This is the channel of Wil van der Aalst, full professor at RWTH Aachen University, where he leads the Process and Data Science (PADS) group. He is also the Chief Scientist at Celonis, part-time affiliated with the Fraunhofer FIT, and a member of the Board of Governors of Tilburg University. Currently, he is also the deputy CEO of the Internet of Production (IoP) Cluster of Excellence and co-director of the RWTH Center for Artificial Intelligence. Van der Aalst received an Alexander-von-Humboldt Professorship and honorary degrees from the Moscow Higher School of Economics (Prof. h.c.), Tsinghua University, and Hasselt University (Dr. h.c.). He is also an IFIP Fellow, IEEE Fellow, ACM Fellow, and an elected member of the Royal Netherlands Academy of Arts and Sciences, the Royal Holland Society of Sciences and Humanities, the Academy of Europe, and the North Rhine-Westphalian Academy of Sciences, Humanities and the Arts.
Пікірлер
❤❤❤❤
Great tutorial! Been researching on process mining for some time. Would love to work with you someday!
How to find trap ,siphons, liveness, and deadlock. It will be helpful if you will explain with example.
At 1:19:40 Why is the left-hand side Petri net NOT deadlock-free? T2 will always fire p1 and p3 at the same time. Once p1 has a token, then p2 will have a token that can fire with the token in p3 together, and then no tokens left in the net. Does it mean it is a deadlock?
Awesome teaching quality. Can you create a course on Coursera? Would be great
many thanks
starts at 39:39
Well Explained
Very usefull
Thank you so much respected sir for giving a wonderful explanation on OCPM 🙏
thank you... all about mathematics
Thank you for your excellent lectures, it's really good for me to understand Process Mining!!!
Thank you..
1:14:03 That should be an f in the righ-most transition right?
Yes, of course! Sorry for the typo.
great info but, plan your lectures better, you stammer and pause a lot
Hello professor Will van der Aalst, this is Elmer Gonzalez professor in a peruvian university in South America and I realize that this lecture is a great complement of your books in Process Mining. I will appreciate if you can help me in finding more information about how to use ProM since I am using it in my classes about Process Mining and how ProM can discover Petri net model, BPM model almost automatically.
Thank you so much for all the information, that you share. You help me to improve my professionality.
Dear Wil van der Aalst, thank you for this lesson. I have been following you on youtube and linkedin for a year and my goal is to do a data science master's and take your course. I hope next year I can :)
I find the comparison of the tools based on the data set of 2000 purchase orders and their modifications extremely interesting. I have implemented a modification idea to the simple directly-folllows approach. Is the data you used to show the shortcomings of some commercial tools available for testing purposes?
I'm not 100% sure about these event logs. This was some time ago, and I have many. The zip file www.dropbox.com/s/6v6cqvll78sc0co/some-event-logs.zip?dl=0 contains three related event logs, and I think these are the ones I used here.
Hay poco material en castellano
Thank you for this wide structure, and concrete examples. Merhaba from Turkey!
Hi Prof! I am a student from China and my graduation project is about process mining. PM is really interesting and thanks for your lecture! Waiting for the update, especially the part of combination with deep learning!!!! I can't find record but only slides from summer school official websites.
The videos will be online soon. However, see lectures 14-16 of the Process Mining Course @ RWTH Aachen University (BPI 2021) kzread.info/head/PLG_1ZxIPXO0uRZtlYxaLgc62kKfko8QQ6
Please I would like to know a process enhancement tool
The more mature higher-end process mining tools privide automation capabilities, e.g., Celonis embeds the Make/Integromat low-code automation tools. This way compliance and performance gaps are translated into actions in the source systems.
Thanks for your tutorial. I would like to know for the third part of process mining which process optimisation are they tools that help to do it or based from the analysis made from the conformance checking phase you must know how to optimise it?
Sure, but this is not in the course itself. Performance and compliance problems identified using process mining should trigger corrective workflows. This is the reason that for example Celonis support Action Flows connecting to over 1000 different types of systems (SAP, etc.) to take corrective actions.
@@wilvdaalst Okay thanks I saw the video on process discovery and compliance checking and it was vague according to me I could not find the techniques to apply , I would like to what are the actions to do in order to do process discovery and compliance checking. I am using Disco as process mining tool.
@@christbryannokamfotso3252 Three conformance checking techniques are discussed in detail. Note that ProM supports these, but Disco does not. Also note that the more mature higher-end process mining tools privide automation capabilities, e.g., Celonis embeds the Make/Integromat low-code automation tools. This way compliance and performance gaps are translated into actions in the source systems.
See response in other lecture.
😳 p̲r̲o̲m̲o̲s̲m̲
What software/package do you use to draw the causal net on your slides? Thank you!
ProM. Use the Heuristic Miner implemented by Felix.
What is easy to say is that BPM did not accomplish the goals that it set out to achieve, but that was mostly due to egregiously overhyped ideations of the amount of effort that it would take to monitor a living business process in a meaningful enough way that the business sponsors of those efforts could feel empowered that they were still in possession of decision making authority. I also think the intention to define OneTrueWay processes undercut the effort because politics invariably come into play when "there can be only one." Further exacerbating the issue is that implementation of BPM-centric solutions requires some fairly decently advanced skillsets, companies naively thought that anyone who had a degree was naturally imbued with these skills, then they death marched them into mine fields that were far too heavily mined for them to traverse successfully. That, in my opinion, was how it "failed."
I agree with what you say. However, I would add that old-school BPM-ers tend to resist data-driven approaches and stick to practices that only make sense in less digitalized environments.
@@wilvdaalstOof, that's a painful truth! It runs face first into the famous quotes about solving problems with the same thinking that caused them. And I've found it is especially challenging to solve problems with the same *_people_* that caused them! They created these systems because they thought they were right, so they don't want to hear us say otherwise.
"We had an army of people building this problem for decades (if not centuries). It's been six months since you graduated - why haven't you fixed it yet?!"
Very good! It's time to expand the BPM to help optimize and improve the delivery of value. It's necessary to integrate different disciplines and tools, to help the business from the Design, Simulation, Analysis, Dashboards and evolve the structure of the applications to record the events and that they are easily extracted for analysis. Long life to BPM!
So the concept of BPM did not fail. The snags it ran into triggered the creation of the two lifelines being Process Mining and RPA.
Agree.
awesome summary! thanks for sharing your knowledge, saludos desde Uruguay
Hello thanks! Have a question: How do you know that you have a good process model?
The model itself can be checked using a soundness check (use the corresponding ProM plug-in).
Great lecture! :) I wonder what plug-in is used in ProM for conformance checking
Alignment-based conformance checking
Could you please provide the csv data
Thank you Sir, really appreciate the insight you provided through your lectures, God Bless <3
This is so underrated thanks
Inductive miner excellently explained. Thank you professor.
Dear Professor @Wil van der Aalst, This was a great presentation and an eye-opener for me. Would it be possible for you to share the slides, please?
thank you great Prof for this. Information on this out there is a bit scarce. im happy i found this
He also wrote a great (i.e. the) book on the topic
Great Lecture :) thank you very much
I was in search of content related to process mining ..Thank you so much for your efforts
Thank you professor this lecture series is highly interesing
Hi Wil van der Aalst, thank you for sharing these interessanti lectures. Is it possible to define the set of complete traces of a petri net as a regular expression?
If the Petri net is bounded it can be converted into a finate automaton. Using Kleene's algorithm transforms a given nondeterministic finite automaton (NFA) into a regular expression. However, the regular expressions will not be readable and may be extremely long and complicated. Hence, it has no real relevance.
Thank you @@wilvdaalst for the reply. My question about regular expression arise in order to solve this problem: given a set of Petri Nets is it possible to determine if there exist complete traces that belong to more then one Petri Net? Maybe there are better way, but regular expression seem a good representation to do this computation.
@@GivanniDrogo If the Petri nets are bounded you can just compute the NFAs for each Petri net and check if the pairwise intersection of the NFAs. The intersection two NFAs will give you the NFA with the intersection of both accepted languages. If the intersection is non-empty, this will be the set of traces that belong to more than one Petri net
Man sieht ja Eschweiler und auch Stolberg ,aber die Aue ist gar nicht gefilmt
All in all, nice course! :)
is it possible to have a reference book for these lectures?
are the slides available for download anywhere?
That's Great Professor
This is really fantastic content - thanks!
My heart goes out to everyone binge learning this before exams 😂
😂 thats me now!
Thank you, these lectures will come in very handy, as I'm starting my Master's thesis in the Process Improvement Explorer project at University of Tartu. I think I will dig more into sequence and episode mining, they seem very relevant for my topic.