Process Mining Camp: Fireside Chat with Bram Vanschoenwinkel
Are you as excited about Process Mining Camp as we are? As a warm-up, we asked some of the speakers for an up-front interview.
Today, you can read the interview with Bram Vanschoenwinkel, a Business Architect at AE. With several projects in payroll accounting, public administration, and postal services, Bram is one of the process mining veterans. At camp, he will share tips, tricks, and points of attention, problems he encountered, and lessons learned.
Interview with Bram
Anne: Can you still remember where and when you first heard about process mining? What exactly caught your attention and fascinated you about the topic?
Bram: I don’t remember exactly when I first heard about process mining, but it must have been somewhere late 2008 or early 2009. At that time I was working for a Belgian management consultancy firm with a strong methodological focus on facts, figures and measurable benefits. In that context, we were always looking for new techniques with a quantitative approach to process analysis. Indeed this is one of the strong points of process mining and thus the right combination of keywords inevitably led Google to direct me to a number of papers by Prof. dr. Wil van der Aalst (and others from the same group at the TU/e), ProM, and the processmining.org website.
Since I have a backgorund in Machine Learning and Data Mining, I was immediatly fascinted about the topic and it also met our requirements in the search for innovative, quantitative process analysis techniques. Particulary the promise to discover AS IS process models very fast and based on facts rather than opinions caught my attention. It looked to me like such discovered process models could serve as the ideal starting point for change, i.e. a basis for further analysis and reflection with the business.
Anne: It’s interesting that you come from a machine learning and data mining background. Sometimes, people are confused about the relationship between process mining and data mining. Do you see process mining as a special data mining technique, or how else would you explain the difference to someone who is new?
Bram: Indeed, for me process mining is a particular form of data mining. Just like text mining and web mining are particular forms of data mining, where the same (or at least similar) techniques are used but specifically tailored to text/web data and corresponding applications.
For me, the main difference is in the representation and preprocessing of the data as well as in the representation and visualisation of the output. Of course, not all techniques that are used will be the same or similar and, therefore, another way to look at this is by considering a simple but generally accepted definition of data mining: “Data Mining is the nontrivial extraction of implicit, previously unknown, and potentially useful information from data.” And isn’t that exactly what process mining is doing?
Anne: Right! Since you said that you have worked with multiple analysis techniques techniques in the past, I would be curious which other techniques you are currently using next to process mining?
Bram: In a more clasical approach to process analysis I typically start by a number of interviews (collect information) and workshops (validate/complete/correct collected information) to get insight into the processes at stake, i.e. both the process flow as well as all relevant information related to the process.
Next, depending on the objectives of the analysis, I use different techniques to analyse the process. For example, Six Sigma to reduce the number of errors or throughput time and improve the stability/predictability of the process, Lean to detect and eliminate waste in the process, workload measurement to optimize resource allocation, time-driven activity based costing, simulation etc. In fact I also do these kind of things when I am using process mining. The big difference is that with process mining I can start from a solid basis for my interviews and workshops and I can easily set up different ‘experiments’ to further analyse the process with one of the analysis techniques I mentioned before.
Before I knew process mining, I also used information from the applications that support the process. But with process mining it becomes easier to fully exploit this information and focus on what is really important: Actually improving the process. So for me process mining is not a standalone technique, but complementary to other techniques and a starting point for change.
Anne: Thanks a lot, Bram, I can’t wait to hear more from you at camp on 4 June!
Would you like to hear more from Bram about his experiences? Are you interested in sharing first-hand knowledge with fellow process miners? Register now to reserve your seat at Process Mining Camp on 4 June in Eindhoven. Tickets are free, but they are limited, and they are going fast…