Formula & Calculator
Average Number in Queue (M/M/1)
Calculates the average number of customers waiting in line (not being served) in a simple single-server queuing system.
Interpretation
Lq = ρ²/(1−ρ). Average number of customers waiting in an M/M/1 queue. Increases nonlinearly with utilisation (ρ). Used to estimate required waiting space and service capacity. Critical in queuing analysis.
Variables
| Symbol | Quantity | Unit |
|---|---|---|
| Lq | Average number waiting in queue | |
| ρ | Traffic intensity (utilization) |
What it means
For an M/M/1 queue (single server, Poisson arrivals, exponential service times), the average number of customers waiting in the queue (Lq) is given by Lq = ρ²/(1−ρ), where ρ = λ/μ (traffic intensity). This formula is derived from the steady‑state probabilities of the system. It shows that Lq grows rapidly as ρ approaches 1; for example, at ρ=0.5, Lq=0.5, but at ρ=0.9, Lq=8.1. This highlights the sensitivity of queue length to utilisation. The metric is used in designing waiting areas, determining buffer sizes in manufacturing, and sizing network routers. It also appears in Little’s Law (Lq = λ Wq) to compute average waiting time. In call centres, Lq helps forecast staffing needs. Understanding this formula is essential for operations managers to predict congestion and to make informed decisions about adding capacity or improving service rates to keep queues manageable and customer satisfaction high.
Worked example
Average Queue Length – Two Examples
Real‑World| Parameter | Value |
|---|---|
| ρ | 0.7 |
| Parameter | Value |
|---|---|
| ρ | 0.9 |
Common mistakes
- ρ: Traffic intensity (λ/μ) – must be < 1.
- Lq: Average number of customers waiting in the queue (not including those being served).
- Formula: Lq = ρ² / (1 – ρ) – valid for M/M/1 queue.
- Stability: If ρ ≥ 1, the queue length grows without bound – the formula gives infinity.
- Units: Lq is in number of customers.
Applications
The average number of customers in queue, Lq = ρ²/(1−ρ) for an M/M/1 queue, provides the expected queue length (excluding the customer in service). This metric is used by operations managers to estimate space requirements, to assess customer waiting experience, and to set service level targets. It grows non‑linearly as utilisation (ρ) increases, highlighting the importance of maintaining moderate utilisation for good service. In manufacturing, Lq helps determine work‑in‑process inventory between workstations. In retail, it guides checkout lane staffing. By calculating Lq, managers can predict congestion, justify capacity investments, and design facilities that minimise customer waiting, thereby improving satisfaction and operational efficiency.
- Design of queuing systems in retail, healthcare, and transportation
- Work‑in‑process inventory control in production lines
- Service level analysis and customer waiting time reduction
- Capacity planning for service and manufacturing operations
- Simulation and modelling of complex queuing networks
Frequently Asked Questions
For an M/M/1 queue, the average number of customers in the queue (waiting, not being served) is Lq = ρ² / (1 – ρ), where ρ = λ/μ is the traffic intensity (server utilization). This assumes ρ < 1.
Assuming queue length grows linearly with utilization. It actually grows explosively as ρ approaches 1. For example, at ρ = 0.9, Lq = 8.1, but at ρ = 0.95, Lq = 18.05, more than double. This is why systems run into trouble near full capacity.
By Little's Law, Lq = λ · Wq, so Wq = Lq / λ. Thus, the average waiting time is directly proportional to the queue length.
In an M/M/1 queue, L = ρ / (1 – ρ). Since L = Lq + ρ (the number in service is ρ), we have Lq = L – ρ = ρ²/(1 – ρ).
Lq increases with λ. As λ approaches μ (i.e., ρ → 1), Lq grows without bound. The relationship is highly non‑linear.
High Lq means long queues, leading to customer dissatisfaction, lost sales, and potential system failure. It may indicate the need for additional capacity or process improvements to reduce service time.
First compute λ = 1 / interarrival time, μ = 1 / service time, then ρ = λ/μ. Then apply the formula Lq = ρ²/(1 – ρ).
For a general G/G/1 queue, the formula for Lq involves the coefficient of variation of interarrival and service times. The M/M/1 formula is a special case with exponential distributions (CV=1).
At ρ = 0.8, Lq = 0.64 / 0.2 = 3.2 customers. This means on average, about 3.2 customers are waiting in line.
- Reduce ρ by increasing service capacity (more servers, faster service).
- Reduce variability in arrivals or service times (smoothing demand, standardising tasks).
- Use queue management strategies (e.g., appointment systems).