Enhancing Robotic Process Automation Task Selection: An Integrated Approach Leveraging Process Mining and Feature Extraction

Abstract

Robotic Process Automation (RPA), an emergent technology, is increasingly being utilized for the automation of straightforward and structured tasks, due to its time efficiency and cost effectiveness. As organizations strive to automate processes, it becomes imperative to discern the most suitable technology for each task to optimize investments in automation. The surge in RPA usage illuminates the challenge of task selection for automation. In response to this challenge, our study presents an integrated approach of process mining and feature extraction to enhance RPA task selection. Organizations provide feature weights, based on which corresponding tasks are extracted. Each task is subsequently ranked, and an overall task rank is computed by summing the products of feature weights and individual feature ranks. This procedure is iteratively performed for all tasks, culminating in a feature matrix, which constitutes the output of this framework. By leveraging historical process data, this combined approach allows for the identification of tasks that exhibit characteristics amenable to automation, such as high frequency, low variability, and distinct decision points. Furthermore, the extraction of task features enables the prioritization of tasks based on their potential for automation, complexity, and anticipated benefits. Through the analysis of process mining data, this study offers an empirical snapshot of organizational activities and suggests tasks that are amenable to RPA. This prioritization of suitable tasks for automation potentially enhances the success of RPA implementation.