Probabilistic Declarative Process Mining



ERC field: PE - Physical Sciences and Engineering

Duration: 24 months, from  28/09/2023 to 27/09/2025

Project code: 20224C9HXAb 

Total budget: 301.388,00 €

Budget funded by MUR :  199.920,80 €


Principal Investigator: Elena Bellodi

UniFe - Università degli Studi di Ferrara

Associate Investigator: Federico Chesani

UniBo - Università degli Studi di Bologna

Associate Investigator: Mario Alviano

UniCal - Università della Calabria


process mining, knowledge discovery, probabilistic logic programming, answer set programming


The modelling of business processes is pivotal to support decision-making in complex industrial and corporate domains.

Recent years have seen the birth of the Business Process Management research area, focused on the analysis and control of process execution quality, and in particular, the rise in popularity of Process Mining, which encompasses a set of techniques to extract valuable information from event logs. 

The rise in popularity is testified by a significant growth in research conducted on the development and improvement of algorithms in the last 25 years, by the birth of dedicated Conference Series, by important academic and industrial frameworks for supporting a wide variety of Process Mining techniques and types of models, and by recent studies on  evidence on the application of Process Mining in organisations.


The PRODE project stands at the forefront of enhancing business process management through the innovative use of Process Mining techniques. Central to its mission is the exploration and implementation of declarative process models derived from event data, emphasizing binary supervised learning within the context of uncertainty. 

A key innovation of PRODE lies in its operational framework, which is grounded in computational logic, specifically combining the strengths of Probabilistic Logic Programming and Answer Set Programming. This fusion is designed to significantly enhance the expressive capabilities of DECLARE frameworks, thereby improving the reasoning and learning tasks associated with process modeling. Through this approach, PRODE intends to tackle real-process issues marking a significant leap towards more effective and comprehensive process management solutions.


PRODE aims to address critical challenges in the business process management field,  balancing the precision and understandability of models, managing inherent log uncertainties, navigating compliance issues through probabilistic assessments, and optimizing model selection to enhance workflow management. 

Drawing on extensive expertise in Process Mining, Artificial Intelligence, Machine Learning, and Logic Programming, PRODE is committed to validating its findings through rigorous formal and empirical analyses across diverse case studies. This initiative not only aims to refine process management practices but also to contribute significantly to the fields of knowledge discovery and organizational efficiency.