CYCLETIME

[impactdynamics]

A feature that I use a great deal in ithink is CYCLETIME. This function calculates how long it takes entities to flow through one or more sequential stock-flows on a FIFO basis. The flow where material enters the chain is 'time stamped' and then the CYCLETIME function is applied to outgoing flow to capture the time it spent in the system between these two flows.

It is critical for situations where controlling variables are dictated by leadtime, for example resources needed to complete an aerospace programme using Earned Value calculations.

I am very keen to convert models with this feature into Vensim. Is there an equivalent function and please point me to the documentation that tells me how to use it

Many thanks

Roger

[Administrator]

Hi Roger,

This has been asked before. See

http://www.ventanasystems.co.uk/forum/v ... .php?t=104

Tony.

[Edited on 13-2-2007 by Administrator]

[impactdynamics]

thanks Tony - yes I remember our discussion and your answer well - however during a more recent discussion with Lee I understood him to say that Vensim could do it.

I had hoped there was some new functionality that I had missed. Could you check with him to confirm I have misunderstood. I use this so frequently I am keen to pursue even if it requires more manual effort to get the same result.


Thanks

Roger

[bob@vensim.com]

Hi Roger,

The normal formulation for cycle time is

average residence time = level / outflow

This assumes that residence is exponentially distributed. And usualy it works very well.

If you truly have a FIFO process, and you need precision to that level of detail. then you will probably want to use either the QUEUE FIFO or the QUEUE FIFO ATTRIB to define the level - then you can get any info you want. If you only want this for cycle time, the the process involves a bunch of levels, you can simply have the QUEUE run parallel to do the accounting. Use QUEUE AGE OLDEST to get the cycle time of the current output.

[Edited on 2007-2-15 by bob@vensim.com]

[impactdynamics]

Many thanks Bob. I will follow this through

Best regards

Roger