M/M/1 Queue Calculator

Queueing theory helps explain how a single server behaves when arrivals are random and service times are unpredictable. The M/M/1 model is the classic one-server queue: Poisson arrivals, exponential service times, and only one item being processed at a time. This calculator turns your arrival rate λ and service rate μ into the standard steady-state measures so you can judge whether a help desk, checkout lane, printer, or API worker is comfortably coping with demand or sliding toward congestion.

M/M/1 inputs: What you enter (and units)

For an M/M/1 queue, you only need two rates, and both must use the same time unit.

• Arrival rate λ: expected arrivals per unit time (e.g., 10 customers/hour).
• Service rate μ: expected services completed per unit time (e.g., 12 customers/hour).

Important: If λ is per hour, μ must also be per hour. The waiting-time outputs, W and Wq, come back in that same unit, so an hourly input will produce hour-based averages that you can convert to minutes if needed.

M/M/1 stability condition (when the formulas apply)

The steady-state M/M/1 formulas only make sense when the single server can keep up on average.

• μ > λ (equivalently, utilization ρ < 1).

If arrival pressure reaches or exceeds service capacity, the queue never settles into a stable long-run average. In that case, L, Lq, W, and Wq stop being useful as steady-state estimates because the line keeps growing instead of balancing out.

M/M/1 formulas and relationships

For an M/M/1 queue, the calculator starts with utilization, which shows what fraction of time the server is busy.

• ρ = λ/μ

From ρ, the standard M/M/1 results follow:

• L: average number of customers/jobs in the system (waiting + in service)
• Lq: average number waiting in line (excluding any in service)
• W: average time in the system (waiting + service time)
• Wq: average time waiting in line (excluding service time)

The relationships:

• L = ρ / (1 - ρ)
• Lq = ρ² / (1 - ρ)
• W = L / λ
• Wq = Lq / λ

These identities are all equivalent ways of reading the same queue. The forms with ρ are intuitive for congestion, while W = 1 / (μ - λ) and Wq = λ / (μ(μ - λ)) make it easy to see why delay rises sharply when arrival and service rates move close together.

The MathML below restates the same relationships for unambiguous rendering.

M/M/1 MathML (for unambiguous rendering)

How to interpret your M/M/1 queue results

• ρ (utilization): the busy fraction of the single server. As ρ approaches 1, the queue becomes fragile: short bursts of arrivals can turn into long waits because there is very little idle capacity left.
• L (avg. in system): the average number of jobs or customers in the system, including the one in service if someone is being served right now. Use it as a rough sense of how many items are tied up end-to-end.
• Lq (avg. in queue): the average number actually waiting. This is the number most people picture when they ask how long the line looks.
• W (avg. time in system): the total elapsed time from arrival to completion, combining both waiting and service.
• Wq (avg. waiting time): the time spent waiting before service begins, which is often the metric people care about when they ask whether the line feels slow.

Because the M/M/1 formulas are nonlinear, a small improvement in μ or a small reduction in λ can have an outsized effect once utilization is already high. That makes this calculator especially useful for comparing staffing or throughput changes before you commit to them.

Worked example: an M/M/1 help desk under moderate load

Here is a simple M/M/1 example using a single help-desk agent that receives about 5 requests per hour and can complete about 7 per hour.

• λ = 5 arrivals/hour
• μ = 7 services/hour

1) Utilization:

• ρ = λ/μ = 5/7 ≈ 0.714

2) Average number in system:

• L = ρ/(1-ρ) = 0.714 / 0.286 ≈ 2.50

3) Average number waiting:

• Lq = ρ²/(1-ρ) = 0.714² / 0.286 ≈ 1.79

4) Average time in system and in queue:

• W = L/λ = 2.50/5 = 0.50 hours ≈ 30 minutes
• Wq = Lq/λ = 1.79/5 = 0.357 hours ≈ 21.4 minutes

A quick consistency check is that the mean service time is 1/μ = 1/7 hour ≈ 8.6 minutes. Adding that service time to Wq gives W, which is why the total time in system comes out to roughly 30 minutes.

M/M/1 formula summary table

Assumptions and limitations of the M/M/1 model

• Poisson arrivals (Markovian): arrivals into the M/M/1 queue are assumed independent, with memoryless interarrival times and a constant λ.
• Exponential service times (Markovian): each job’s service duration is assumed memoryless, with a constant average rate μ.
• Single server: exactly one job/customer can be served at a time; this is not a multi-server or pooled-capacity model.
• Queue discipline: the model usually assumes first-come, first-served. Other rules can change who waits longest, even if the average formulas sometimes stay similar under special conditions.
• Infinite queue capacity and calling population: the model does not account for balking, blocking, or a small finite population that can run out of arrivals.
• Steady-state requirement: the closed-form averages require λ < μ, otherwise the line does not settle.
• Real-world variability: breaks, batching, priorities, time-of-day swings, or non-exponential service can push actual waits away from the textbook M/M/1 result.

Historical note: Where the M/M/1 model came from

The M/M/1 model grew out of early telephone-traffic studies, where engineers needed a simple way to estimate how a single operator or line would behave under random call arrivals. The same idea still matters whenever one server has to absorb unpredictable demand, from contact centers to web services and small production lines.

Privacy: Calculations stay in your browser, so the M/M/1 rates you enter are not sent to a server by default.

How to use this M/M/1 queue calculator

1. Enter Arrival Rate λ using the unit or time period you want to model.
2. Enter Service Rate μ using that same unit so arrival pressure and service capacity are comparable.
3. Run the calculation, then try a second λ/μ pair to see how the M/M/1 wait changes before you make staffing or throughput decisions.