It’s funny how some things generate more interest than others.
I was going through my twitter feed last week and was intrigued by a tweet about the Process Mining Manifesto that was released recently by the IEEE.
Process mining is a topic that been around for some time.
I’ve heard of it, but never personally see it work.
It’s an interesting topic, well related to BPM. So to overcome my general ignorance on the subject I did a bit of research.
Now if I understand correctly, process mining is all about discovering and improving processes by extracting data from event and transaction logs.
Process mining enables to extract the process model from the data. Discovering the real process and actual business rules. Documenting your factual ‘as is’ processes.
There are some process mining tools in the market:
Fourspark , Fujitsu, Futura Process Intelligence, Open Connect, Perceptive Software, Process Sphere, QPR, Software AG, StereoLOGIC, TU/e, UPC, Verint
(Apologies to any other process mining vendors that I missed)
Reading the vendor’s publications, there is a bit of confusion in the messaging.
Some call it “process discovery”, some explain its value in improving the process by analysing the real data (although you could probably do most of this using your standard reporting tools)
The value of process mining, it seems, is in basing the objective analysis of real data, not the subjective views of different stakeholders (each 2 stakeholders have 3 views on how the process currently runs and 4 views on how it should run..) Building the process model based on objective data, not the subjective views of humans.
Anyway, back to my digging…
Adam Deane 
Hi Adam:
Model discovery is only one of the dimensions process mining can help to understand what is really going on a process. The scope of analysis can be enlarged if the process is executed out of the company boundaries across companies, or other stakeholders.
There are other disciplines where process mining can truly help to improve processes: socialization, reasoning and performance. I understand that some practitioners can be headed to the modeling because they want to understand what is happening, but process mining is much more than that.
Regarding socialization process mining can identify how people interact and behave who they rely on, if they are highly coupled or not, who they share information to solve a particular problem or if there are bottlenecks regarding to much inbound to low outbound, if resources are balanced to collaborate and if escalation occurs naturally among people when peak / unpredictable events happens. Who perform best? What are the right teams to assemble regarding past experience? This kind of analysis is very important under social execution paradigm and it cannot be express on a process map. there is no process map that can show this kind of interactions.
Regarding reasoning and execution patterns is clear how people make decisions: things like batch processing (that increases cycle time), how they take decisions (to whom they distribute work at if that distribution is balanced), what are the task they pick first to execute (are the best ones? that argument that knowledge workers know what to do every time does not holds most of the times).
Performance: an amazing new area. Why we are performing poorly? What are the causes? Why it takes so long to do something that exceeds the business rule?
One last word: conformance. With experience as an auditor I realized that in middle 90’s with systems spreading across companies, was getting much more difficult to get the facts. Today is almost human possible. Try to figure it out if your company is paying supplier invoices of things that never entered in the company, on lets say an automotive plant, manually.
Process Mining is about speed, yes speed and accuracy. It leaves far behind old school approaches of process improvement. It’s like an addition, once you tasted you figure it out your approaches are a thing of the past.
By: Alberto Manuel on 15/12/2011
at 9:12 am
Hi Adam (again)
This is just in: Ana Airports uses Process Mining to improve Changes Order Process, more details (in Portuguese) follow bellow link
http://www.process-sphere.pt/
A copy of the case study is available, just send us an e-mail: http://goo.gl/oDBZh
Regards
By: Alberto Manuel on 15/12/2011
at 2:30 pm
Adam
Process Discovery is also pretty useful in an unstructured process where you can advise the “Next Best Action” based on analysis of previous cases (no pun intended) and suggesting a course of action or routing logic.
By: Pieter van Schalkwyk on 16/12/2011
at 1:48 am
[…] started making public the case study (if you want a copy send an e-mail) a lot of noise part 1 and part 2 appeared around process mining. Good I thought. A lot of discussion, fear, anger and anxiety means […]
By: A Process Mining Project « End to End BPM on 18/12/2011
at 11:56 pm
Hi Adam,
I enjoy and learn a lot from your blog posts, and follow you on Twitter as well.
Just wish to add Fluxicon to your list of vendors.
http://fluxicon.com
They have an educational blog about process mining, typified by the recent “7 Objections Against Process Mining”
http://fluxicon.com/blog/2012/01/7-typical-objections-against-process-mining/
I am interested in application of process mining to electronic health records. If anyone else is too, feel free to contact me at
http://chuckwebster.com/ehr-workflow-contact
or tweet me on Twitter
https://twitter.com/EHRworkflow
Looking forward to your further posts and tweets.
Regards,
Chuck
By: Chuck Webster, MD, MSIE, MSIS on 30/01/2012
at 12:20 pm
Thanks Chuck,
I’ve asked Anne for a company description, and I’ll add Fluxicon to the BPM vendor list
I’ve also started doing a bit of home on EMR
Cheers,
Adam
By: Adam Deane on 31/01/2012
at 10:27 am