Towards a Taxonomy of Cognitive RPA Components

Robotic Process Automation (RPA) is a discipline that is increasingly growing hand in hand with Artificial Intelligence (AI) and Machine Learning enabling the so-called cognitive automation. In such context, the existing RPA platforms that include AI-based solutions clas sify their components, i.e. constituting part of a robot that performs a set of actions, in a way that seems to obey market or business deci sions instead of common-sense rules. To be more precise, components that present similar functionality are identified with different names and grouped in different ways depending on the platform that provides the components. Therefore, the analysis of different cognitive RPA platforms to check their suitability for facing a specific need is typically a time consuming and error-prone task. To overcome this problem and to pro vide users with support in the development of an RPA project, this paper proposes a method for the systematic construction of a taxonomy of cognitive RPA components. Moreover, such a method is applied over components that solve selected real-world use cases from the industry obtaining promising results