– the workload (arrival rate) la for each node a (note that, provided that no
resources are overloaded, the throughput for each node is equal to its arrival
rate);
– the processing time Ta and the response time Ra, for each behaviour element or
service;
– the utilisation Ur, for each resource r