Qua2Pro Quality feature based quantification of production risks

由于产品开发周期的缩短、变种范围的扩大和产品复杂性的增加，制造企业面临着越来越多的技术风险，这些风险强烈地影响着产品的质量、可靠性，从而影响产品的成败和竞争力。对风险概率和风险的潜在货币后果进行量化和有效的分析和评估是确保产品和过程安全的关键挑战。到目前为止，风险管理的量化方法主要集中在公司的财务部门和领域，尚未建立操作风险即价值链风险的扩张。现有的确定和评估技术风险管理的方法，如FMEA，没有将定性评估联系起来。除了对潜在故障的量化，当流程中只发生小的变化时，险些失手的影响被低估了，这代表着高风险。缺少识别、分析和系统处理生产中这些险情的适当途径和方法。本研究项目的总体目标是开发方法，以量化评估发生概率和相关的货币后果的形式，得出描述生产中的故障风险的综合关键绩效指标。起始点是以质量特征的形式对产品和过程的要求进行分类。因此，开发了一种程序来识别和表征这些质量特征。作为次要目标，将开发确定风险概率的模型和确定风险货币损失的部分成本函数。对于建模，将研究统计评估方法的使用，如过程能力指数和描述罕见事件的方法(极值理论)。其中，通过基于模拟的验证，确保了科学结果的可转移性。技术风险的量化评估是质量管理中最大的科学挑战之一。对评估技术风险的统计方法的可转移性进行调查是研究项目内的主要挑战。通过故障事件的量化可预测性和相关的货币后果，为产品和工艺创新的成功持续提供了更好的基础。可以更有针对性地实施风险处理的预防措施，并可以更准确地评估其效果。由于缺乏关于发生概率和成本的准确陈述，研究项目的结果使得以前难以证明技术风险管理的好处成为可能。

Due to shorter product development cycles, higher range of variants and increasing product complexity manufacturing companies are facing increasing technical risks that strongly affect the quality, reliability, and thus the success of a product as well as the competitiveness. A quantified and valid analysis and assessment of risk probability and potential monetary consequences of risks represents a key challenge to ensure the safety of products and processes. So far, quantifying approaches of risk management mainly concentrate on the financial sector and areas of a company, so that the expansion on operational risks, i. e. risks along the value chain is not established yet. Existing approaches for identification and assessment of technical risk management, such as FMEA, do not link the qualitative assessment. Besides the quantification of potential failures, the effect of Near-Misses, that represent high risks when only small changes in a process occur, are underestimated. Suitable approaches and methods for identification, analysis and systematic handling of these Near-Misses in production are missing.The overall objective of this research project is the development of methods to derive aggregated key performance indicators describing a failure risk in production, in the form of quantified assessment of the probability of occurrence and the associated monetary consequences. Starting point is the classification of requirements on product and process in form of quality features. Therefore, a procedure is developed to identify and characterize these quality features. As a secondary objective, models will be developed to determine the probability of risk and partial cost functions for determination of the monetary losses of a risk. For modeling, the use of statistical evaluation methods will be investigated, such as process capability indices and approaches for describing rare events (extreme value theory). Amongst others, through a simulation-based validation the transferability of the scientific result is assured. The validation of the developed methods is performed by means of a turbine blade.The quantified assessment of technical risks is one of the biggest scientific challenges of quality management. The investigation of the transferability of statistical approaches to the assessment of technical risks represents the main challenge within the research project. Through quantified predictability of a failure event and the connected monetary consequences, a better-grounded basis for the successful continuance of product and process innovations is provided. Preventive measures of risk treatment can be implemented more focused and their effect can be evaluated more precisely. The previously hard demonstration of the benefit of technical risk management, due to the lack of precise statements about the probability of occurrence and costs, is made possible by the results of the research project.