Actor-Enriched Time Series Forecasting of Process Performance

2025-10-13

AI Summary

Existing business process throughput time (TT) prediction methods neglect the dynamic behavioral patterns of process participants, limiting predictive accuracy. Method: We propose an actor-enriched time series modeling paradigm that explicitly incorporates time-varying participant-centric features—including participation intensity, activity frequency, and task duration—as multivariate inputs alongside TT sequences. Using real-world event logs, we construct joint time-series representations and train end-to-end machine learning regression models. Contribution/Results: Experimental evaluation demonstrates statistically significant improvements over TT-only baselines across all key metrics: RMSE (−18.7%), MAE (−19.3%), and R² (+0.12). These results empirically validate that modeling participant behavior dynamics substantially enhances TT prediction fidelity. Our approach establishes a novel, interpretable, and scalable paradigm for process performance prediction, bridging a critical gap between human-centric process behavior and data-driven forecasting.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Predictive Process Monitoring (PPM) is a key task in Process Mining that aims to predict future behavior, outcomes, or performance indicators. Accurate prediction of the latter is critical for proactive decision-making. Given that processes are often resource-driven, understanding and incorporating actor behavior in forecasting is crucial. Although existing research has incorporated aspects of actor behavior, its role as a time-varying signal in PPM remains limited. This study investigates whether incorporating actor behavior information, modeled as time series, can improve the predictive performance of throughput time (TT) forecasting models. Using real-life event logs, we construct multivariate time series that include TT alongside actor-centric features, i.e., actor involvement, the frequency of continuation, interruption, and handover behaviors, and the duration of these behaviors. We train and compare several models to study the benefits of adding actor behavior. The results show that actor-enriched models consistently outperform baseline models, which only include TT features, in terms of RMSE, MAE, and R2. These findings demonstrate that modeling actor behavior over time and incorporating this information into forecasting models enhances performance indicator predictions.

Problem

• Incorporating actor behavior as time-varying signals improves process performance forecasting
• Modeling actor-centric features enhances throughput time prediction accuracy
• Actor-enriched time series outperform baseline models in predictive process monitoring

Innovation

• Incorporating actor behavior as time-varying signals
• Constructing multivariate time series with actor-centric features
• Actor-enriched models outperform baseline performance indicators