Iman Heidari - Masters Student of Industrial Engineering, Tarbiat Modares University
Mohammad Amin Pirian - Masters Student of Industrial Engineering, Tarbiat Modares University
Toktam Khatibi - Associate Professor at the Faculty of Industrial and Systems Engineering, Tarbiat Modares University
Mohammad Mehdi Sepehri - Professor at the Faculty of Industrial and Systems Engineering, Tarbiat Modares University

Process Analytics methodologies empower organizations to optimize Business Process Management and continuous improvement by leveraging process-related data for knowledge extraction, enhancing process performance, and facilitating data-driven decision-making across the organizational spectrum. The aggregated process execution data contains valuable insights and actionable intelligence, enabling the identification of performance bottlenecks, cost reduction strategies, insights derivation, and resource utilization optimization. These methodologies encompass information extraction from event logs, facilitating process model discovery, monitoring, and refinement. A critical application within process analytics is the predictive monitoring of business processes, aiming to forecast quantifiable metrics for ongoing process instances through the development of predictive models. In this paper, we provide an outline of fundamental principles and present a comprehensive evaluation of the domain of predictive process monitoring, We also perform a thorough and methodical examination of the utilization of deep learning methods in predictive monitoring for business processes. This review encompasses a wide array of existing methodologies and their potential contributions to the enhancement of predictive capabilities within Business Process Management systems.

Predictive process monitoring, business process management, Process mining, Deep learning