What Is M/M/1 Queueing Theory?
The M/M/1 queue is the simplest and most fundamental model in queueing theory. It describes a single-server queue where arrivals follow a Poisson process (rate λ) and service times are exponentially distributed (rate μ). The system is stable when the utilization ρ = λ/μ is less than 1.
M/M/1 Queue Formulas
Utilization
Fraction of time the server is busy.
Avg Customers in System
Expected number of customers including the one being served.
Avg Queue Length
Expected number waiting in queue (not being served).
Avg System Time
Expected total time a customer spends in the system.
Avg Wait Time
Expected time waiting in queue before being served.
Probability of n Customers
Probability of exactly n customers in the system.
Understanding the M/M/1 Notation
In Kendall notation, M/M/1 means:
- First M: Markovian (Poisson) arrival process
- Second M: Markovian (exponential) service time distribution
- 1: Single server
Stability Condition
The M/M/1 queue is stable (reaches steady state) only when ρ = λ/μ < 1, meaning the service rate must exceed the arrival rate. If ρ >= 1, the queue grows without bound.
Real-World Applications
- Call centers: Predicting hold times and staffing needs.
- Network traffic: Modeling packet queues in routers and switches.
- Retail: Optimizing checkout lane staffing.
- Healthcare: Emergency room wait time estimation.
- Manufacturing: Production line bottleneck analysis.
Little's Law
Little's Law (L = λW) connects the average number of customers in the system to the arrival rate and average system time. It holds for any stable queueing system and is used to verify the M/M/1 results: L = λ / (μ - λ) and W = 1 / (μ - λ), confirming L = λW.