Detecting and Explaining Deviations in Regulated Processes
Inside the loan process
In this Research in a Minute video, postdoctoral researcher Michael Grohs explains how data-driven methods can be used to analyze and improve such processes, using the loan application as an example. This process underlies many regulations and stakeholder interests, such as banking regulations that limit risks, anti-discrimination laws that ensure fairness, and customer expectations for fast decisions and transactions.
Aligning performance, compliance, and fairness
A data-driven approach helps identify whether all these requirements are being met by automatically detecting deviations, explaining their causes, and even predicting potential future issues. This has a tangible impact:
- It helps organizations ensure compliance with regulations
- It contributes to fairer outcomes by identifying potential discrimination
- It improves efficiency and service quality, leading to faster and more transparent processes
Watch the Research in a Minute video to get the insights from the study in one minute.
Further Readings:
The full research papers can be read here:
- A Task Taxonomy for Conformance Checking.
- Proactive conformance checking: An approach for predicting deviations in business processes.
Research by the same authors on this topic:
- Beyond Log and Model Moves in Conformance Checking: Discovering Process-Level Deviation Patterns.
- A Procedural Framework for Assessing the Desirability of Process Deviations.
- Large Language Models Can Accomplish Business Process Management Tasks.
Watch this one-minute video here:
