Celosphere 2019 Keynote: Process Mining - Past, Present, and Future
wvdaalst
19 views
28 slides
Nov 02, 2025
Slide 1 of 28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
About This Presentation
Keynote by Wil van der Aalst at the first Celosphere conference organized by Celonis in 2019. The first Celosphere attracted over 1000 participants. The keynote discusses the Past, Present, and Future of process mining.
From April 2nd to 4th, 2019, the process mining community gathered in Munich...
Keynote by Wil van der Aalst at the first Celosphere conference organized by Celonis in 2019. The first Celosphere attracted over 1000 participants. The keynote discusses the Past, Present, and Future of process mining.
From April 2nd to 4th, 2019, the process mining community gathered in Munich's Eisbach Studios for the first Celosphere.
The presentation “Process Mining: Past, Present, and Future” provides a reflective overview of how process mining has evolved over two decades, where it stands today, and how it will shape the future of data-driven process management. It begins by looking back at the origins of the field, highlighting the shift from traditional process management by modeling—rooted in Petri nets, BPM, workflow management, and simulation—to process management by mining, where real event data are used to discover, monitor, and improve processes. Key milestones mark the field’s growth, from the first algorithms such as the Alpha and Heuristic Miner and the development of the open-source tool ProM, to the founding of Celonis in 2011, the release of the first process mining books and MOOCs, and the recognition of process mining as a mainstream technology by Gartner in 2018.
Despite the vast potential, adoption remains hindered by human and data challenges: limited awareness, lack of training, reluctance toward transparency, and persistent data quality issues. A crucial distinction is drawn between process mining and traditional data mining—while the latter focuses on patterns in data, process mining explicitly reconstructs and evaluates process behavior over time.
Looking ahead, the presentation envisions a future “towards better processes”, in which process mining becomes more forward-looking and prescriptive. The focus shifts toward hybrid models seamlessly integrating process discovery and conformance checking, comparative process mining for benchmarking, and causal analysis to suggest real improvements rather than reinforce correlations. Making results actionable—through integration with robotic process automation (RPA) and other automation technologies—is seen as essential, as is embracing responsible process mining that safeguards fairness, confidentiality, and trust. The talk concludes with optimism: process mining has matured into a thriving global discipline, culminating in the launch of the International Conference on Process Mining (ICPM 2019), and continues to bridge data science and business improvement in transformative ways.