fabricated and used. In the physical process chain, phases extend from the production of the raw material, manufacture and use of the product right up to its re-use or disposal. Uncertainty occurs in all phases, which has a critical influence on process properties and consequently on product properties. In particular in the case of products with loadcarrying function, incorrect assessment due to uncertainty may have catastrophic consequences in terms of safety and profitability of the product, [1] and [2]. The objective of controlling uncertainty in mechanical engineering is to significantly enhancing safety, reliability and economic efficiency in development, production and use, and conserving natural resources. Essentially, uncertainty has to be taken for granted; however, its influence during the product lifetime from the material/semi-finished product through production and usage right up to its re-use and during the product lifecycle from market introduction, through growth and saturation right up to its decline can be controlled and hence minimized. Particularly in the area of light-weight construction, high demands are made with regard to efficiency such as low weight and low production cost with adequate load-bearing capacity. This means that in this field, and above all, the control of uncertainty is of particular importance. It is also the focus of the research conducted by the Collaborative Research Centre SFB 805 at Technische Universität Darmstadt, Germany, host of the 1st International Conference on Uncertainty in Mechanical Engineering ICUME 2011 and funded by the Deutsche Forschungsgemeinschaft DFG. As a first step to reach the goals mentioned above, known methods and technologies for the development, production and utilisation of load-carrying systems, up until their re-use need to be evaluated, with regard to their uncertainty potential. Based on this information, uncertainty can be described in process models and assessed, so that it can eventually be controlled with the help of new methods and technologies to be developed. 1. Advanced methods of robust design, mathematical optimisation methods for robust product design and mathematical models for the combination of active and passive load-carrying components within a system network, appropriate information models for representation and visualisation of uncertainties and new assessment methods will be developed in product development. 2. In production, process chains will be optimised with the help of the mathematical methods described in above. Metal-forming and metal-cutting methods will be rendered more flexible whilst maintaining a consistent level of production quality. Functional materials for active components will be integrated at an early stage. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, (# /04/18,20:01:41)

2 3. During the use of the load-bearing system, new usage monitoring methods will ensure the permanent acquisition of actual loads, whilst advanced mechatronic and adaptronic or adaptive technologies will stabilise and attenuate the load-carrying structure. Finally, structure/property relationships derived from the usage process may provide information on the quality and suitability of the product under actual conditions of use with feedback into development and production. There are numerous definitions and descriptions of uncertainty, depending on the area of sciences, e.g. data uncertainty, measurement uncertainty, uncertainty of information and many more. However, most definitions do not fully cover combinations and dependencies of uncertainty throughout processes of a product s full life time. Especially when providing high sensitive safety aspects, it is important to know how uncertainty propagates from one process to another, like from processes of development to production, from production to use and, if necessary to reuse. Well known approaches that classify uncertainty are worth to be mentioned: Knetsch [3] for example distinguishes between aleatoric and epistemic uncertainty. Aleatoric uncertainty includes random variation of parameters which can not be further reduced in general [4]. Aleatoric uncertainty mainly appears in real product life cycles. Epistemic uncertainty is caused by insufficient information about the product or the process. Due to lack of knowledge, reality is depicted incompletely, incorrectly and insufficiently detailed. Uncertainty could be classified into four types, depending on the ability of making a statement about probabilities and the ability to determine a characteristic consequence as well as a meaning of uncertainty [5] and [6]. If probability and consequences are known, the uncertainty is stochastic and therefore well controllable. However, if probability is known but the consequences are not, there is a certain element of ambiguity. If the consequences are known but the probability is not, it is a matter of simple uncertainty. If neither probability nor consequences are known, ignorance prevails. International standards like DIN-EN-ISO [7] for machine safety or 2006/42/EG [8] for machine standards define the term uncertainty as - a safety function of a machine that, in case of failure, may increase different risks, - reliability of a machine to fulfil a certain task for a certain time without failure, - risk in combination of probability and the extend of failure as well as - safety components, faults, failure etc. The term uncertainty is mostly used in an indirect manner in the field of measuring uncertainty or legal uncertainty. However, a practical, consistent, systematic and comprehensive description and evaluation of uncertainty as well as, eventually, a way to control uncertainty throughout the product s lifetime are still missing. Therefore, the SFB 805 formulated the following working hypothesis describing uncertainty: Uncertainty occurs when process properties of a system can not, or only partially be

3 determined [1]. According to this hypothesis, the SFB 805 developed a model of uncertainty to provide a reasonable categorization of uncertainty properties into: unknown uncertainty, estimated uncertainty and stochastic uncertainty. No knowledge or unknown uncertainty occurs when effects and resulting deviation of a regarded property of uncertain processes are unknown. Based on this state of knowledge, no decisions can be made to control uncertainty. With little knowledge or estimated uncertainty, the probability distribution of the resulting deviation is only known partially. Stochastic uncertainty occurs when the effects and the resulting deviations of a considered uncertain property are sufficiently (ideally completely) described by a probability distribution. Stochastic uncertainty is present after extensive analyses of properties in terms of quantifiable experiments and measurements. While differentiating between the three categories, no sharp boundary can be drawn. The transitions between the categories are fluent. As a general rule, unknown uncertainty becomes stochastic uncertainty if the amount of available and secure information increases. The Organizing committee of the first International Conference on Uncertainty in Mechanical Engineering ICUME is pleased to present several works from an international community and from the SFB 805 giving an academic and industrial perspective to describe, evaluate and to control uncertainty in: 1 Development 2 Production and 3 Usage. The editors hope to meet the interest of a broad readership with the selection of the following contributions and like to motivate for further investigations. Holger Hanselka, Peter Groche and Roland Platz

4 About ICUME The aim of ICUME is to discuss methods and technologies to describe, evaluate and control uncertainty in mechanical engineering applications. International scholars and specialists come together to provide a broad forum to discuss the description, evaluation, avoidance, elimination of and adaptation to uncertainty in planning, development, production and usage of mechanical structures, systems and machines throughout their complete lifetime. Engineers, mathematicians and other areas of expertise working in uncertainty evaluation exchange latest research results and application of uncertainty control. The proceedings show some new approaches and examinations for controlling uncertainty in mechanical engineering. By controlling uncertainty, safety margins between mechanical loading and strength will be lowered, oversizing will be reduced, resources will be preserved, range of application will be widened and economic advantages will be achieved. The Organizing Committee likes to thank the Deutsche Forschungsgemeinschaft DFG for funding the Collaborative Research Centre SFB 805 at Technische Universität, and for helping to realize ICUME.

5 Local Organizing Committee E. Abele Production Engineering and Cutting Machine Tools, Technische Universität R. Anderl Computer Integrated Design, Technische Universität H. Birkhofer Product Development and Machine Elements pmd, Technische Universität A. Bohn Product Development and Machine Elements pmd, Technische Universität P. Groche Production Engineering and Forming Machines, Technische Universität, Conference Co-Chair H. Hanselka System Reliability and Machine Acoustics, Technische Universität and Fraunhofer Institute of Structural Durability and System Reliability LBF, Germany, Conference Chair and Head of SFB 805 H. Kloberdanz Product Development and Machine Elements pmd, Technische Universität U. Lorenz Mathematical Optimization, Technische Universität P. Pelz Fluid Systems Technology, Technische Universität R. Platz Fraunhofer Institute of Structural Durability and System Reliability LBF, Germany S. Ulbrich Mathematical Optimization, Technische Universität International Scientific Committee J. Allwood Department of Engineering, Low Carbon Materials Processing, University of Cambridge, UK G. Diana Dipatermento di Meccanica, Politecnico di Milano, Italy P. Göransson Department of Aeronautics and Vehicle Engineering Royal Institute of Technology, Sweden E. Macha Department of Mechanics and Machine Design, Technical University of Opole, Poland A. Plummer Department of Mechanical Engineering, Centre for Power Transmission and Motion Control University of Bath, UK

6 References [1] Hanselka, H., Platz, R.: Ansätze und Maßnahmen zur Beherrschung von Unsicherheit in lasttragenden Systemen des Maschinenbaus (Controlling Uncertainties in Load Carrying Systems), VDI-Zeitschrift Konstruktion, Ausgabe November/Dezember 11/ , (2010), S [2] Engelhardt, R., Enss, G., Koenen, J., Sichau, A., Platz, R., Kloberdanz, H., Birkhofer, H., Hanselka, H.: A Model to Categorise Uncertainty in Load-Carrying Systems, Conference MMEP Modelling and Management of Engineering Processes, in Cambridge/UK, (2010), pp [3] Knetsch, T.: Unsicherheiten in Ingenieurberechnungen (Uncertainty in Engineering Calculation), Shaker-Verlag, Aachen, [4] Chalupnik, M., J., Wynn, D., C., Clarkson, P., J.: Approaches to Mitigate the impact of uncertainty in development processes, In: Proceedings of the 17th International Conference on Engineering Design, Stanford, 2009, pp [5] Andrews, C.J., Hassenzahl, D.M., Johnson, B.B.: Accommodating Uncertainty in Comparative Risk, Risk Analysis, Vol. 24, No. 5, [6] Stirling, A.: Risk, uncertainty and precoution: Some instrumental implications from the social sciences, Negotiation Change, [7] DIN EN ISO , Safety of machinery - Basic concepts, general principles for design - Part 1: Basic terminology, methodology - Amendment 1 (ISO :2003/Amd 1:2009) [8] 2006/42/EG Directive of the European Parliament and Council from for Machines, change of Directive 95/16/EG (revised version). In: Official Register of the European Union,