Cancel
Showing results for 
Search instead for 
Did you mean: 

Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Valued Contributor
Valued Contributor

Hello dear community,

I have a problem which I can't solve.

I would like to change the dismantling table of the station "AufteilungProben" with two distributions when a MU is entered. For this I have the following method:

Code:


var x, y, z: integer

x := z_demp(1,.Modelle.Probenvorbereitung.Sägen.Anzahl_Proben_Verteilung)

for z := 1 to .modelle.probenvorbereitung.sägen.AufteilungProben.DemontageListe.ydim
.modelle.probenvorbereitung.sägen.AufteilungProben.DemontageListe[2,z] := 0
next

for z := 1 to x
y := z_demp(1,.Modelle.Probenvorbereitung.Sägen.Zuschnittsverteilung)
.modelle.probenvorbereitung.sägen.AufteilungProben.DemontageListe[2,y] := .modelle.probenvorbereitung.sägen.AufteilungProben.DemontageListe[2,y] + 1
next

 

You can see the tables as a screenshot in the appendix.

Unfortunately, the result are about 10% MU's "Aufweitversuch" and all other MU's are generated by approx. 5%.

Does anyone see my mistake?
Have I the distribution tables for z_demp set up correctly?

The helpdesk did not help me, so I would be grateful for any help.

 

Best regards,
Henner

_________________________________________________________________________________

 

Hallo liebe Community,

 

ich habe ein Problem an dem ich verzweifle.

Ich möchte die Demontagetabelle der Station "AufteilungProben" mit zwei Verteilungen beim Eintreten eines BE's ändern. Dazu habe ich folgende Methode:

 

Code:
var x, y, z: integer

x := z_demp(1,.Modelle.Probenvorbereitung.Sägen.Anzahl_Proben_Verteilung)

for z := 1 to .modelle.probenvorbereitung.sägen.AufteilungProben.DemontageListe.ydim
.modelle.probenvorbereitung.sägen.AufteilungProben.DemontageListe[2,z] := 0
next

for z := 1 to x
y := z_demp(1,.Modelle.Probenvorbereitung.Sägen.Zuschnittsverteilung)
.modelle.probenvorbereitung.sägen.AufteilungProben.DemontageListe[2,y] := .modelle.probenvorbereitung.sägen.AufteilungProben.DemontageListe[2,y] + 1
next


Die Tabellen seht ihr als Screenshot im Anhang.

Leider werden im Ergebnis gut 10% des BE's Aufweitversuch und ansonsten alle BE's ca. 5% erzeugt.

Erkennt jemand so einen Fehler?
Habe ich die Verteilungstabellen für z_demp richtig aufgestellt?

Die Hilfe hat mir nicht geholfen, ich wäre daher für jede Hilfe dankbar.

 

Beste Grüße,
Henner

 

8 REPLIES

Re: Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Gears Esteemed Contributor Gears Esteemed Contributor
Gears Esteemed Contributor

können Sie Ihr Modell (oder einen Auszug daraus) posten? So ist das etwas schwierig zu beurteilen.

Steffen Bangsow
freelance simulation specialist  
web: www.bangsow.eu
mail: steffen@bangsow.net

Re: Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Valued Contributor
Valued Contributor

Natürlich. Es geht um das Netzwerk .modelle.probenvorbereitung.sägen    

Re: Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Siemens Phenom Siemens Phenom
Siemens Phenom

Hi Henner,

your question contains an interesting problem.

Your changed correctly the dismantle list in the entrance control.

The variable x is a random integer number 1 … 50. It is the number of generated movable element for each incoming part.

The variable y is a random integer number 1 ... 17, which describes the exiting movable elements. Therefore the sum of column 2 of the dismantle list is x.

What is the resulting frequency of one of the 17 parts?

You want to validate your model, but what frequency do you expect?

 

I changed the entrance control a little bit and saved your model. The attached model can be opened by Plant Simulation 13.0

 

This is the entrance control:

 

var x:integer := z_demp(1, Anzahl_Proben_Verteilung)

var DismantleList:table := AufteilungProben.DemontageListe
for var z := 1 to DismantleList.ydim
 DismantleList[2,z] := 0
next

for var z := 1 to x
 var y:integer := z_demp(1, Zuschnittsverteilung)
 DismantleList[2,y] := DismantleList[2,y] + 1
next

 

Regards, Peter

Re: Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Valued Contributor
Valued Contributor

Hello Peter,

 

thanks for the simplification of my entrance control.

 

For the validation of the dismantle process I count the outgoing MU's now in a table.  As you see in the attached ScreenShot, the MU "Aufweitversuch" is (as it should be) the most frequent. But the other MU's are likely as frequent as each other.

 

I would expect the same distribution as in the table "Zuschnittsverteilung".

 

I really do not understand why it isn't, because the method "Zuschnitt_festlegen" always uses the distribution from the table "Zuschnittsverteilung" to decide which MU should be generated.

 

Any idea how this distribution could result?

 

Greets,

Henner

Re: Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Valued Contributor
Valued Contributor

I got the Problem! Smiley Very Happy

 

It seems like the dismantle station do not except a number of 0 in the dismantletable. As you see in the attached screenshot the dismantle station generates at least 1 MU of every kind.

 

Is it maybe just not possible to generate 0 MU's?

 

Best regards,

Henner

Re: Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Valued Contributor
Valued Contributor

Now I just erase the rows in which the number of MU’s is 0 and it works perfectly.

 

var x:integer := z_demp(1, Anzahl_Proben_Verteilung)
var a: integer
var DismantleList:table := AufteilungProben.DemontageListe

for a := 1 to 17
	DismantleList[1,a] := Demontagetab_leer[1,a]
	DismantleList[2,a] := 0
	DismantleList[3,a] := 1
next

for var z := 1 to DismantleList.ydim
 DismantleList[2,z] := 0
next

for var z := 1 to x
 var y:integer := z_demp(1, Zuschnittsverteilung)
 DismantleList[2,y] := DismantleList[2,y] + 1
next


a := 17

while a /= 1
	if DismantleList[2,a] = 0
		DismantleList.cutRow(a)
	end
	a:=a-1
end

Re: Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Siemens Phenom Siemens Phenom
Siemens Phenom

Great,

I did not expect this behavior.

Therefore I improved the entrance control.

The idea is simple: I delete the row with 0 in the second column.

Please see the attached model.

Re: Probability distributions for dismantling process/Wahrscheinlichkeitsverteilung bei Demontage

Valued Contributor
Valued Contributor

Once again it looks better than my solution. I'll take it Smiley Very Happy

 

Thank you very much and best regards,

Henner