Infinite recursion when it should not be. 1

[Degine]

I'm trying to simulate cycle times as a function of work released at a station.

Please see attached worksheet.

Mathcad 14

http://files.engineering.com/getfil...-8bbd-4755-bc66-bbc536ff534&file=Example.xmcd

Mathcad 11

http://files.engineering.com/getfil...-6705-4e79-a83d-583bae91b224&file=Example.mcd

The problem is, for the CT(w), it requires CT(w-1).

While understandable this leads to recursion, CT(1) is very easily calculated since some parts reduce to 0 for example:
w-1/CT(w-1)=0 for w=1

No need to know CT(w-1) as 0 divided by anything will always be 0.

So how can I avoid Mathcad worrying about 'infinite recursion' / overflow?

Note: w=1,2...32 at most. So definitely not infinite.

 

[IRstuff]

There are several approaches. Attached is the indexed variable approach. Your second term contains redundancies.

http://files.engineering.com/getfil...f2-980c-b9c3fa4749df&file=Example_indexed.mcd

TTFN

FAQ731-376

 

[Degine]

Wow great thanks, I never knew about the second kind of subscript. (<.> vs <[>) It seems like a great solution for many things.

However think I've oversimplified the expression, please see attached files with expressions to describe multi-station lines. The expression is completely wrong, but describes what I'm trying to do. (Unfortunately, the second expression no longer contains redundancies.)

I guess a solution would be to separate all the CT_j(w) and sum them up later, i.e. move everything to different expressions like you might calculate everything in different cells in Excel.
But then each expression refers to another expression that has not yet been calculated -> every expression should be solved at the same time. Another form of problematic recursion I guess.

Note: CT_j should be a function of w rather than j. Didn't manage to formulate that in Mathcad.

Mathcad 11

http://files.engineering.com/getfil...bdbd-4de1-97fb-20cd16e145d3&file=Example2.mcd

Mathcad 14

http://files.engineering.com/getfil...781-4c4a-9065-84fe7bea7b21&file=Example2.xmcd

 

[IRstuff]

Just file into your brain that an indexed variable is NOT a vector; that will save you loads of grief later.

Engineering.com appears to be non-responsive at the moment. I'll post the file later

TTFN

FAQ731-376

 

[IRstuff]

Let's try again

http://files.engineering.com/getfil...6-a755-1a95e318d605&file=Example2_indexed.mcd

TTFN

FAQ731-376

 

[Degine]

Thank you very much, this has been of great help.