Temporal Stability in Predictive Process Monitoring

TL;DR

This paper introduces the concept of temporal stability in predictive process monitoring, evaluates existing methods on stability and accuracy, and proposes techniques to improve stability without significantly sacrificing accuracy.

Contribution

It defines temporal stability for binary classification in process monitoring and demonstrates how to enhance it through hyperparameter tuning and smoothing techniques.

Findings

XGBoost and LSTM exhibit high temporal stability

Hyperparameter optimization improves stability of classifiers

Smoothing techniques can increase stability with minor accuracy loss

Abstract

Predictive process monitoring is concerned with the analysis of events produced during the execution of a business process in order to predict as early as possible the final outcome of an ongoing case. Traditionally, predictive process monitoring methods are optimized with respect to accuracy. However, in environments where users make decisions and take actions in response to the predictions they receive, it is equally important to optimize the stability of the successive predictions made for each case. To this end, this paper defines a notion of temporal stability for binary classification tasks in predictive process monitoring and evaluates existing methods with respect to both temporal stability and accuracy. We find that methods based on XGBoost and LSTM neural networks exhibit the highest temporal stability. We then show that temporal stability can be enhanced by hyperparameter-optimizing random forests and XGBoost classifiers with respect to inter-run stability. Finally, we show that time series smoothing techniques can further enhance temporal stability at the expense of slightly lower accuracy.