I know I promised a relatively quick update cycle to my blog – but I am having trouble with my Clone Management Interface (the interface to the three versions of myself that never sleep and are eagerly beavering away at all hours of the day).
BPMN Process Modeling
The BPM Process Modeling Fundamentals courses in London, LA and Washington DC were all pretty much sold out (we couldn’t fit any more in the room). The course for Sydney on June 7th and 8th is also virtually full already (register here). Perhaps we will run another one at the end of the following week before I head back to the UK. Not surprisingly, the Ensuring BPM Project Success course is less well attended (as it is designed for a different audience).
But one thing has struck me as I reflect on the BPMN aspect of the training. A lot of people seem to expect a lot more methodology out of BPMN. In the end, BPMN itself is method independent – that it allows companies, individuals and tool vendors can apply the Notation to any number of methods. Moreover, adding simulation into the mix, while it may be useful in some situations, is not part of BPMN. It belongs in some extended method (which is relatively poorly supported by the attributes of BPMN icons).
Indeed, one could argue that if your challenge is to understand the process (as it is in the early stages of most BPM initiatives), applying simulation to the mix is a complete waste of time as it involves a lot of effort in gathering data about the process and validating that the distributions and estimates of time/resource usage is correctly applied in the model. Moreover, many of these (simulation) models are constructed with the perspective of proving the benefits of the approach (proving to management the benefits in terms of money saved or revenue generated). As such, the modeler is often (unconsciously) constructing a model that is already pre-disposed to supporting the aim … a model that buries the assumptions rather than surfacing them.
In that upstream activity (of understanding the domain to identify the 20% of functionality that will deliver 80% of the value), what is needed is the ability to compare and contrast different perspectives on the process … looking not just at the orchestration (ordered sequence of activities), but also the choreography (the sets of interactions between the roles), and the boundary conditions of the chunks. Because by understanding the process better, the end-users can really identify the areas that will make a difference.
It is in the downstream implementation of that defined scope that BPMN comes into its own as an implementation oriented graphical language. The point is that applying detailed BPMN modeling and simulation too early in the modeling endeavour is inappropriate.
BPM Focus Web Site
Well suffice to say that it is about to go through a major overhaul. We have been busy beavering away implementing a commercial BPM Suite under the covers that will make the whole experience an order of magnitude easier to handle. All those people who are currently on the BPM Focus mailing list will be invited to update their profile, such that we can more effectively target the messages and communication that we send. Moreover, this will facilitate the rapid introduction of a whole range of new services that have been in development for some time. Instead of worrying about the implementation detail, a new service becomes nothing more than a set of robust (BPMN) models. But more on that later.