Approach / method for analyzing a process (production)

J

Good morning people!

I try to somehow methodically pursue / investigate everything in my Master thesis. Overall, the goal is to optimize a sub-process in production, especially at the organizational / employee level.

For this I have to determine the actual state of course in the actual part.
So how does the process go?
What's the productivity, wins, number of pieces, OEE.

For this, however, I would need some system / approach / method, so I also cover everything relevant. (All environmental influences, all relevant indicators, etc.)

Do you have an idea for this?

Best regards
!

Avatar for rewt

I did something similar for my masters' dissertation. Where I optimised a unit operation in a manufacturing plant during a placement year (meant I did masters dissertation before I had even started). I will give you some advice from my work and hopefully, some of it might be relevance.

You need to decide your boundary conditions - where does it start, where does it end, what can you vary, what external conditions are fixed? If you look at too many variables it can very complicated fast, so you want to focus only on the most important aspects. If you get lots of results you can look at more variables and extreme conditions but set it at a manageable level to start with. Talk about processes upstream and downstream and how they affect the project but just assume they are fixed and produce/require fixed conditions. It makes it so much easier if you can initially say that the previous process is always working.

Define your conditions. For my project, I decided that profit was the absolute indicator of performance and everything else was constraints. Ie my constraints were; keep waste below 25%, not to increase manpower required, produce consistent products. So if all the constraints were met I was only looking at one value for comparison.

Methodology! This varies between processes but I bet there isn't much literature on your process. So create your own literature and vary everything from there. I knew 3 settings always worked, I had those settings incredibly well documented with all the outputs and costs recorded for nearly all external conditions. Therefore whatever happened or whatever method I used, I had something I could compare to, meaning most of my data was useful.

Finally, talk with the operators/technicians/users. They usually have a very different understanding of the process than office staff that can be invaluable.

Hope this helps and good luck,
rewt

54594