Paper Type

Complete

Abstract

This study examines the shift from rule-based Robotic Process Automation (RPA) to goal-driven agentic AI through the Business Process Management (BPM) lifecycle. A structured qualitative artifact analysis of leading automation platforms identifies a “lifecycle collapse,” in which modeling, execution, and optimization converge into a continuous autonomous loop, dissolving design-time/run-time boundaries central to governance. While this integration mitigates RPA’s rigidity and brittle exception handling, it creates two tensions: an accountability gap, as absent pre-approved process models undermine auditability, and a “deskilling paradox,” where humans elevated to intent architects lose procedural literacy needed to detect subtle reasoning failures. This study frames agentic automation as both capability expansion and governance risk, contributing to BPM and responsible AI research by identifying the governance tensions that emerge when autonomous systems dissolve established design-time and run-time boundaries. For service science, the findings advance understanding of how redistributed agency reshapes value co-creation and human oversight in autonomous service environments.

Paper Number

1760

Comments

SIG SVS

Download

Share

COinS
 
Aug 15th, 12:00 AM

Process Automation Revisited: From Rule-Based to Agentic AI in Service Systems

This study examines the shift from rule-based Robotic Process Automation (RPA) to goal-driven agentic AI through the Business Process Management (BPM) lifecycle. A structured qualitative artifact analysis of leading automation platforms identifies a “lifecycle collapse,” in which modeling, execution, and optimization converge into a continuous autonomous loop, dissolving design-time/run-time boundaries central to governance. While this integration mitigates RPA’s rigidity and brittle exception handling, it creates two tensions: an accountability gap, as absent pre-approved process models undermine auditability, and a “deskilling paradox,” where humans elevated to intent architects lose procedural literacy needed to detect subtle reasoning failures. This study frames agentic automation as both capability expansion and governance risk, contributing to BPM and responsible AI research by identifying the governance tensions that emerge when autonomous systems dissolve established design-time and run-time boundaries. For service science, the findings advance understanding of how redistributed agency reshapes value co-creation and human oversight in autonomous service environments.