Stuck Model

Hi FlexSim Community!

Good day!

As I ran my model, it somehow gets stuck before finishing the desired end time. Can anyone help me figure out what's causing this? I have attached my model.

Thank you!

Best,
Winwin

Your first problem is the MTBF and MTTR fields in Unplanned Delays.

You are trying to use item to lookup the time in the table, but item doesn't exist in those fields so you are getting 0 as your MTBF and MTTR values.

You could use first(involved) instead of item, but then you'll run into the same issue if the involved object (processor) doesn't have an item in it when the MTBF or MTTR functions are called (hence why you don't have a reference to the item).


Once you fix that, your simulation will still run really slowly because you are applying the Speed Delay MTBF/MTTR object to a set of states, but you left the Accuracy at 100 and the Range Cutoff at 0. Thus, it will continuously be creating a bunch of events for very small values checking the states as it tries to time the breakdowns perfectly.

You can read more about the Accuracy and Range Cutoff in the MTBF/MTTR page of the User Manual.

I'm not sure what you are trying to do with this Speed Delay, but there is probably a better way to model what you are trying to simulate than what you are doing with this MTBF/MTTR object having such small values for MTBF and MTTR.

Hi Phil,

Good day!

Thank you so much for your reply.

Indeed, the model ran very slowly despite increasing the run speed.

For the Speed Delays, these are minor delays in the machines like jams and will be attended to by the Operator per Machine. The Unplanned Delays are major stoppages that will be attended to by one technician supporting the three machines. However, the MTBF and MTTR for the Unplanned Delays will differ depending on the item type. Both delays will only occur when the lot is being processed. I could assume that at least 70% of the time, machines are under the "processing" state.

Can you help me out more?

Thanks a lot!

Best,
Winwin

If the jams are attended to by the operator that is currently working on the machine, the simplest way to model them would to just include them in the processing time.

The level of fidelity that you need to simulate depends on what questions you are trying to answer with the simulation model. If the goal of the simulation isn't directly related to how these minor delays are affecting things, it is probably best to abstract them away into the processing time. If the goal of the simulation is to understand how these little delays are affecting things, then you might want to change the accuracy to 70 and the cutoff to a value that is a small portion of the downtime, such as 0.001 for your 0.1 downtime.