Mathematical Framework for Agentic Business Process Management Systems Published
A study outlines the mathematical principles underlying agentic business process management systems, which stand apart from conventional BPM by utilizing several autonomous decision-makers known as agents. These systems enhance process specifications with clear objectives designated for each agent, who strive to achieve these targets while considering the anticipated actions of others through suitable strategies. The organization implementing the process can set boundaries on the agents' decision-making abilities at the strategy level using these specifications. The research investigates four core challenges of agentic BPM within three primary contexts, offering formal mathematical frameworks for such systems. This work, published on arXiv in the Computer Science > Artificial Intelligence category, enriches the theoretical framework of autonomous agent systems in organizational settings.
Key facts
- Agentic BPM systems use multiple autonomous decision-makers called agents
- Process specifications include explicit objectives assigned to agents
- Agents pursue goals under assumptions about others' behavior
- Organizations can impose guardrails on agents' decision-making capabilities
- The paper establishes mathematical foundations for such systems
- Four foundational problems of agentic BPM are analyzed
- Research covers three key settings for agentic BPM
- Published on arXiv under Computer Science > Artificial Intelligence
Entities
Institutions
- arXiv