Conceptual Bot Platform

A meta repository for the components of the conceptual bot platform

The conceptual bot platform serves as a software system for modeling and analyzing conceptual RPA bots based on the ontology of RPA operations. Furthermore, modeled conceptual RPA bots can be linked, i.e. translated, to existing RPA tools, currently Robot Framework and taskt.

Demonstration of the modeling component, including data resources and variables:

Screencast of the platform featuring the modeling (video does not show modeling of data added later) and translation components (https://youtu.be/Pq5FIS9KtqA):

For illustrations of the newer components, see the respective modules for abstraction and the complexity metrics.

Available components

So far, the platform consists of a front end (the modeler), and a back end for storing and linking bots.

Conceptual RPA Bot Modeler

This modeler currently comprises the following functionality:

• Parsing the ontology of RPA operations
• Mapping from BPMO concepts to BPMN
• Modeling conceptual bots using the Business Process Model and Notation using operations from the ontology
• Support for different types of operations, such as triggers, decisions, and context containers
• Parsing BPMN models to generic process trees that are sent to the back end for storing
• Analzying stored bots for included operations and concepts, i.e. searching the bot repository for bots automating certain applications or using specific operations
• Apply different abstraction strategies to models and view the result
• Automatically compute various complexity metrics for models

Conceptual RPA Bot Back End

The back end, serving as the bot repository, currently offers the following features:

• Storing the generic process tree and the visual BPMN representation of RPA bots
• Linking, i.e., translating, stored bots to either Robot Framework (.robot) or taskt files (.xml)

Installation

Requirements:

• node.js
• Docker or directly MongoDB

1. You need to clone/download the sources for the back end and the front end.
2. Run npm install in each directory (BE+FE)
3. Make sure you have MongoDB running at port 27017.
4. Run npm run dev in each directory (BE+FE)