Abstract Module-based plant design opens up the opportunity for the (bio-)chemical industry to reduce lead times, which is crucial for future competitiveness. Equipment modules are designed once such that they can cover a wide range of process tasks and conditions. The time-consuming equipment design step is replaced by selecting the most suitable equipment module from an equipment module database so that engineering work is reused. It is the aim of this work to develop a structured approach for the determination of features to characterize industrial process tasks. On the one hand, these characteristic features determined are required for the generation of an equipment module database based on cluster analysis of industrial process tasks (Eilermann et al., 2017). On the other hand, the characteristic features are a measure for similarity of process tasks and support the selection of the same equipment module for similar tasks. Thereby, reuse of engineering is enabled. The approach for the determination of characteristic features presented is based on those used in data mining, whereas the quality of the characteristic features is traded off against the computational effort required. The approach developed is exemplarily applied to industrial liquid/liquid heat exchanger and condenser tasks kindly provided by Evonik.

中文翻译：

---

作为设备模块数据库生成和应用基础的工业过程任务表征方法

摘要 基于模块的工厂设计为（生物）化工行业提供了减少交货时间的机会，这对未来的竞争力至关重要。设备模块设计一次，因此它们可以涵盖广泛的过程任务和条件。通过从设备模块数据库中选择最合适的设备模块来代替耗时的设备设计步骤，以便重复使用工程工作。这项工作的目的是开发一种结构化方法，用于确定表征工业过程任务的特征。一方面，这些确定的特征是基于工业过程任务的聚类分析生成设备模块数据库所必需的（Eilermann et al., 2017）。另一方面，特征特征是衡量过程任务相似性的一种度量，支持为相似任务选择相同的设备模块。从而，实现了工程的重用。用于确定特征特征的方法基于数​​据挖掘中使用的方法，而特征特征的质量与所需的计算工作量进行权衡。所开发的方法示例性地应用于赢创提供的工业液体/液体热交换器和冷凝器任务。而特征特征的质量与所需的计算工作量进行权衡。所开发的方法示例性地应用于赢创提供的工业液体/液体热交换器和冷凝器任务。而特征特征的质量与所需的计算工作量进行权衡。所开发的方法示例性地应用于赢创提供的工业液体/液体热交换器和冷凝器任务。