Prediction-based Resource Allocation using Bayesian Neural Networks and Minimum Cost and Maximum Flow Algorithm
This addresses the gap in business process management where predictions lack actionable steps, offering a domain-specific solution for improving efficiency in processes like those in financial organizations.
The paper tackles the problem of connecting predictive business process monitoring to concrete improvements by optimizing resource allocation in a non-clairvoyant online environment, resulting in a method that integrates Bayesian Neural Networks for predictions with an extended minimum cost and maximum flow algorithm, validated on artificial and real-life event logs.
Predictive business process monitoring aims at providing predictions about running instances by analyzing logs of completed cases in a business process. Recently, a lot of research focuses on increasing productivity and efficiency in a business process by forecasting potential problems during its executions. However, most of the studies lack suggesting concrete actions to improve the process. They leave it up to the subjective judgment of a user. In this paper, we propose a novel method to connect the results from predictive business process monitoring to actual business process improvements. More in detail, we optimize the resource allocation in a non-clairvoyant online environment, where we have limited information required for scheduling, by exploiting the predictions. The proposed method integrates the offline prediction model construction that predicts the processing time and the next activity of an ongoing instance using Bayesian Neural Networks (BNNs) with the online resource allocation that is extended from the minimum cost and maximum flow algorithm. To validate the proposed method, we performed experiments using an artificial event log and a real-life event log from a global financial organization.