🤖 AI Summary
This work addresses the limitations of traditional process discovery methods, which struggle to fully exploit the expressive power of Petri nets due to reliance on predefined structures and inability to capture long-range dependencies. To overcome these constraints, the paper proposes a monotonicity-guided, bottom-up approach to Petri net discovery that departs from conventional top-down paradigms. By leveraging local monotonicity analysis, pruning candidate places, and applying combinatorial optimization, the method efficiently constructs high-quality process models capable of representing concurrency, free-choice behavior, and long-term dependencies. Experimental results demonstrate that, on both synthetic and real-life event logs, the approach generates expressive and structurally flexible Petri nets using limited computational resources, significantly improving the accuracy and adaptability of process discovery.
📝 Abstract
Process discovery is one of the central challenges in process mining. Petri nets are particularly attractive because simple local constructs can express complex behavior, including concurrency. While their global behavior may be difficult to analyze, individual places can be efficiently characterized using monotonic properties, enabling bottom-up discovery. Unlike top-down approaches such as the Inductive Miner, which rely on predefined constructs for sequences, choices, loops, and concurrency, our approach allows such structures to emerge organically and can exploit the full expressive power of Petri nets, including free-choice constructs and long-term dependencies. The main challenge is the exponential number of candidate places and their combinations. We present the SPECpp framework which implements strategies to obtain high-quality models under time and resource constraints. SPECpp supports rapid experimentation and is used to evaluate these strategies using both synthetic and real-life event data.