Generating multiperspective process traces using conditional variational autoencoders

In recent years, trace generation has emerged as a significant challenge within the Process Mining community. Deep Learning (DL) models have demonstrated accuracy in reproducing the features of the selected processes. However, current DL generative models are limited in their ability to adapt the learned distributions to generate data samples based on specific conditions or attributes. This limitation is particularly significant because the ability to control the type of generated data can be beneficial in various contexts, enabling a focus on specific behaviours, exploration of infrequent patterns, or simulation of alternative “what-if” scenarios. Moreover, many existing models primarily focus on generating traces that capture only the control-flow and temporal perspectives, neglecting crucial aspects such as resource and data perspectives, which are essential to understanding business process executions. In this work, we address these challenges by introducing a conditional model for multiperspective process data generation based on a conditional variational autoencoder (CVAE). Conditional models offer control over the generation process by tuning input conditional variables, enabling more targeted and controlled data generation. Unlike other domains, CVAE for process mining faces specific challenges due to the multiperspective nature of the data and the need to adhere to control-flow rules while ensuring data variability. Specifically, we focus on two main objectives: (i) generating complete trace executions that include control flow, temporal data, and other data attributes, with a particular focus on trace attributes and resources, as they are the most common attributes in business processes; and (ii) conditioning the trace generation on specific control flow and temporal conditions, enabling the production of traces that align to the desired execution scenarios defined by the condition context. The generated traces are then evaluated using common metrics for generative model assessment, along with additional metrics to evaluate the quality of the conditional generation.