Queueing Model Input
Use the stacked page layout below. Input fields switch to 3 columns on large screens, 2 on medium screens, and 1 on mobile.
Formula Used
Core definitions: λ is the arrival rate, μ is the service rate per server, c is the number of servers, and K is system capacity.
Traffic intensity: ρ = λ / (cμ). For M/M/1, this becomes ρ = λ / μ.
M/M/1: P0 = 1 - ρ, Lq = ρ² / (1 - ρ), L = ρ / (1 - ρ), Wq = Lq / λ, and W = 1 / (μ - λ).
M/M/c: Erlang C is used. The idle probability P0 comes from the standard normalizing sum. Then P(wait), Lq, Wq, W, and L are computed from that steady-state form.
Finite-capacity models: M/M/1/K and M/M/c/K use birth-death state probabilities. The calculator normalizes all states from 0 to K, finds P(K) as blocking probability, then computes throughput λe = λ(1 - P(K)).
Little’s Law: Wherever the system is in steady state, L = λeW and Lq = λeWq.
How to Use This Calculator
1. Select the queueing structure that matches your service system.
2. Enter the average arrival rate and service rate using the same time basis.
3. Add the number of servers for multi-server systems.
4. Add capacity only when the system has a maximum limit.
5. Click the calculate button to display utilization, queue length, waiting time, throughput, and blocking performance.
6. Download the result summary as CSV or PDF for reporting.
Example Data Table
| Scenario | Model | Arrival Rate λ | Service Rate μ | Servers c | Capacity K | Time Unit |
|---|---|---|---|---|---|---|
| Single counter checkout | M/M/1 | 4 | 6 | 1 | Infinite | minutes |
| Three-agent support desk | M/M/c | 12 | 5 | 3 | Infinite | minutes |
| Limited waiting room clinic | M/M/1/K | 6 | 8 | 1 | 5 | minutes |
| Server cluster with buffer cap | M/M/c/K | 20 | 8 | 3 | 7 | seconds |
FAQs
1. What does a queueing model measure?
A queueing model measures how customers, tasks, or packets arrive, wait, get served, and sometimes get rejected. It estimates congestion, delays, queue length, throughput, and service utilization.
2. When should I use M/M/1 instead of M/M/c?
Use M/M/1 for one service channel, such as one cashier or one processor. Use M/M/c when several identical servers work in parallel and share the same arrival stream.
3. What does the capacity K mean?
Capacity K is the maximum number of customers allowed inside the whole system, including both waiting and being served. When the system is full, new arrivals are blocked or lost.
4. Why does the calculator mark some systems as unstable?
Unlimited models become unstable when arrivals meet or exceed service capacity. In that case, average queue length and waiting time keep growing, so steady-state results are no longer finite.
5. What is utilization telling me?
Utilization shows the share of available service capacity that is busy on average. Very high utilization often increases waiting sharply, even when the system still looks technically stable.
6. What is the difference between W and Wq?
Wq is the average time spent waiting before service begins. W is the total average time in the system, so it includes both queueing time and service time.
7. Can I use different units for arrival and service rates?
No. Both rates must use the same time basis. For example, if arrivals are per minute, service rate must also be per minute. Mixed units create misleading results.
8. Why are finite-capacity results still available at high demand?
A finite-capacity system cannot grow forever because it has a hard limit. Excess arrivals are blocked, so the state space stays bounded and steady-state probabilities can still be computed.