modeling production line

[rezasoltani]

Hello Incontrol community

I have a production line of a two-step processing of the healthy parts of the distribution of normal and defective parts will follow a Poisson distribution to model the problem, how do I do?
Then, if all the parts are produced in the queue, how can I control the quality of all the products I sample to study?
Regards,reza soltani

[Nick]

Hi Rezasoltani,

Please see the attached example model. The Two-step example production line consists of a Source, a Queue, two Servers and two Sinks.

Source - Trigger on exit :
Each product that leaves the source gets a label: 73% of the products are marked as 'healthy'.

Code: Select all

Label([normalordefective],i) := Bernoulli(73,1,2) {1 is 'healthy'; 2 is defective}

Queue - Trigger on Entry:
The icon is changed based on the label value to clearly see the difference between the healthy and defective parts.

Code: Select all

If(Label([normalordefective],i) = 1,
   icon(i):= 44,
   icon(i):= 22 {this value can be found in the Resource manager --> |> Main Menu |> Tools |> Resourcemanager}
)

Server - Cycle time
Healthy parts follow a normal distribution with a mean of 100 seconds and a standarddeviation of 20. Defective parts follow a Poisson distribution with a mean of 60 seconds.

Code: Select all

If(
 Label([normalordefective], last(c)) = 1, 
 normal(100,20),
 Poisson(mins(1))
)  

Server - Send to
The healthy parts are sent to a different Sink than the defective parts.

Code: Select all

if(Label([normalordefective], first(c)) = 1, 1, 2)

You can adjust the Bernoulli formula on the Source atom to change the quality level (percentage defective).

Regards, Nick

[rezasoltani]

Hi Nick,

thank you for your detailed reply.

rezasoltani