ABSTRACT Engineering design automation has been an active field of research and application for more than five decades. Despite a multitude of available methods stemming from research fields such as Knowledge-Based Engineering (KBE) and Computational Design Synthesis (CDS) the application context in industry is mostly limited to routine tasks. Here we propose a means to foster the transition of academic methods to industrial practice through a comprehensible and comprehensive design automation task categorisation that allows practitioners to grasp the opportunities state-of-the-art design automation offers. The categorisation is based on a critical investigation of 77 papers stemming from KBE and CDS, respectively, and is detailed with respect to knowledge needed for complete formalisation of a task, i.e. inputs, outputs and goals, as well as available automation methods. A thorough discussion indicates the context of the work with current design practice and provides a consolidation of the research fields KBE and CDS including the derivation of technological requirements for design automation task definition by practitioners. Thus, this paper contributes not only by consolidating two research fields but also by presenting technological requirements for further streamlining the process of realising design automation applications in industry.

中文翻译：

---

用于识别设计自动化机会的任务分类

摘要 五十年来，工程设计自动化一直是一个活跃的研究和应用领域。尽管有多种可用方法源自基于知识的工程 (KBE) 和计算设计综合 (CDS) 等研究领域，但工业中的应用环境大多仅限于常规任务。在这里，我们提出了一种方法，通过可理解和全面的设计自动化任务分类来促进学术方法向工业实践的过渡，使从业者能够抓住最先进的设计自动化提供的机会。分类基于对分别来自 KBE 和 CDS 的 77 篇论文的批判性调查，并详细介绍了完成任务形式化所需的知识，即输入、输出和目标，以及可用的自动化方法。彻底的讨论表明了当前设计实践的工作背景，并提供了 KBE 和 CDS 研究领域的整合，包括从业者对设计自动化任务定义的技术要求的推导。因此，本文不仅通过整合两个研究领域做出贡献，而且还提出了进一步简化在工业中实现设计自动化应用的过程的技术要求。