What BPM Pros Really Think About AI and A/B Testing Process Change
The application of DevOps principles to business process management (BPM) aims to facilitate continuous business process improvement. The AB-BPM methodology, specifically, integrates AB testing and reinforcement learning to enhance the speed and quality…
The application of DevOps principles to business process management (BPM) aims to facilitate continuous business process improvement. The AB-BPM methodology, specifically, integrates AB testing and reinforcement learning to enhance the speed and quality of improvement efforts. This approach has been assessed for its requirements, risks, opportunities, and other aspects through a qualitative analysis combining grounded theory with a Delphi study. This included semi-structured interviews and follow-up surveys with a panel of business process management experts.
The analysis indicates the necessity of human oversight during reinforcement learning-driven experiments. Furthermore, it emphasizes the importance of aligning the methodology culturally and organizationally within the respective setting and advocates for an integrated process execution platform.
Business processes are essential for value creation and the delivery of products and services. Enhancing these processes is crucial for gaining a competitive edge, improving value delivery, and increasing efficiency and customer satisfaction. Business Process Improvement (BPI) is a critical component of BPM, which involves overseeing organizational work to ensure consistent outcomes and capitalize on improvement opportunities.
The application of DevOps principles to business process management (BPM) aims to facilitate continuous business process improvement.
DevOps, a convergence of development and operations, is designed to minimize the time between committing changes to a system and their deployment into production while maintaining quality. In BPM, the integration of DevOps principles, such as AB testing, is proposed to enable continuous BPI through the AB-BPM method. AB testing evaluates different software feature versions with real users in a production environment and retires the current version only if test data supports the improvement hypothesis. The AB-BPM methodology extends traditional AB testing by proposing reinforcement learning to dynamically route incoming process instantiation requests based on performance measurements.
The AB-BPM approach has not yet been systematically analyzed from the perspective of BPM practitioners. Additional research is required to enhance confidence in this approach. Insights from BPM practitioners are valuable for identifying challenges and opportunities for increased automation in process (re-)design. This study presents an analysis of BPM experts' views on the impact, advantages, and challenges of the AB-BPM method in an industry setting. It examines their overall sentiment, perceived risks, potential use cases, technical feasibility, and software support requirements. A panel of BPM experts from a large software company contributed to this study through interviews and follow-up surveys, employing a mix of grounded theory and Delphi research methodologies.
The complete paper is available on arXiv under the CC BY 4.0 DEED license.
Based on reporting by hackernoon.com.