Advances in Rare Event Simulation and Process Mining

The field of rare event simulation and process mining is moving towards the development of more efficient and accurate methods for modeling and analyzing complex systems. Researchers are exploring new approaches to simulate rare events, such as the use of time-sensitive importance functions, and to integrate data from various sources, including IoT devices. The goal is to improve the accuracy and reliability of predictions, and to enable more effective fault diagnosis and maintenance in cyber-physical systems. Noteworthy papers in this area include:

  • Time-Sensitive Importance Splitting, which introduces a new approach to rare event simulation using a time-sensitive importance function.
  • Process mining-driven modeling and simulation to enhance fault diagnosis in cyber-physical systems, which presents a novel unsupervised fault diagnosis methodology that integrates collective anomaly detection, process mining, and stochastic simulation.

Sources

Time-Sensitive Importance Splitting

Timed Prediction Problem for Sandpile Models

An object-centric core metamodel for IoT-enhanced event logs

Process mining-driven modeling and simulation to enhance fault diagnosis in cyber-physical systems

Built with on top of