Remco Bunder and Jacco Vogelsang are two pioneering information analysts at Nederlandse Spoorwegen (Dutch Railway). Remco and Jacco had joined camp in 2016 and got inspired by the talk of Paul Kooij. As innovators, they were eager to get started and find a good use case to show the value of process mining within their organisation. But where to start and how to do it?

They decided to simply apply process mining on every dataset they could put their hands on. By doing this, they gained knowledge and experience and started making unexpected observations. One of the first experiments was to track the OV-Bike (a bike rental service) from ‘rented’ to ‘return’. They saw that a lot of the bikes seemed to be reported as stolen upon return. This was unexpected and further investigation revealed that many of the bikes got reported as stolen because the ‘report stolen’ button was too close to the ‘return’ button. This minor mistake lead to the ordering of too many new bikes.

Eager to find more, they started looking at how the lockers at the stations were being used. Especially if lockers are abandoned for more than 94 hours, it was required to check the locker and empty the content out of the locker. With process mining they were able to show that it would be better to wait an additional 48 hours before emptying the locker.

They continued to analyse even more complex processes such as the reporting and resolution of broken windows, escalators, elevators, etc. by benchmarking the quality and efficiency of resolving these failures for different channels, stations, and vendors. The experiments from Remco and Jacco are a fantastic example of how enthusiasm and persistence can help you grow from a process mining novice to achieving great results within just 1 year.