Agentic Business Process Management: The Past 30 Years And Practitioners' Future Perspectives

2025-03-23

arXiv.org

AI Summary

Generative AI–driven software agents exhibit inherent stochasticity, necessitating tailored governance mechanisms for their secure and effective deployment in organizational contexts—yet existing business process management (BPM) frameworks lack systematic responses to the paradigm shift toward agentic systems. Method: We conduct a systematic literature review (1994–2024) and in-depth interviews with 18 BPM practitioners to ground our analysis empirically. Contribution/Results: We introduce “Agentic BPM” as a novel conceptual framework—deliberately decoupled from current generative AI hype and prioritizing methodological rigor over tool-centricity. The study identifies practitioners’ core expectations: process autonomy, human–agent collaboration, and predictive process insights. It further surfaces critical practice bottlenecks—including data inconsistency and excessive manual intervention—and pinpoints key governance challenges: algorithmic bias, opacity, and workforce displacement. This work provides an original theoretical and operational pathway for evolving BPM theory and governance practice in the agentified era.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Agent/AI Theories and Architectures
• Philosophy and Ethics of AI: Artificial General Intelligence

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

With the advent of generative Artificial Intelligence (genAI), the notion of an agent has seen a resurgence in popularity. This has also led to speculation about the extent to which business process management, as a discipline and research field, may impact and be impacted by the deployment of genAI-based agents. To better ground such speculations into the state-of-the-art, we draw from the past 30 years of research on agents and business process management to establish the concept of Agentic Business Process Management (agentic BPM) that is only loosely coupled to the genAI hype. We conduct a series of interviews with BPM practitioners to explore their understanding, expectations, and concerns related to agent autonomy, adaptability, human collaboration, and governance in processes. The findings reflect both challenges with respect to data inconsistencies, manual interventions, identification of process bottlenecks, actionability of process improvements, as well as the opportunities of enhanced efficiency, predictive process insights and proactive decision-making support. While the technology offers potential benefits, practitioners also anticipate risks such as biases, over-reliance, lack of transparency, and job displacement within organizations. These concerns underscore the need for a robust methodological framework for managing agents in organizations.

Problem

• Governance challenges in deploying AI agents in business processes
• Identifying benefits and risks of AI agent adoption in organizations
• Aligning traditional BPM with agentic AI for effective integration

Innovation

• Qualitative study with 22 BPM practitioners
• Six key recommendations for AI adoption
• Align traditional BPM with agentic AI