The next level: pursuing a PhD

I’m back!

After my final presentation and master thesis round-up I took a one-month holiday. Last Monday the holiday ended and I started my job: PhD candidate.

Not much changed actually: I sit in the same office at the same desk, I still travel more than 2,5 hours (2 times 1:15) each day to go to work and home again, I still like my colleagues and I still work on process mining (and even on XESMa a little bit).
There are some differences of course: I’m not a student any more but a ‘real’ employee with all kinds of things attached (from a mail box to all kinds of administrative things). Another thing I already notice is that much more different things come ‘my way’. During a master project you’re mainly working on that. As a PhD candidate you have your assigned project (see below) but you’re also supposed to help in education (but not too much, although I think I’ll like it), set up your own education plan (which courses do you need/want to follow) and many more fun things that come your way. Another benefit is that you belong ‘more’ to the group. Instead of only 6 months you’re now in this group for 4 years (at the very least). This means that other people invest (even) more time and have (even) more interest in you and vice versa.
Although I never though that I would pursue a PhD it does feel good, this first week at least.

The project: CoSeLoG

Update (20-5-2010): There is now an official CoSeLoG website!

So, the project, the main reason for me to start this PhD. If you don’t like the project you’re on, you won’t survive the 4 years.

The goal of the project, in brief, is to implement configurable process models in a SaaS (Software as a Service) environment to support municipalities in executing their processes.

The idea is that processes within municipalities are much alike. Registering an unborn child is similar in most municipalities. By applying process mining, configurable process models should be discovered. These models should describe the process executed in each of the municipalities. The next step is to implement these models in a SaaS system and provide formal semantics for execution. Because the models are configurable, each municipality will execute their own variant of the process. Furthermore, during execution the correct data should be recorded to enable the application of process mining on the SaaS system. This will enable municipalities to compare themselves with others and improve their process execution.
More information about this project can be found at the links at the end of this post.

One of the main advantages of this project is that it is very practical. 2 IT companies cooperate: D!MPACT, a Dutch IT cooperative for and by 20 municipalities and Pallas Athena, the well known creators of BPM|One used by +/-250 of the 440 Dutch municipalities.
Furthermore, 10 municipalities support the project, providing case data to use. These municipalities are Bergeijk, Bladel, Coevorden, Eersel, Emmen, Gemert-Bakel, Hellendoorn, Noordoostpolder, Reusel de Mierden and Zwolle. Half of these municipalities are located around Eindhoven. The other half however are located in the north of the Netherlands so I’m prepared for some long train rides. 🙂 Not all municipalities are a member of D!MPACT.
And, also very important, another PhD candidate will join the project in a few weeks. Together we try to realize the goals mentioned before. The other candidate will focus more on the practical implementation of the process models in the SaaS environment. My goal is to discover the process models using process mining, generate a configurable process model from these models (using a process mining algorithm to be developed) and enable the application of process mining in the resulting SaaS environment.

Overall I’m very pleased with the content of the project. Although configurable process models are new for me they are very interesting. Furthermore, the main goal of the project will be applicable to many companies too. Think for instance about the salary payment process of an international company, this could be captured in a configurable process model… So the future after my PhD also looks promising.

That’s it for now. Expect some topics about configurable process model in the future, although I’m planning to continue development on XESMa.




Update (20-5-2010): There is now an official CoSeLoG website!

(Google Cache) FILLED vacancy on TU/e website


Official announcement at

Official announcement at

News announcement at TU/e website

News announcement at D!MPACT website

Computable! article