STATISTICSExploratoryMathematics Calculator
📊

Queueing Theory — M/M/1, M/M/s, Little's Law

Operations research: arrival rate λ, service rate μ. Utilization ρ = λ/μ. Little's Law: L = λW. Erlang formulas for queue length and waiting time.

Concept Fundamentals
λ/μ
ρ
λW
L
1 server
M/M/1
s servers
M/M/s
Queue MetricsErlang & Little

Why This Mathematical Concept Matters

Why: Queueing theory models waiting lines: call centers, traffic, healthcare. Utilization ρ < 1 for stability. Little's Law: avg in system = arrival rate × avg time.

How: M/M/1: exponential arrivals and service, 1 server. M/M/s: s parallel servers. L = ρ/(1−ρ) for M/M/1. Erlang C for probability of wait.

  • Little's Law: L = λW holds for any steady-state queue.
  • ρ ≥ 1: queue grows without bound (unstable).
  • Erlang formulas: Danish engineer A.K. Erlang (1909), telephone traffic.
📊
OPERATIONS RESEARCHQueue & Waiting

Queueing Theory — M/M/1, M/M/s, Little's Law

Erlang formulas, utilization, queue length distribution. Call centers, traffic, healthcare.

Sample Examples

Grocery Checkout

Single cashier grocery store checkout line

Arrival Rate: 10
Service Rate: 12
Servers: 1
Model: M/M/1

Customer Support Center

Call center with multiple service agents

Arrival Rate: 15
Service Rate: 4
Servers: 5
Model: M/M/s

Bank Tellers

Bank with multiple tellers serving customers

Arrival Rate: 8
Service Rate: 5
Servers: 2
Model: M/M/s

Fast Food Restaurant

Fast food restaurant drive-through window

Arrival Rate: 20
Service Rate: 22
Servers: 1
Model: M/M/1

Inputs

Average number of customers arriving per hour
Average number of customers that can be served per hour

⚠️For educational and informational purposes only. Verify with a qualified professional.

🧮 Fascinating Math Facts

📊

Little's Law: L = λW. Avg customers = arrival rate × avg time in system

— Operations Research

📞

Erlang (1909) developed queueing theory for telephone exchanges

— History

📋 Key Takeaways

  • Little's Law: L = λW. Applies to virtually any queue
  • M/M/1: Single server, Poisson arrivals, exponential service
  • Utilization ρ < 1 required for stability
  • Erlang formulas for call centers, traffic, healthcare

💡 Did You Know?

📞Erlang developed queueing theory for telephone exchanges in 1909Source: History
🚗Traffic flow uses M/M/1: cars = arrivals, road capacity = service rateSource: Transport
🏥ERs use queue models for staffing. ρ > 0.9 means long waitsSource: Healthcare
📊L = ρ/(1-ρ) for M/M/1. As ρ→1, queue length explodesSource: Formulas
⏱️Wq = Lq/λ. Average wait = queue length / arrival rateSource: Little's Law
🔢M/M/s: s servers. Same ρ = λ/(sμ) for stabilitySource: Multi-server

What is Queueing Theory?

Queueing theory is the mathematical study of waiting lines (queues). It analyzes the behavior of service systems where customers arrive, wait for service, and eventually depart after being served.

Developed by Danish engineer A.K. Erlang in the early 1900s to model telephone exchange congestion, queueing theory has become essential for optimizing service systems in various industries including telecommunications, healthcare, manufacturing, and customer service.

Applications

  • Call centers and help desks
  • Hospital emergency rooms
  • Bank teller services
  • Manufacturing production lines
  • Network data packet handling
  • Traffic flow management

Key Benefits of Queueing Analysis

  • Predict system performance metrics like waiting times and queue lengths
  • Determine optimal staffing levels to balance costs and service quality
  • Identify capacity constraints and bottlenecks
  • Evaluate different service policies and their impact
  • Guide design decisions for new service systems

Kendall's Notation

Queueing systems are commonly classified using Kendall's notation in the form A/B/C/D/E/F, which concisely describes the key characteristics of a queue:

SymbolDescriptionCommon Values
AArrival processM Markovian (Poisson)D DeterministicG General
BService time distributionM Markovian (Exponential)D DeterministicG General
CNumber of servers1, 2, ..., ∞
DSystem capacityK (maximum customers allowed) or ∞
EPopulation sizeN (finite source) or ∞
FService disciplineFIFO First In First OutLIFO Last In First OutPR Priority

For example, M/M/1 represents a queue with Poisson arrivals, exponential service times, and a single server. Similarly, M/M/s is the multi-server extension with s servers.

Key Performance Metrics

System Utilization (ρ)

rho=fraclambdasmu\\rho = \\frac{\\lambda}{s\\mu}

The fraction of time servers are busy. For stability, ρ must be less than 1.

Average Queue Length (Lq)

Lq=textAveragenumberinqueueL_q = \\text{Average number in queue}

The average number of customers waiting in line (not being served).

Average System Length (L)

L=Lq+fraclambdamuL = L_q + \\frac{\\lambda}{\\mu}

The average number of customers in the entire system (queue + service).

Average Waiting Time (Wq)

Wq=fracLqlambdaW_q = \\frac{L_q}{\\lambda}

The average time a customer spends waiting in line before service starts.

Average System Time (W)

W=Wq+frac1muW = W_q + \\frac{1}{\\mu}

The average total time a customer spends in the system (waiting + service).

Little's Law

L=lambdaWL = \\lambda W

A fundamental relationship that applies to virtually all queueing systems.

M/M/1 Queue

The simplest queueing model with a single server, Poisson arrivals, and exponential service times.

Key Formulas

ρ = λ/μ
L = ρ/(1-ρ)
Lq = ρ²/(1-ρ)
W = 1/(μ-λ)
Wq = ρ/(μ-λ)
P₀ = 1-ρ

M/M/1 Example: Bank Teller

A bank teller serves customers at rate μ = 12 customers/hour, while customers arrive at rate λ = 10 customers/hour.

The system utilization is ρ = 10/12 = 0.83 or 83%.

Average queue length: Lq = 0.83²/(1-0.83) ≈ 4.1 customers

Average waiting time: Wq = 4.1/10 ≈ 0.41 hours (≈ 25 minutes)

M/M/s Queue

A multi-server queue model with Poisson arrivals and exponential service times. Key formulas are more complex and involve calculating P₀ first.

Performance Metrics

Utilization
ρ = λ/(s·μ)
Probability of empty system
P₀ = [Σₙ₌₀ˢ⁻¹((λ/μ)ⁿ/n!) + ((λ/μ)ˢ/(s!))·(sμ)/(sμ-λ)]⁻¹
Average queue length
Lq = ((λ/μ)ˢ·P₀·ρ)/(s!·(1-ρ)²)

M/M/s Example: Call Center

A call center has 5 agents, each serving at rate μ = 4 calls/hour, with calls arriving at rate λ = 15 calls/hour.

The system utilization is ρ = 15/(5×4) = 0.75 or 75%.

With 5 agents, the queue is much shorter than if there were only one agent with the same utilization.

Real-World Applications

1

Healthcare

Hospitals use queueing theory to optimize:

  • Emergency room staffing
  • Operating room scheduling
  • Bed capacity planning
  • Patient appointment systems
2

Telecommunications

Network operators apply queueing models to:

  • Call center staffing
  • Network traffic management
  • Bandwidth allocation
  • Server capacity planning
3

Transportation

Transportation systems use queueing theory for:

  • Traffic signal timing
  • Toll booth optimization
  • Airport security checkpoint design
  • Public transit scheduling

Interactive Exploration

Use these interactive tools to better understand queueing theory concepts and visualize how different parameters affect system performance.

System Utilization (ρ)

System utilization represents the fraction of time that servers are busy. It's calculated as:

ρ = λ / (s·μ)

Where:

  • λ (lambda) = average arrival rate
  • μ (mu) = average service rate per server
  • s = number of servers

Important:

A queueing system is only stable when ρ < 1. If ρ ≥ 1, the queue will grow indefinitely.

Effect of Utilization on Performance

Low Utilization (30-50%)
  • Fast customer service
  • Short wait times
  • Short queue lengths
  • Idle servers (higher cost)
High Utilization (90-99%)
  • Long customer wait times
  • Long queue lengths
  • Customer dissatisfaction
  • Efficient resource usage

Key Insight: As utilization increases, queue length and waiting time grow exponentially. A system at 95% utilization doesn't have just slightly longer queues than one at 80% - it has dramatically longer queues.

Compare Queue Models

See how adding multiple servers can dramatically reduce queue length, even with the same utilization level.

Interactive Queue Model Comparison

How this works: This interactive tool demonstrates how multiple servers can dramatically reduce queue length compared to a single server, even at the same utilization level. The visualization shows both queue models side-by-side with their respective queue lengths.

10%50%95%
Current: 70%
3

M/M/1 Queue (Single Server)

Queue Visualization

Average Queue Length: 1.63 customers

M/M/3 Queue (Multiple Servers)

Queue Visualization

Average Queue Length: 10.80 customers

Performance Improvement

Using 3 servers reduces queue length by approximately -562% compared to a single server at 70% utilization.

This demonstrates why many service systems use multiple servers rather than a single fast server. Even though the total service capacity remains the same, multiple servers handle variability better.

👈 START HERE
⬅️Jump in and explore the concept!
AI