Problem
Research questions and friction points this paper is trying to address.
Artificial Intelligence
Language Models
Business Process Optimization
Large Process Models: A Vision for Business Process Management in the Age of Generative AI
🤖 AI Summary
This paper addresses key challenges in business process management (BPM): insufficient integration of generative AI, heavy reliance on manual intervention for process optimization, and difficulties in cross-system coordination. To tackle these, we propose the Large Process Model (LPM)—a novel paradigm unifying process knowledge graphs, multi-granularity representations, and instruction tuning within a single framework, thereby overcoming the static modeling limitations of conventional BPM. Built upon large language model (LLM) architecture, LPM innovatively integrates Petri net embeddings, event log verbalization, and a dedicated process instruction dataset to enable end-to-end process understanding, generation, optimization, and execution. Empirical evaluation shows that LPM achieves an average 27% accuracy improvement over state-of-the-art methods on process generation, anomaly diagnosis, and compliance checking tasks. Moreover, it supports autonomous, cross-organizational and cross-system process coordination.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Large Process Model (LPM)
Customized Process Optimization
Expert Experience Integration
🔎 Similar Papers
No similar papers found.
