Last Monday I did a lecture on Big Data and Process Mining for local government representatives of the Municipality of Eindhoven. The Uni. Eindhoven is home of process mining (see also: www.processmining.org) and the audience existed of people with a degree on Math, Statistics or other fields. They all had in common that they had a desire to better understand Big Data, Data analytics, Business Process Management and Process Mining. Since they are or were all working at the local government, I focussed my lecture on Process Mining within ‘small’ (governmental) organizations.
My assignment was to take people along in the practical side of process mining and to show them some cases. There were a lot of questions and there was also a lot of discussion. What should have been a 1,5 hour lecture started at 12:30 and ended at 17:30. A great afternoon with great input and remarks. With these people with all kinds of backgrounds the angles to look at Big Data and Process Mining were refreshing.
In the end we determined that for Process Mining to work in organizations these are 15 important considerations:
- Process Mining is a rather new field of science and expertise. It is still much driven by scientists. Translation to practical means is important and the input of people from practices is cardinal.
- Big Data is hot and therefore all adjacent fields of research seem to be hot as well. However Process Mining and Process Analytics are more than a hype. They are the next step in BPM and the practical side of Data Science.
- Digitalization and other developments such as IoT will create opportunities (and need) for Process Mining rapidly.
- Small organizations and local governments currently are ‘smelling’ and ‘tasting’, but it will take some time before they can start ‘cooking’ themselves. Collaboration and cooperation will be crucial for success.
- For Process Mining to work later, you need to make long term decisions on (target, system, enterprise) architecture now, in terms of event-logging, process supporting/enabling systems, Service oriented architecture, mash-ups and data gathering.
- To start with Process Mining; the Conformance or Extension method will probably work best and demand less effort and investments compared to Process Discovery.
- For Process Mining to work you should start with creating value-streams (step 1). High level chains that shows the business functions and added-value in your organization and the way this added-value is delivered through processes.
- You don’t have to model processes in detail since that is one of the big advantages of Process Mining (step 2.).
- However, there is big (change management and awareness) value in the workshops, brown-papers and interviews as well. Therefore you might want to consider to use these instruments after you have created process models from event-log data.
- For Process Mining to work you need to have knowledge of (or access to) data analytics.
- Be aware what your strategic (process) goals are (specially in fast changing environments).
- Uniformity and cost reduction may no longer be your biggest trigger. More and more organizations are aware that customization and personalization are becoming more important. Therefore inefficient processes that are customer focussing may no longer fall out of the bell-chart.
- Create a clear overview of costs for implementation and maintenance of the event-logs, data, algorithms, process models and the benefits in terms of optimization.
- Be aware of the seven big data V’s (http://dataconomy.com/seven-vs-big-data/).
- Take notice of compliance regulations from privacy and security legislation and (internal) principles. (http://www.theguardian.com/technology/2014/jun/20/little-privacy-in-the-age-of-big-data and http://www.computerweekly.com/feature/How-to-tackle-big-data-from-a-security-point-of-view)
In the end we concluded that Big Data will change. Data gathering will no longer be the goal. Big Data and Data gathering will be used for clear purposes such as real-time Process Management by Process Mining and Analytics. The developments are unstoppable and fastening rapidly. Organizations, also small ones (including governmental) had better be ready! Just like a lot of ‘traditional’ organizations missed the web-shop boat, you don’t want to mis the Process Mining and Data analytics boat!