I am relatively a new user in discrete event simulation and my background is completely different to production engineering. I am a mechanical engineer but I always felt that I am an all rounder, given the generic bachelor normally done in undergrad degrees.
For a research I am working on, I would like to implement a series (standard) production of mechanical excavators (keep in mind output pace is based on weeks work in progress with 2 week intervals/progress stages, volumes are low and the chassis are externally sourced. After 'X' amount of weeks, the product is complete). Basically, a lot of assembly is involved by different teams/trade workers and some minor re-works/modifications. A planning is in place of who does what and for how long (at least on paper).
Currently the series production happens on a shop floor but lean production is far from reality and work on the product starts where space is available with everything shifting around.
I would like to show that, if we work in dedicated space boxes, and shift to a pull system, with clear WIP stages, not only our WIP falls but also our throughput time falls, which would mean, unknown future orders can be accommodated with a clear delivery timeframe.
There is a debate whether the product should move or the teams should move, given the weight of the product and while in the beginning of the discussion, moves were planned at each two-week interval to clearly signal a pull system, consensus has been reached that heavy lift moves should be limited to two times, with first move occurring in one place and last move bringing the product closer to the exit (as the products gets heavier and wider), before being transported out. Any for/against opinions?
It has also been difficult for me to convince that one piece flow brings a lot of advantages in particular for lean principles and a pull approach. Any for/against opinions?
If anyone is still intrigued by the problem and I haven't got you bored and would like to share some Jaamsim tips on how to proceed (whether its possible or not and if an example on this scale is available) as well as some theoretical background to the more qualitative questions.
Thanks in advance for your help and insight that you would like to share.
Kind Regards,
Mark Azzopardi