Process mining techniques, which reconstruct how processes are executed, can provide valuable input for process improvement initiatives but do not automatically identify concrete process weaknesses. Identifying such weaknesses still requires the extensive involvement of domain experts. At the same time, process problems experienced by customers are frequently shared on social media platforms. The goal of this project is, therefore, to address the problem of the manual work in the context of process improvement by developing techniques that automatically link process weaknesses described in social media posts to specific events from event logs, and rank the aligned weaknesses, so that they can effectively serve as input for process improvement.
Against the background of intense competition and rapidly changing demands from customers, many organizations strive for continuously improving their business processes. Over the last couple of years, process mining techniques have become increasingly popular to support this endeavor. They provide insights into what exactly happens during process executions by analyzing so-called event logs, which are extracted from IT systems. These techniques allow to detect undesired patterns, bottlenecks, and compliance issues, among others, but still require significant manual effort from domain experts to identify concrete weaknesses in the process execution that can be picked up for improvement (e.g., the waiting time for receiving a voucher). This project aims to address the problem of the manual effort and to develop techniques that automatically identify process weaknesses relating to specific events. To achieve this, we exploit a widely available textual resource that captures the customer’s perspective and thereby allows to detect weaknesses that could not have been identified by analyzing the event logs only: social media posts. The core idea is to extract process-related weaknesses described in social media posts using natural language processing tools, to align these weaknesses with the events of event logs using optimisation rules, and to cluster and rank the aligned weaknesses, so that they can effectively serve as input for process improvement initiatives.
Kühne Logistics University (KLU)