Hello Sravan,
If the processors have got a capacity of one it is just a matter of the priorities you set at the processors for transport and process.
On the other hand if you have got a capacity greater than 1 it becomes tricky, because now it is a problem of the engine runtime and has nothing to do with logic. But you can force the system to use a special pass to method.
In this method you collect all transport tasksequences in the dispatcher, which you expect to happen for a cycle. Then you dispatch only the transport tasksequences, even if you get tasksequences to process items. Then you pass only the process tasks to the operator and you collect the next transport tasks in two different groups. The transport tasksequences to deliver to machine 2 belong to the first group. The transport tasksequences to transport to machine 1 belong to the other group. Maybe you can use another dispatcher to store the tasksequences at different places. If you want to pass the taskesequences of only one group you have to open the output of the dispatcher. The other dispatcher's output should be closed then.
The control logic depends on the variation of close and open outputs of the dispatchers and the right collection of taskesequences for a cycle. then you can prevent a mixture and choatic work on tasksequences, which will result in a block of the model.
Jörg
|
