Coping with Aging in Manual Assembly Systems: Age-differentiated Analyses and Mathematical Modeling for Predicting the Time Structure of Sensorimotor-skill Acquisition for Assembly in Series Production with Numerous Product Variants

该研究项目的重点是应对具有众多产品变种的手动系列装配中的老化问题。研究目的是建立典型装配任务学习过程的年龄差异模型，并预测学习过程的时间结构。研究目标由几个子目标组成。第一步是开发一个基于理论的学习过程时间结构的数学模型，其中既包括当前的学习理论，也包括进一步的概念方面。这既适用于工人的早期和核心过程，也适用于一段时间内学习进度的数学公式。通过在研究项目中扩展现有的知识，该模型将被陆续验证。为此，将进行几个年龄区分的研究。这些研究将分析年龄对学习和学习时间的影响，以及其他因素。研究的重点将是对尚未分析或已作为其他技术、业务或社会参数的一部分进行分析的因素进行实证研究，如不同类型的工作计划和不同的工作指导方法。此外，每项任务执行后的休息时间长短和时间类型对学习的影响也将是分析的一部分。这些影响可能在人类表现方面是同时存在的。所获得的结果和知识将被巩固，以便推导和开发有效的学习时间预测方法。该方法将为工业系列化生产中的典型手工任务的学习时间提供准确和稳健的估计。这将通过一种统计方法来实现，我在初步研究中已经成功地使用了这种方法。首先必须确定学习效果的数学描述，其参数将作为预测的一部分进行估计。与初步研究相比，将考虑其他因素。此外，还将使用更详细和更先进的统计方法，如向量自回归模型和周期向量自回归模型，其结构可以通过理论和经验数据来证实。在该项目中开发的模型的预测有效性将在最终的实验室研究中进行评估。

The research project focuses on coping with aging in manual series assembly with numerous product variants. The research objective is to produce an age-differentiated model and forecast of the time structure of learning processes for typical assembly tasks. The research objective consists of several sub-objectives. The first step is to develop a theory-based mathematical model of the time structure of learning processes that includes both current theories of learning and further conceptual aspects. This applies both to workers' early and central processes, as well as to the mathematical formulation of the learning progress over the course of time. By expanding existing knowledge during the research project, the model will be successively validated.To this end, several age-differentiated studies will be conducted. The impact of age on learning and learning time, as well as additional factors, will be analyzed in these studies. The focus of the studies will be on empirical research on factors that have not yet been analyzed or have been analyzed as part of other technical, operational or social parameters, such as different types of work plans and different job instruction methods. In addition, the influences of the length of breaks (after each task execution) and of chronotype on learning will be part of the analyses. These influences are probably conjunct in terms of human performance.The results and knowledge obtained will be consolidated in order to derive and develop a valid prediction method for learning times. The method will provide an exact and robust estimation of the learning time for typical manual tasks in industrial series production. This will be achieved with the help of a statistical approach which has already been used successfully in my preliminary studies. A mathematical description of the learning effect has first to be identified, the parameters of which are to be estimated as part of the prediction. Additional factors - compared to the preliminary studies - will be taken into account. In addition, more detailed and more advanced statistical approaches such as vector autoregressive and periodic vector autoregressive models, whose structure can be substantiated by theory and empirical data, will be used. The predictive validity of the model developed in the project will be assessed in a final laboratory study.