Dear AnyLogic community,
I'm a novice AnyLogic user and hope someone of you can help me!
I have a simple process model, consisting of multiple sources, queue, seize, delay, release and sink (Process Model). The system I'm modelling is a server capacity problem. I have agents with differing service times and limited server capacity, the KPI I'm interested in is the number of customers that don't receive proper service when the resource pool is exhausted. Currently I'm letting customers timeout at the queue block when all resources are being used, but thats not an accurate representation of how the system performs in real life.
In reality a new incoming agent may preempt and terminate the service of an agent being served, more specifically the agent with the lowest service time remaining. I've seen the preemption options in the seize block, however I'm having trouble to define whose service should be terminated.
Would someone of you know how I can model the preemption of the agent with the lowest service time remaining while eliminating the queue entirely?
Many thanks for your time & help!