Queueing Theory Calculator
Test cases replayed : 3 of 3 match · Methodology · No outside source listed
Enter the average arrival rate, the average service rate of one server (both per hour) and the number of servers (1 for an M/M/1 queue).
Result
Average wait in line Wqminutes
Offered load a = λ ÷ μ (erlangs)
- Utilization of each server ρ = a ÷ c
- Probability the system is empty P₀
- Probability an arrival has to wait (Erlang C)
- Average number waiting in line Lq
- Average number in the system L = Lq + a
- Average time in the system W = Wq + 1/μ (minutes)
Estimate for planning only, not tax, legal, financial or medical advice. This calculator uses the method explained on this page.
This browser did not allow copying. The sentence is in the box below, already selected: copy it with your keyboard or the menu.
How to use this calculator
Fill in the fields in this order; each step takes you to its field.
- Arrival rate λ (customers per hour): enter 0 or more.
- Service rate μ of one server (customers per hour): enter 0 or more.
- Number of servers c (1 to 20): enter a number from 1 to 20.
- Read the result: Average wait in line Wq; the panel also shows Offered load a = λ ÷ μ (erlangs), Utilization of each server ρ = a ÷ c, Probability the system is empty P₀, Probability an arrival has to wait (Erlang C), Average number waiting in line Lq, Average number in the system L = Lq + a and Average time in the system W = Wq + 1/μ (minutes).
How this calculator works
This is the M/M/c model of queueing theory: random (Poisson) arrivals at rate λ, exponential service times with rate μ per server, c identical servers and one line. It is stable only when the utilization ρ = λ ÷ (cμ) is below 100%; otherwise the line grows without limit and the results show —.
With a = λ/μ, P₀ = 1 ÷ [Σₖ₌₀^(c−1) aᵏ/k! + aᶜ/(c!(1 − ρ))]. The Erlang C formula (A. K. Erlang, 1917) gives the chance of waiting: C = aᶜ/(c!(1 − ρ)) × P₀. Then Lq = C ρ/(1 − ρ), Wq = Lq/λ, W = Wq + 1/μ and L = λW (Little's law).
For one server (M/M/1) these reduce to Lq = ρ²/(1 − ρ) and W = 1/(μ − λ). Times are shown in minutes because the rates are per hour.
Worked example
| Arrival rate λ (customers per hour) | 10 |
|---|---|
| Service rate μ of one server (customers per hour) | 6 |
| Number of servers c (1 to 20) | 2 |
| Offered load a = λ ÷ μ (erlangs) | 1.6667 |
|---|---|
| Utilization of each server ρ = a ÷ c | 83.33% |
| Probability the system is empty P₀ | 9.09% |
| Probability an arrival has to wait (Erlang C) | 75.76% |
| Average number waiting in line Lq | 3.7879 |
| Average number in the system L = Lq + a | 5.4545 |
| Average wait in line Wq | 22.7273 minutes |
| Average time in the system W = Wq + 1/μ (minutes) | 32.7273 |
Frequently asked questions
A single cashier serves 5 customers an hour and 4 arrive per hour. How long is the wait?
ρ = 80%, Lq = 3.2 customers and Wq = 0.8 hour = 48 minutes; total time in the system W = 60 minutes.
Why does the wait explode near 100% utilization?
Lq contains the factor 1/(1 − ρ): at 90% it is 10, at 99% it is 100, so small increases in traffic cause long lines.