Given:
System type: M/D/1 queue, deterministic distribution
Arrival rate: \(\lambda=1\) Service rate: \(\mu=1/0.9\)
Find:
\(W_q\)
\(W\)
\(L_q\)
\(L\)
\(\rho\)
Mean arrival rate:
\[\lambda=1\]
Mean service rate:
\[\mu=\frac{1}{meanservicetime}\]
\[\mu=1/0.9\]
Server utilization:
\[\rho=\frac{\lambda}{\mu}\]
\[\rho=\frac{1}{1/0.9}=0.9\]
Customers in queue:
\[L_q = (1/2)*\frac{\rho^2}{1 - \rho}\]
\[L_q = (1/2) * \frac{0.9^2}{1 - 0.9} = 4.05 \quad customers in queue\]
Customers in system:
\[L= \rho+(1/2)*\frac{\rho^2}{1 - \rho}\]
\[L = 0.9+(1/2) * \frac{0.9^2}{1 - 0.9} = 4.95 \quad customers in system\]
Total time in system:
\[W = \frac{1}{\mu} +\frac{p}{2*\mu*(1-\rho)}\]
\[W = \frac{1}{1/0.9} +\frac{0.9}{2*(1/0.9)*(1-0.9)} = 4.95 \quad mins\]
Time in queue:
\[W_q = W-E(s) = W - (1/\mu)\]
\[W_q = 4.95-(1/(1/0.9)) = 4.05 \quad mins\]
Build a Simio model for the previous problem and compute the relevant metrics.
Model setup
Results, 24 hours
Experiment, 48 hours, 10 replications
Experiment, 48 hours, 10 replications
We see that a 48-hour simulation with a 10-hour warm up period returns closer results to the calculated results. The simulation returns 4.5 for average number in system, compared to the calculated 4.95.