On Process Mining and ACM – Keith Swenson
Process mining is especially important to Adaptive Case Management (ACM). When it is impossible to predict the process ahead of time, knowledge workers need to move forward without a predefined process. Process mining promises a way to retrospectively see what the pattern of work turned out to be, even though it was planned as the work was being one. Process mining can give you metrics about the process which can be used to evaluate how well the work is being done. Even though the process is not automated, you can get many of the benefits normally associated with BPM.
On BPM and Rules – Nicholas Kitson
My prediction for the future is that more and more applicaitons will seek to abstract business logic in separate tools, where they can be accesses, created updated and amended as necessary in natural (or close approximaiton of ) language like oracles OIPA solution. Rules and process management, will become like database management is today, as nautural to us as mom’s apple pie
On BPM and Agile – Tim Huenemann
Some aspects of Agile are good tenets for BPM efforts: close collaboration between the business and IT; smaller, more achievable implementations that show business value sooner rather than later. Regardless if you use a true Agile methodology, Agile concepts can be a good influence on BPM efforts.
Adam Deane 
Adam, thanks for your always very interesting set of BPM links. The ones you chose this week drive me to a comment.
Keith Swenson: Process Mining is an oxymoron. If something is already a process there is no sense in mining it. One can mine generic user interactions in social collaobration or email or in case management and see if they are being used to perform repeatable processes. But as Keith pointed out, most knowledge work doesn’t repeat, but yes there may be some sub-processes hidden in there. If process mining makes sense for ACM then it is a form of BPM and it should support process flow capabilities too, which is however something that Keith disputes. A process is something that happened or how it could happen but never how it IS or WILL BE.
Nicholas Kitson: A mix and match of independent rule engine products and BPMS requires substantial change management and deployment efforts and is totally lost without embedded master data management. These projects completely kill compan agility because of it. It is irrational to recommend it.
Tim Huenemann: The necessary BPMS implementation efforts are in fact the antithesis to AGILE. As Tim properly points out: The two don’t go together. So how can anyone still claim that BPM actually increases the agility of a business if not even an agile process creation approach is possible.
All this leaves me wondering: How much more irrational has the arguing for BPMS to become? It reminds of the CEO of AIG, who said two month before the 2008 crash that he couldn’t see how his company could loose money on the derivative insurances.
Thanks again, and keep it up! Max
By: Max J. Pucher on 10/12/2011
at 10:48 am
Adam, an additional note on process mining is necessary. As the UK management thinker John Seddon says: ‘Industrialization of service work is a myth. Anytime you reduce variety (i.e. through BPMS flowcharts) you drive up the cost by failure demand. What has to be managed is the value demand, which means to not standardize but focus on customer outcome.’
Which means: What would you want to learn from mining except how to standardize and thus how to drive up cost?
Thanks again, Max
By: Max J. Pucher on 10/12/2011
at 10:55 am
Hi Max:
Sorry to disagree 100% with the your reasoning about Process Mining. Process Mining is not only for structured processes. Process mining can be used to unstructured processes. The fact that Process Mining can be used (this one of typical dimensions) for process model discovery does not mean that is applies only to the structured world.
The argument you use that unstructured work will never repeat again is false. Unstructured work follows most of the times reasoning patterns (reasoning not step sequence) that are closed linked the way a person thinks and execute, this is so true and it becomes real when a person acquainted with corporate culture and with it’s workforce can only understand when is looking to the event data. These patterns are highly important to improve articulation and supporting social arrangements, thus benefits ACM.
Even in situations where there are no references, no patterns, Process Mining can shows you information (it can be what ever you want) that can help you understand to nurture and foster ACM, how people interact, what kind of information they share, what channels they use, when people stop executing because something was missing. These kind of facts, yes facts, not assumptions like what happens with 99,99% of the times can truly help managers, participants to find out how to improve the way they work and what to do they work.
Sorry Max but your reasoning shows that you never experienced 200% a Process Mining Project.
Regards.
By: Alberto Manuel on 10/12/2011
at 2:41 pm
With respect to process mining and ACM, Keith and Alberto are right on the money here. Process Mining makes more sense and provides more insight the more unstructured and less predefined your processes are, in fact.
If you want to read (a whole lot) more about this topic, here is my PhD thesis about process mining in flexible environments: http://www.processmining.org/blogs/pub2009/process_mining_in_flexible_environments.
Best,
Christian
By: Christian W. Günther (@cwg) on 10/12/2011
at 5:53 pm
It seems that the only people who don’t ‘get’ process mining are the ones that have never looked at it in any detail. There is a lot of good literature on process mining including the recent manifesto –
Click to access fetch.php
By: George on 16/12/2011
at 9:31 am