Welding-scroller

A Day in the Life of a PhD Student: Wolfgang Koehler

In a fast changing world, lifelong learning becomes a necessity in many professions. I received my first Master’s degree in electrical engineering in Germany, back in 1991. During that time I got hooked on the subject of industrial automation, which has me moving around the globe ever since. For example I spent excessive time in North, Middle and South America as well as Australia, China and Russia. Needless to say, I spent a lot of time, away from my family, in hotel rooms. To keep my sanity, I decided in 2007 to start an online Master’s degree in Internet IT. Without the pressure of having to succeed, in order to get a job, I did exceptionally well and finished as the best student of my graduate year of 2011. In 2016 the urge to learn overcame me once again and I contacted my Master’s supervisor from 2011, who had moved on to a position at Coventry University. After talking about the research I was undertaking for my industrial customers, he suggested I write up a thesis proposal about mining industrial manufacturing processes, which was accepted late 2016.

The most memorable days of my part-time PhD student life, are the days that allow me to attend events with like-minded researchers. This includes lectures and conferences as well as even – yes – my PRP (Progress Review Panel) reviews. I enjoy when people ask me questions about my research. It gives me new angles and approaches. Something to think about. Reasons for sleepless nights.

To understand my excitement I’ll have to give you a bit of the boring part. For my research I developed, in the course of my regular work, software that automatically determines which datapoints of industrial equipment to monitor. Since these records are, due to many reasons, incomplete and flawed I needed to research how to correctly group the data together and how to determine if the data is complete. I spent the last 12 month 50+ hours a week working with the same dataset of about 500,000 records and I was sure I knew everything about it. I even had developed algorithms that can automatically spot deficiencies within the manufacturing processes.

During one of the lectures I attended, I got pointed to the discipline of Process Mining, which is a research topic for business processes. I read up on the subject and concluded that this could be a way to analyse my data, once I got the data cleaned. Therefore the Process Mining requirements on the data set became the guidelines for my research.

It just so happened, that once I reached the above described milestone in my research, the Process Mining Camp 2018 took place in Eindhoven. Going there I was doubtful that Process Mining actually would be of any benefit to my research. That feeling got even stronger while listening to the speakers and attending the workshops. After the second day I had chalked it off as a nice networking opportunity and a chance I had to get away from the daily grind. After all ‘I knew everything about my data’ and Process Mining didn’t offer me any tools to automatically discover anything in the data. For me it just was a tool to visualize data in a different way.

The day after however I decided to load my data into one of the Process Mining tools anyway, just to see whether the format and quality I had achieved conformed to the tools requirements. What followed was an unexpected surprise. The output of a, in my mind, straight forward process, was actually what is know in the community as a spaghetti process with many unexpected connections between events. For business processes this is a bad thing, because it means things have gone astray. For me, this was mind-blowing as all of a sudden there was a vast additional amount of information that I didn’t know existed.  Again another angle to view my problem. And I learned something else: Process Mining is not a tool. It’s a methodology about how to research any process. Something I have been doing all along by developing my own purpose build tools, with features exceeding those available for business processes. I guess I’m on the right track. What followed were two sleepless nights; a draft presentation for next years Process Mining Camp and a lot of motivation. What a thrill.

To conclude, I have to say that getting into the part-time PhD program, at Coventry University, for me was the right choice, because it gives me the freedom to really dig deep into subject matters of my interest without getting held back by corporate cost / benefit considerations. On top of it the whole experience is lots of fun for me and keeps my mind young. With the completion of the program I’m hoping to be able to get a position as a lecturer, possibly online, so I can stop traveling and spend more quality time with my family.

Comments

comments