Cunha,Artur Manuel Barros daHeinemann,Dominik2026-04-142026-04-142025http://hdl.handle.net/10400.5/118030Trabalho Final de Mestrado, Gestão de Sistemas de Informação, ISEG, 2025.As business processes grow more complex and interconnected, organizations face increasing pressure to choose the right analytical tools to understand their operations. Traditional Process Mining (TPM), which relies on case-centric event logs, has long been the standard approach due to its tool maturity and ease of use. However, it often struggles to capture the nuances of multi-entity systems. Object-Centric Process Mining (OCPM) offers an alternative by preserving relationships between multiple object types, enabling more detailed insights into coordination and concurrency. This thesis explores the comparative strengths and limitations of TPM and OCPM through a mixed-methods approach. A literature-based framework was developed and applied to two contrasting datasets: a structured administrative workflow from the BPI Challenge 2017 and a complex, dynamic object-centric log derived from Age of Empires II game telemetry. The comparison focused on eight analytical dimensions, including scalability, model complexity, and interpretability. Based on these findings, a decision framework is proposed to help practitioners identify which technique is more suitable for their context. While TPM remains a strong option for straightforward processes and fast implementation, OCPM proves advantageous in capturing inter-object interactions and revealing deeper insights in complex environments. The framework aims to support more informed, case-specific method selection in both academic and applied process mining work.application/pdfengProcess MiningObject-Centric Process MiningTraditional Process MiningDecision Frameworkmaster thesis