Study guide

9.1 Queuing Theory: Little's Law, M/M/1, and Multi-Server Systems

FE Industrial and Systems Study Guide Exam

Track your progress as you read

Finished sections get a check mark.

1 Chapter 1: Exam Overview, CBT Strategy & Handbook Navigation

1.1 FE Industrial Exam Structure, CBT Logistics, and Pacing 1.2 Using the NCEES FE Reference Handbook and Approved Calculators

2 Chapter 2: Mathematics for Industrial Engineers

2.1 Analytic Geometry, Coordinate Systems, and Trigonometry 2.2 Calculus: Differential, Integral, and Series Approximations 2.3 Linear Algebra: Matrix Operations, Systems of Equations, and Vectors

3 Chapter 3: Engineering Sciences Core

3.1 Thermodynamics Fundamentals and Fluid Flow Principles 3.2 Statics, Dynamics, and Mechanics of Materials Core 3.3 Electrical Circuits, DC/AC Analysis, and Power in Industrial Systems

4 Chapter 4: Ethics and Professional Practice

4.1 NCEES Model Rules of Professional Conduct and Technical Society Ethics 4.2 Engineering Licensure, Public Welfare, and Professional Liability 4.3 Engineering Contracts, Intellectual Property, and Conflicts of Interest

5 Chapter 5: Engineering Economics

5.1 Time Value of Money, Discount Factors, and Equivalence 5.2 Evaluating Alternatives: Present Worth, Equivalent Annual Cost, and IRR 5.3 Incremental Rate of Return, Replacement Analysis, and Break-Even 5.4 Depreciation Methods, MACRS, Income Taxes, and Inflation

6 Chapter 6: Probability Foundations & Distributions

6.1 Basic Probability Concepts, Conditional Probability, and Bayes' Theorem 6.2 Discrete Probability Distributions 6.3 Continuous Probability Distributions

7 Chapter 7: Statistical Inference, Regression & Design of Experiments

7.1 Sampling Distributions, Central Limit Theorem, and Confidence Intervals 7.2 Hypothesis Testing: Means, Proportions, and Variances 7.3 Linear Regression, Correlation Analysis, and Design of Experiments (ANOVA)

8 Chapter 8: Data, Logic, Modeling & Linear Programming

8.1 Optimization Modeling: Decision Variables, Objectives, and Constraints 8.2 Linear Programming: Graphical Solutions and Simplex Fundamentals 8.3 Duality, Sensitivity Analysis, and Integer Linear Programming 8.4 Data, Logic Development, and Analytics

9 Chapter 9: Stochastic Models & Simulation

9.1 Queuing Theory: Little's Law, M/M/1, and Multi-Server Systems 9.2 Discrete-Time Markov Chains: Transition Matrices and Steady-State Probabilities 9.3 Discrete-Event Simulation Principles and Monte Carlo Modeling

10 Chapter 10: Engineering Management, Project Planning & Performance Measurement

10.1 Organizational Structure, Leadership, and Worker Motivation Theories 10.2 Project Management: WBS, Critical Path Method (CPM), PERT, and Crashing 10.3 Earned Value Management (EVM), Decision Trees, and Risk Preference 10.4 Performance Measurement: Productivity, KPIs, Wage Systems, and the Balanced Scorecard

11 Chapter 11: Manufacturing & Production Systems

11.1 Manufacturing Processes, Material Processing, and Tolerancing 11.2 Assembly Line Balancing, Takt Time, Cycle Time, and Throughput 11.3 Sustainable Manufacturing, Waste Minimization, and Life-Cycle Assessment

12 Chapter 12: Production Planning, Inventory Control & Sequencing

12.1 Demand Forecasting Methods and Forecast Error Tracking 12.2 Aggregate Planning, Master Production Scheduling (MPS), and MRP 12.3 Deterministic Inventory Models: EOQ, EPQ, and Safety Stock 12.4 Job Sequencing, Dispatching Rules, and Flow Shop Scheduling

13 Chapter 13: Lean Systems & Theory of Constraints

13.1 Lean Foundations: Value Stream Mapping, Seven Wastes, and 5S 13.2 Pull Systems, Kanban, Just-In-Time (JIT), and SMED Setup Reduction 13.3 Theory of Constraints: Bottleneck Identification and Drum-Buffer-Rope

14 Chapter 14: Facilities Planning & Material Handling

14.1 Facility Location Analysis: Qualitative Factor Rating and Centroid Method 14.2 Facility Layout Design and Flow Analysis 14.3 Material Handling Equipment, Warehousing, and AS/RS

15 Chapter 15: Supply Chain Design & Logistics

15.1 Supply Chain Network Architecture and Transportation Modes 15.2 Distribution Strategies, Cross-Docking, and Vehicle Routing 15.3 The Bullwhip Effect, Risk Pooling, and Supply Chain Coordination

16 Chapter 16: Human Factors & Cognitive Ergonomics

16.1 Cognitive Engineering, Human-Machine Interface, Displays, and Controls 16.2 Anthropometric Principles, Percentile Design, and Biomechanical Posture 16.3 Environmental Ergonomics: Industrial Noise, Lighting, and Heat Stress

17 Chapter 17: Physical Ergonomics & Workplace Safety

17.1 Physical Ergonomics: The NIOSH Manual Lifting Equation 17.2 Cumulative Trauma Disorders and Postural Assessment 17.3 Workplace Safety: OSHA Standards, Hazard Analysis, and Control Hierarchy

18 Chapter 18: Work Design & Measurement

18.1 Methods Engineering, Process Charting, and Principles of Motion Economy 18.2 Direct Stopwatch Time Study: Observed Time, Performance Rating, and Allowances 18.3 Predetermined Time Systems, Work Sampling, and Learning Curves

19 Chapter 19: Quality Management, Six Sigma & Analytical Tools

19.1 Quality Philosophies, Total Quality Management, and Cost of Quality 19.2 Six Sigma DMAIC Methodology and Quality Function Deployment 19.3 The Seven Basic Quality Tools, Root Cause Analysis, and Taguchi Loss Function

20 Chapter 20: Statistical Process Control

20.1 Statistical Process Control Principles, Variation, and Run Rules 20.2 Variable Control Charts: X-bar & R, X-bar & S, and Individual-Moving Range 20.3 Attribute Control Charts: Defectives (p, np) and Defects (c, u)

21 Chapter 21: Process Capability & Acceptance Sampling

21.1 Process Capability Indices and Defect Levels 21.2 Acceptance Sampling Concepts and Operating Characteristic Curves 21.3 Sampling Risks, AQL, LTPD, and Standard Sampling Plans

22 Chapter 22: Systems Engineering & Life-Cycle Management

22.1 The Systems Engineering V-Model, Stakeholder Requirements, and Functional Architecture 22.2 Configuration Management, Interface Control, and Verification & Validation 22.3 Life-Cycle Costing, Maintainability, and System Availability

23 Chapter 23: Reliability Engineering & Risk Analysis

23.1 Reliability Mathematics, Hazard Rates, and the Bathtub Curve 23.2 System Reliability Configurations: Series, Parallel, and k-out-of-n Systems 23.3 Failure Modes and Effects Analysis and Fault Tree Analysis

9.1 Queuing Theory: Little's Law, M/M/1, and Multi-Server Systems

Key Takeaways Explain This Topic AI

Little's Law (L = λW and L_q = λW_q) is a universal conservation relationship governing any steady-state queuing system regardless of arrival distribution, service distribution, or queue discipline, with system time decomposing as W = W_q + 1/μ.

For an M/M/1 queue, system stability strictly requires server utilization ρ = λ / μ < 1; state probabilities follow the geometric progression P_n = (1 - ρ)ρ^n, yielding average system length L = λ / (μ - λ) and average queue length L_q = λ² / [μ(μ - λ)].

Kendall's notation (A/B/s/K/N/Z) standardizes queuing classification: A denotes arrival distribution, B service distribution, s number of parallel servers, K system capacity, N calling population, and Z queue discipline (defaulting to M/M/s/∞/∞/FCFS).

In multi-server M/M/s configurations, overall utilization is ρ = λ / (sμ) < 1; server sharing reduces queue build-up significantly compared to isolated single-server lines due to pooled capacity and stochastic buffering.

Economic optimization of queuing systems trades off the linear cost of providing server capacity against the convex cost of customer or WIP waiting time, minimizing total expected cost per unit time TC = s · C_s + L · C_w.

Last updated: September 2026

Queuing theory is the mathematical study of waiting lines, congestion, and service delays in stochastic systems. In industrial and systems engineering, queuing models govern the design and optimization of manufacturing cells, automated material handling networks, maintenance repair depots, telecommunication switches, hospital emergency departments, and logistics supply chains. The NCEES FE Reference Handbook collects essential queuing formulas under its Industrial and Systems Engineering section. To excel on the FE exam, candidates must master Kendall's notation, recognize system stability limits, apply Little's Law across arbitrary system boundaries, calculate operating characteristics for single-server ( M / M / 1 M/M/1 M / M /1 ) and multi-server ( M / M / s M/M/s M / M / s ) systems, and execute economic trade-off analyses between service capacity and waiting costs.

  • 1. Elements of a Queuing System
  • Every queuing system consists of six fundamental structural components: Calling Departing
  • Population ──> [ Arrival Process (λ) ] ──> [ Queue / Buffer (K) ] ──> [ Service Facility (s servers, μ) ] ──>
  • Queue Discipline (FCFS)

Arrival Process : Characterizes how entities enter the system over time. The standard assumption in stochastic queuing models is a Poisson arrival process with constant mean arrival rate λ \lambda λ (customers/units per unit time). Under a Poisson arrival process, the number of arrivals in an interval of length t t t follows a Poisson distribution with parameter λ t \lambda t λ t , and the time between consecutive arrivals (interarrival times) is independent and identically distributed ( i.i.d. ) according to an Exponential distribution with mean 1 / λ 1/\lambda 1/ λ .

Service Mechanism : Characterizes the duration required to serve an entity. Service times are typically modeled as random variables. In Markovian queues, service duration follows an Exponential distribution with mean service time 1 / μ 1/\mu 1/ μ , where μ \mu μ is the mean service rate per busy server (units served per unit time).

Number of Servers ( s s s or c c c ) : The number of identical, parallel service channels available to process arriving entities. If s = 1 s = 1 s = 1 , arriving entities form a single line for a single server. If s > 1 s > 1 s > 1 , entities typically wait in a single common queue and advance to the first available server.

Queue Discipline : The priority rule dictating the order in which waiting entities are selected for service:

FCFS / FIFO : First-Come, First-Served (First-In, First-Out). The default baseline for most physical queuing systems.

LCFS / LIFO : Last-Come, First-Served (e.g., stack-based inventory retrieval or push-down data stacks).

SIRO : Service in Random Order.

Priority Disciplines : Entities possess priority attributes. In preemptive priority , a higher-priority arrival interrupts a lower-priority customer currently in service; in non-preemptive priority , the higher-priority customer moves to the head of the queue but waits for the current service to finish.

Processor Sharing : All waiting entities share available server capacity equally (common in computer operating systems and web server thread allocation).

System Capacity ( K K K ) : The maximum number of entities allowed in the system concurrently (including both entities waiting in queue and entities actively in service). An infinite capacity system ( K = ∞ K = \infty K = ∞ ) can accommodate an arbitrary queue length. In a finite capacity system ( K < ∞ K < \infty K < ∞ ), when the system contains K K K entities, any subsequent arrival is turned away (blocked or lost), such as in telecommunications call centers.

Calling (Source) Population ( N N N ) : The pool of potential customers. An infinite calling population assumes the arrival rate λ \lambda λ is unaffected by the number of customers currently in the system. A finite calling population ( N < ∞ N < \infty N < ∞ ) assumes each customer currently in the system reduces the arrival rate of remaining potential customers, typical in internal machine maintenance pools (e.g., a technician maintaining a shop floor of N = 10 N = 10 N = 10 stamping presses).

2. Kendall's Notation

David G. Kendall formalized a standardized shorthand notation to classify queuing systems using the format: A / B / s / K / N / Z A \, / \, B \, / \, s \, / \, K \, / \, N \, / \, Z A / B / s / K / N / Z

Where:

  • A A A : Arrival interarrival time distribution
  • B B B : Service time distribution
  • s s s : Number of parallel servers ( s = 1 , 2 , 3 , … s = 1, 2, 3, \dots s = 1 , 2 , 3 , … )
  • K K K : System capacity (maximum units in queue plus service; omitted if ∞ \infty ∞ )
  • N N N : Calling population size (omitted if ∞ \infty ∞ )
  • Z Z Z : Queue discipline (omitted if FCFS/FIFO)
  • Standard Distribution Codes

M M M : Markovian (exponential interarrival or service times, possessing the memoryless property; Poisson arrival process)

D D D : Deterministic / constant time (zero variance, such as an automated transfer indexing line)

E k E_k E k ​ : Erlang distribution with shape parameter k k k (modeling phases of independent exponential operations)

G G G or G I GI G I : General independent distribution (arbitrary mean and variance)

  • Kendall Notation Arrival Distribution Service Distribution Servers Capacity Population Discipline
  • M / M / 1 M/M/1 M / M /1 Exponential (Poisson) Exponential 1 1 1 ∞ \infty ∞ ∞ \infty ∞ FCFS
  • M / M / s M/M/s M / M / s Exponential (Poisson) Exponential s s s ∞ \infty ∞ ∞ \infty ∞ FCFS
  • M / G / 1 M/G/1 M / G /1 Exponential (Poisson) General 1 1 1 ∞ \infty ∞ ∞ \infty ∞ FCFS
  • M / D / 1 M/D/1 M / D /1 Exponential (Poisson) Constant 1 1 1 ∞ \infty ∞ ∞ \infty ∞ FCFS
  • M / M / s / K M/M/s/K M / M / s / K Exponential (Poisson) Exponential s s s K K K (Finite) ∞ \infty ∞ FCFS
  • M / M / 1 / N / N M/M/1/N/N M / M /1/ N / N Exponential (Poisson) Exponential 1 1 1 N N N N N N (Finite) FCFS

Exam Shorthand Convention : When only three descriptors are provided ( A / B / s A/B/s A / B / s ), the system implicitly assumes infinite capacity ( K = ∞ K = \infty K = ∞ ), infinite calling population ( N = ∞ N = \infty N = ∞ ), and FCFS queue discipline ( Z = FCFS Z = \text{FCFS} Z = FCFS ). Thus, M / M / 1 M/M/1 M / M /1 denotes M / M / 1 / ∞ / ∞ / FCFS M/M/1/\infty/\infty/\text{FCFS} M / M /1/∞/∞/ FCFS .

  • 3. Traffic Intensity and Stability Conditions
  • Single-Server Utilization

For a single-server system ( s = 1 s = 1 s = 1 ), the server utilization factor ρ \rho ρ represents the fraction of time the server is busy: ρ = λ μ \rho = \frac{\lambda}{\mu} ρ = μ λ ​

Steady-State Stability Criterion : A queuing system reaches a stationary, steady-state operating equilibrium if and only if: ρ = λ μ < 1 ⟺ λ < μ \rho = \frac{\lambda}{\mu} < 1 \quad \iff \quad \lambda < \mu ρ = μ λ ​ < 1 ⟺ λ < μ

If λ > μ \lambda > \mu λ > μ , arrivals outpace service capacity. The queue grows infinitely large over time without bound, and no steady-state distribution exists.

If λ = μ \lambda = \mu λ = μ , stochastic variability guarantees that the queue will eventually wander toward infinity; hence ρ = 1 \rho = 1 ρ = 1 is unstable for infinite capacity queues.

Multi-Server Utilization

For a multi-server system with s s s identical parallel servers each operating at mean rate μ \mu μ , the total combined service capacity is s μ s\mu s μ . The system utilization factor ρ \rho ρ (average utilization per server) is defined as: ρ = λ s μ \rho = \frac{\lambda}{s \mu} ρ = s μ λ ​

The term r = λ / μ r = \lambda / \mu r = λ / μ is the offered load (or traffic intensity in Erlangs), representing the expected number of servers required to handle the traffic. The multi-server stability condition requires: ρ = λ s μ < 1 ⟺ λ < s μ \rho = \frac{\lambda}{s \mu} < 1 \quad \iff \quad \lambda < s \mu ρ = s μ λ ​ < 1 ⟺ λ < s μ

4. Little's Law and Fundamental Conservation Equations

John D.C. Little formulated what is arguably the most powerful and widely applied relationship in industrial engineering and operations research: Little's Law .

Mathematical Formulation

In any steady-state queuing system where entities do not vanish: L = λ W L = \lambda W L = λW L q = λ W q L_q = \lambda W_q L q ​ = λ W q ​

Where:

  • L L L : Average number of entities in the system (both waiting in line and actively receiving service)
  • L q L_q L q ​ : Average number of entities waiting in the queue
  • W W W : Average total time an entity spends in the system (waiting time plus service time)
  • W q W_q W q ​ : Average time an entity waits in the queue prior to the start of service

λ \lambda λ : Long-run average arrival rate to the system │◄─────────────────────── W (System Time) ──────────────────────►│

  • │◄──────── Wq (Queue Time) ────────►│◄──── 1/μ (Service Time) ──►│
  • ┌───────────────────────────────────┬────────────────────────────┐
  • Arrivals (λ) ───────────────>│ Queue (Lq) │ Service (L - Lq) │───────────> Departures
  • └───────────────────────────────────┴────────────────────────────┘
  • │◄─────────────────────── L (System Units) ─────────────────────►│
  • Time and Entity Decomposition

The total time an entity spends in the system is the sum of its waiting time in queue and its active service time: W = W q + 1 μ W = W_q + \frac{1}{\mu} W = W q ​ + μ 1 ​

Multiplying this entire relationship by the arrival rate λ \lambda λ yields the entity decomposition formula: λ W = λ W q + λ μ ⟹ L = L q + λ μ \lambda W = \lambda W_q + \frac{\lambda}{\mu} \quad \implies \quad L = L_q + \frac{\lambda}{\mu} λW = λ W q ​ + μ λ ​ ⟹ L = L q ​ + μ λ ​

For a single-server system, λ / μ = ρ \lambda / \mu = \rho λ / μ = ρ , so L = L q + ρ L = L_q + \rho L = L q ​ + ρ . The quantity λ / μ \lambda / \mu λ / μ represents the average number of entities currently being served.

Universal Scope of Little's Law

Little's Law is exceptionally robust because it requires no assumptions regarding:

  • The arrival distribution (Poisson, deterministic, or arbitrary general arrivals)
  • The service distribution (exponential, normal, uniform, or empirical)
  • The number of servers ( s = 1 , 2 , … , ∞ s = 1, 2, \dots, \infty s = 1 , 2 , … , ∞ )
  • The queue discipline (FCFS, LCFS, priority, or processor sharing)

As long as the system is stable and in steady state, Little's Law holds across the entire system, across the queue alone, or across any designated sub-process.

5. The M / M / 1 M/M/1 M / M /1 Queuing Model

The M / M / 1 M/M/1 M / M /1 queue is the foundational analytical model in queuing theory. It assumes Poisson arrivals (rate λ \lambda λ ), exponential service times (rate μ \mu μ ), 1 server, infinite capacity ( K = ∞ K = \infty K = ∞ ), infinite calling population, and FCFS queue discipline.

State Probabilities

Let P n P_n P n ​ denote the steady-state probability that there are exactly n n n customers in the system ( n = 0 , 1 , 2 , … n = 0, 1, 2, \dots n = 0 , 1 , 2 , … ): P 0 = 1 − ρ = 1 − λ μ P_0 = 1 - \rho = 1 - \frac{\lambda}{\mu} P 0 ​ = 1 − ρ = 1 − μ λ ​ P n = ( 1 − ρ ) ρ n = ( 1 − λ μ ) ( λ μ ) n P_n = (1 - \rho) \rho^n = \left( 1 - \frac{\lambda}{\mu} \right) \left( \frac{\lambda}{\mu} \right)^n P n ​ = ( 1 − ρ ) ρ n = ( 1 − μ λ ​ ) ( μ λ ​ ) n

Idle Probability ( P 0 P_0 P 0 ​ ) : The probability that the system is completely empty (server is idle).

Server Busy Probability : 1 − P 0 = ρ = λ / μ 1 - P_0 = \rho = \lambda / \mu 1 − P 0 ​ = ρ = λ / μ .

Probability of n n n or More Customers in System : Summing the geometric series from n n n to ∞ \infty ∞ : P ( N ≥ n ) = ∑ k = n ∞ P k = ρ n = ( λ μ ) n P(N \ge n) = \sum_{k=n}^{\infty} P_k = \rho^n = \left( \frac{\lambda}{\mu} \right)^n P ( N ≥ n ) = k = n ∑ ∞ ​ P k ​ = ρ n = ( μ λ ​ ) n

Operating Characteristics Formulas

From the state probabilities, the standard performance metrics provided in the NCEES Reference Handbook are derived:

Performance Metric Formula in Terms of ρ \rho ρ Formula in Terms of λ \lambda λ and μ \mu μ

Average number in system ( L L L ) L = ρ 1 − ρ L = \frac{\rho}{1 - \rho} L = 1 − ρ ρ ​ L = λ μ − λ L = \frac{\lambda}{\mu - \lambda} L = μ − λ λ ​

Average number in queue ( L q L_q L q ​ ) L q = ρ 2 1 − ρ = L − ρ L_q = \frac{\rho^2}{1 - \rho} = L - \rho L q ​ = 1 − ρ ρ 2 ​ = L − ρ L q = λ 2 μ ( μ − λ ) L_q = \frac{\lambda^2}{\mu(\mu - \lambda)} L q ​ = μ ( μ − λ ) λ 2 ​

Average time in system ( W W W ) W = 1 μ ( 1 − ρ ) W = \frac{1}{\mu(1 - \rho)} W = μ ( 1 − ρ ) 1 ​ W = 1 μ − λ W = \frac{1}{\mu - \lambda} W = μ − λ 1 ​

Average waiting time in queue ( W q W_q W q ​ ) W q = ρ μ ( 1 − ρ ) W_q = \frac{\rho}{\mu(1 - \rho)} W q ​ = μ ( 1 − ρ ) ρ ​ W q = λ μ ( μ − λ ) W_q = \frac{\lambda}{\mu(\mu - \lambda)} W q ​ = μ ( μ − λ ) λ ​

Hyperbolic Growth as ρ → 1 \rho \to 1 ρ → 1 : Notice that as server utilization ρ \rho ρ approaches 1.0 1.0 1.0 , the denominator ( μ − λ ) → 0 (\mu - \lambda) \to 0 ( μ − λ ) → 0 . As a result, L , L q , W , L, L_q, W, L , L q ​ , W , and W q W_q W q ​ do not grow linearly—they explode hyperbolically toward infinity. Operating a system at 95 % 95\% 95% utilization produces an average queue nineteen times larger than operating at 50 % 50\% 50% utilization ( L = 0.95 / 0.05 = 19 L = 0.95 / 0.05 = 19 L = 0.95/0.05 = 19 vs. L = 0.50 / 0.50 = 1 L = 0.50 / 0.50 = 1 L = 0.50/0.50 = 1 ).

6. Multi-Server Systems: The M / M / s M/M/s M / M / s Queuing Model

In an M / M / s M/M/s M / M / s queue, entities arrive according to a Poisson process with rate λ \lambda λ and are served by s s s parallel identical servers, each with an exponential service rate μ \mu μ . If an arriving entity finds any server idle, service begins immediately. If all s s s servers are busy, arriving entities wait in a single shared queue.

State Probabilities and Idle Probability ( P 0 P_0 P 0 ​ )

The state-dependent service rate μ n \mu_n μ n ​ reflects that when n < s n < s n < s customers are present, only n n n servers are active: μ n = { n μ , 0 ≤ n < s s μ , n ≥ s \mu_n = \begin{cases} n\mu, & 0 \le n < s \\ s\mu, & n \ge s \end{cases} μ n ​ = { n μ , s μ , ​ 0 ≤ n < s n ≥ s ​

The probability that the system is completely idle ( P 0 P_0 P 0 ​ ) is: P 0 = [ ∑ n = 0 s − 1 ( λ / μ ) n n ! + ( λ / μ ) s s ! ( 1 1 − λ s μ ) ] − 1 = [ ∑ n = 0 s − 1 r n n ! + r s s ! ( 1 1 − ρ ) ] − 1 P_0 = \left[ \sum_{n=0}^{s-1} \frac{(\lambda / \mu)^n}{n!} + \frac{(\lambda / \mu)^s}{s!} \left( \frac{1}{1 - \frac{\lambda}{s\mu}} \right) \right]^{-1} = \left[ \sum_{n=0}^{s-1} \frac{r^n}{n!} + \frac{r^s}{s!} \left( \frac{1}{1 - \rho} \right) \right]^{-1} P 0 ​ = [ n = 0 ∑ s − 1 ​ n ! ( λ / μ ) n ​ + s ! ( λ / μ ) s ​ ( 1 − s μ λ ​ 1 ​ ) ] − 1 = [ n = 0 ∑ s − 1 ​ n ! r n ​ + s ! r s ​ ( 1 − ρ 1 ​ ) ] − 1

Where r = λ / μ r = \lambda / \mu r = λ / μ is the offered load and ρ = λ / ( s μ ) < 1 \rho = \lambda / (s\mu) < 1 ρ = λ / ( s μ ) < 1 is the server utilization.

Probability That Arriving Entities Must Wait (Erlang C Formula)

An arrival must wait in the queue if all s s s servers are occupied ( N ≥ s N \ge s N ≥ s ). This probability is given by the Erlang C delay formula: P ( N ≥ s ) = r s s ! ( 1 − ρ ) P 0 = ( λ / μ ) s s ! ( 1 − λ s μ ) P 0 P(N \ge s) = \frac{r^s}{s! (1 - \rho)} P_0 = \frac{(\lambda / \mu)^s}{s! \left( 1 - \frac{\lambda}{s\mu} \right)} P_0 P ( N ≥ s ) = s ! ( 1 − ρ ) r s ​ P 0 ​ = s ! ( 1 − s μ λ ​ ) ( λ / μ ) s ​ P 0 ​

Multi-Server Operating Characteristics

From P 0 P_0 P 0 ​ , the expected queue length is: L q = r s ρ s ! ( 1 − ρ ) 2 P 0 = ( λ / μ ) s ( λ s μ ) s ! ( 1 − λ s μ ) 2 P 0 L_q = \frac{r^s \rho}{s! (1 - \rho)^2} P_0 = \frac{(\lambda / \mu)^s \left( \frac{\lambda}{s\mu} \right)}{s! \left( 1 - \frac{\lambda}{s\mu} \right)^2} P_0 L q ​ = s ! ( 1 − ρ ) 2 r s ρ ​ P 0 ​ = s ! ( 1 − s μ λ ​ ) 2 ( λ / μ ) s ( s μ λ ​ ) ​ P 0 ​

Once L q L_q L q ​ is determined, Little's Law immediately yields the remaining metrics without requiring further integration: W q = L q λ W_q = \frac{L_q}{\lambda} W q ​ = λ L q ​ ​ W = W q + 1 μ W = W_q + \frac{1}{\mu} W = W q ​ + μ 1 ​ L = L q + λ μ = λ W L = L_q + \frac{\lambda}{\mu} = \lambda W L = L q ​ + μ λ ​ = λW

Pooled Capacity Principle : A single queue feeding s s s parallel servers ( M / M / s M/M/s M / M / s ) always outperforms s s s independent single-server queues ( s × M / M / 1 s \times M/M/1 s × M / M /1 ) each handling an arrival rate of λ / s \lambda / s λ / s . Server pooling eliminates the scenario where one server sits idle while customers wait in line at an adjacent busy server.

7. Economic Analysis of Queuing Systems

Industrial engineers frequently evaluate queuing models to balance the economic trade-off between the cost of providing service capacity and the cost of delays. Cost ($/hr)

  • ▲
  • │ Total Cost Curve TC(s) = s*Cs + L*Cw
  • │ \ /
  • │ \ Optimal /
  • │ \_ s* _/
  • │ \_______/ Service Cost: s * Cs
  • │ │ /
  • │ Waiting Cost: L * Cw │ /
  • │ \ │ /
  • │ \ │ /
  • │ \_ │ /
  • │ \______________ │ _______/
  • └───────────────────────────┴────────────────────────► Number of Servers (s)
  • Total Expected Cost Formulation

The total expected cost per unit time ( T C TC TC ) as a function of the number of servers s s s is: TC ( s ) = s ⋅ C s + L ⋅ C w \text{TC}(s) = s \cdot C_s + L \cdot C_w TC ( s ) = s ⋅ C s ​ + L ⋅ C w ​

Or, if waiting costs apply strictly to entities waiting in the queue rather than those in active service: TC q ( s ) = s ⋅ C s + L q ⋅ C w \text{TC}_q(s) = s \cdot C_s + L_q \cdot C_w TC q ​ ( s ) = s ⋅ C s ​ + L q ​ ⋅ C w ​

Where:

C s C_s C s ​ : Cost of operating each server per unit time (including capital equipment leasing, operator wages, space, and energy)

C w C_w C w ​ : Cost of entity waiting time per unit time (including in-process inventory holding cost, expediting charges, customer ill-will, or machine downtime penalties)

Because s s s is an integer ( s = 1 , 2 , 3 , … s = 1, 2, 3, \dots s = 1 , 2 , 3 , … ), finding the optimal number of servers s ∗ s^* s ∗ requires marginal analysis:

Compute TC ( s ) \text{TC}(s) TC ( s ) for increasing integer values of s s s starting from the minimum stable value ( s > λ / μ s > \lambda / \mu s > λ / μ ).

The optimal configuration s ∗ s^* s ∗ satisfies the discrete optimality condition: TC ( s ∗ − 1 ) ≥ TC ( s ∗ ) ≤ TC ( s ∗ + 1 ) \text{TC}(s^* - 1) \ge \text{TC}(s^*) \le \text{TC}(s^* + 1) TC ( s ∗ − 1 ) ≥ TC ( s ∗ ) ≤ TC ( s ∗ + 1 )

  • 8. Step-by-Step Worked Engineering Examples
  • Worked Example 9.1.1: M / M / 1 M/M/1 M / M /1 Tool Regrinding Station Analysis

Problem : An automated CNC tooling bay routes dull end-mills to a dedicated regrinding technician. Cutting tools arrive according to a Poisson process at an average rate of λ = 6 \lambda = 6 λ = 6 tools per hour. The technician re-sharpens tools with exponentially distributed service times, averaging 8 8 8 minutes per tool.

Determine the technician's utilization ρ \rho ρ and the probability that the station is completely idle P 0 P_0 P 0 ​ .

Compute the average number of tools waiting in the queue ( L q L_q L q ​ ) and the average number of tools in the entire station ( L L L ).

Calculate the average waiting time a tool spends in queue ( W q W_q W q ​ ) and total turnaround time in the bay ( W W W ).

What is the probability that there are more than 3 tools in the bay?

Solution :

Convert Parameters to Consistent Hourly Rates :

Arrival rate: λ = 6 \lambda = 6 λ = 6 tools/hour

Mean service time: E ( S ) = 8 E(S) = 8 E ( S ) = 8 minutes = 8 / 60 = 2 / 15 = 8 / 60 = 2 / 15 = 8/60 = 2/15 hours

Service rate: μ = 1 / E ( S ) = 60 / 8 = 7.5 \mu = 1 / E(S) = 60 / 8 = 7.5 μ = 1/ E ( S ) = 60/8 = 7.5 tools/hour

Utilization factor: ρ = λ μ = 6.0 7.5 = 0.80 ( 80.0 % busy ) \rho = \frac{\lambda}{\mu} = \frac{6.0}{7.5} = 0.80 \quad (80.0\% \text{ busy}) ρ = μ λ ​ = 7.5 6.0 ​ = 0.80 ( 80.0% busy )

Since ρ = 0.80 < 1 \rho = 0.80 < 1 ρ = 0.80 < 1 , the system is stable in steady state.

Idle probability: P 0 = 1 − ρ = 1 − 0.80 = 0.20 ( 20.0 % idle ) P_0 = 1 - \rho = 1 - 0.80 = 0.20 \quad (20.0\% \text{ idle}) P 0 ​ = 1 − ρ = 1 − 0.80 = 0.20 ( 20.0% idle )

Calculate Average Tool Inventory ( L L L and L q L_q L q ​ ) :

Average number in system: L = λ μ − λ = 6.0 7.5 − 6.0 = 6.0 1.5 = 4.0 tools L = \frac{\lambda}{\mu - \lambda} = \frac{6.0}{7.5 - 6.0} = \frac{6.0}{1.5} = 4.0 \text{ tools} L = μ − λ λ ​ = 7.5 − 6.0 6.0 ​ = 1.5 6.0 ​ = 4.0 tools

Average number in queue: L q = λ 2 μ ( μ − λ ) = 6.0 2 7.5 ( 1.5 ) = 36.0 11.25 = 3.2 tools L_q = \frac{\lambda^2}{\mu(\mu - \lambda)} = \frac{6.0^2}{7.5(1.5)} = \frac{36.0}{11.25} = 3.2 \text{ tools} L q ​ = μ ( μ − λ ) λ 2 ​ = 7.5 ( 1.5 ) 6. 0 2 ​ = 11.25 36.0 ​ = 3.2 tools

Check via decomposition: L = L q + ρ = 3.2 + 0.80 = 4.0 L = L_q + \rho = 3.2 + 0.80 = 4.0 L = L q ​ + ρ = 3.2 + 0.80 = 4.0 tools (Consistent).

Calculate Average Turnaround Times ( W q W_q W q ​ and W W W ) :

Average waiting time in queue: W q = L q λ = 3.2 6.0 = 0.5333 hours = 32.0 minutes W_q = \frac{L_q}{\lambda} = \frac{3.2}{6.0} = 0.5333 \text{ hours} = 32.0 \text{ minutes} W q ​ = λ L q ​ ​ = 6.0 3.2 ​ = 0.5333 hours = 32.0 minutes

Average total time in system: W = L λ = 4.0 6.0 = 0.6667 hours = 40.0 minutes W = \frac{L}{\lambda} = \frac{4.0}{6.0} = 0.6667 \text{ hours} = 40.0 \text{ minutes} W = λ L ​ = 6.0 4.0 ​ = 0.6667 hours = 40.0 minutes

Check: W = W q + 1 / μ = 32.0 + 8.0 = 40.0 W = W_q + 1/\mu = 32.0 + 8.0 = 40.0 W = W q ​ + 1/ μ = 32.0 + 8.0 = 40.0 minutes (Consistent).

Probability of More Than 3 Tools in System ( N > 3 N > 3 N > 3 ) : P ( N > 3 ) = P ( N ≥ 4 ) = ρ 4 = ( 0.80 ) 4 = 0.4096 ≈ 41.0 % P(N > 3) = P(N \ge 4) = \rho^4 = (0.80)^4 = 0.4096 \approx 41.0\% P ( N > 3 ) = P ( N ≥ 4 ) = ρ 4 = ( 0.80 ) 4 = 0.4096 ≈ 41.0%

Worked Example 9.1.2: Economic Optimization of Single- vs. Dual-Server Configuration

Problem : A circuit board assembly plant generates test failures requiring diagnostic rework at an average rate of λ = 8 \lambda = 8 λ = 8 boards per hour (Poisson arrivals). Management is evaluating two alternatives:

Alternative A : Keep the current single technician ( s = 1 s = 1 s = 1 ) who processes μ = 10 \mu = 10 μ = 10 boards per hour (exponential service times).

Alternative B : Add a second identical technician to create an M / M / 2 M/M/2 M / M /2 pooled rework station ( s = 2 s = 2 s = 2 , each with μ = 10 \mu = 10 μ = 10 boards per hour).

Technician operating cost is C s = $ 40.00 C_s = \text{\textdollar}40.00 C s ​ = $ 40.00 per hour per technician. Work-in-process (WIP) waiting cost is C w = $ 60.00 C_w = \text{\textdollar}60.00 C w ​ = $ 60.00 per hour per board in the system. Which alternative minimizes total hourly cost?

Solution :

Evaluate Alternative A ( M / M / 1 M/M/1 M / M /1 ) :

λ = 8 \lambda = 8 λ = 8 boards/hr, μ = 10 \mu = 10 μ = 10 boards/hr, ρ = 8 / 10 = 0.80 \rho = 8 / 10 = 0.80 ρ = 8/10 = 0.80 .

Average boards in system: L A = λ μ − λ = 8 10 − 8 = 8 2 = 4.0 boards L_A = \frac{\lambda}{\mu - \lambda} = \frac{8}{10 - 8} = \frac{8}{2} = 4.0 \text{ boards} L A ​ = μ − λ λ ​ = 10 − 8 8 ​ = 2 8 ​ = 4.0 boards

Total hourly cost for Alternative A: TC A = s ⋅ C s + L A ⋅ C w = 1 ( $ 40.00 ) + 4.0 ( $ 60.00 ) = $ 40.00 + $ 240.00 = $ 280.00 / hour \text{TC}_A = s \cdot C_s + L_A \cdot C_w = 1(\$40.00) + 4.0(\$60.00) = \$40.00 + \$240.00 = \$280.00 / \text{hour} TC A ​ = s ⋅ C s ​ + L A ​ ⋅ C w ​ = 1 ( $40.00 ) + 4.0 ( $60.00 ) = $40.00 + $240.00 = $280.00/ hour

Evaluate Alternative B ( M / M / 2 M/M/2 M / M /2 ) :

Offered load: r = λ / μ = 8 / 10 = 0.80 r = \lambda / \mu = 8 / 10 = 0.80 r = λ / μ = 8/10 = 0.80 .

System utilization: ρ = λ / ( s μ ) = 8 / ( 2 × 10 ) = 0.40 \rho = \lambda / (s\mu) = 8 / (2 \times 10) = 0.40 ρ = λ / ( s μ ) = 8/ ( 2 × 10 ) = 0.40 .

Calculate P 0 P_0 P 0 ​ for s = 2 s = 2 s = 2 : P 0 = [ ∑ n = 0 1 r n n ! + r 2 2 ! ( 1 1 − ρ ) ] − 1 = [ 0.8 0 0 ! + 0.8 1 1 ! + 0.8 2 2 ( 1 − 0.40 ) ] − 1 P_0 = \left[ \sum_{n=0}^{1} \frac{r^n}{n!} + \frac{r^2}{2!} \left( \frac{1}{1 - \rho} \right) \right]^{-1} = \left[ \frac{0.8^0}{0!} + \frac{0.8^1}{1!} + \frac{0.8^2}{2(1 - 0.40)} \right]^{-1} P 0 ​ = [ n = 0 ∑ 1 ​ n ! r n ​ + 2 ! r 2 ​ ( 1 − ρ 1 ​ ) ] − 1 = [ 0 ! 0. 8 0 ​ + 1 ! 0. 8 1 ​ + 2 ( 1 − 0.40 ) 0. 8 2 ​ ] − 1 P 0 = [ 1 + 0.80 + 0.64 2 ( 0.60 ) ] − 1 = [ 1.80 + 0.64 1.20 ] − 1 = [ 1.80 + 0.5333 ] − 1 = 1 2.3333 = 3 7 ≈ 0.4286 P_0 = \left[ 1 + 0.80 + \frac{0.64}{2(0.60)} \right]^{-1} = \left[ 1.80 + \frac{0.64}{1.20} \right]^{-1} = \left[ 1.80 + 0.5333 \right]^{-1} = \frac{1}{2.3333} = \frac{3}{7} \approx 0.4286 P 0 ​ = [ 1 + 0.80 + 2 ( 0.60 ) 0.64 ​ ] − 1 = [ 1.80 + 1.20 0.64 ​ ] − 1 = [ 1.80 + 0.5333 ] − 1 = 2.3333 1 ​ = 7 3 ​ ≈ 0.4286

Calculate L q L_q L q ​ for Alternative B: L q = r 2 ρ 2 ! ( 1 − ρ ) 2 P 0 = ( 0.64 ) ( 0.40 ) 2 ( 0.60 ) 2 ( 3 7 ) = 0.256 0.72 ( 3 7 ) = ( 0.3556 ) ( 0.4286 ) ≈ 0.1524 boards L_q = \frac{r^2 \rho}{2!(1 - \rho)^2} P_0 = \frac{(0.64)(0.40)}{2(0.60)^2} \left( \frac{3}{7} \right) = \frac{0.256}{0.72} \left( \frac{3}{7} \right) = (0.3556)(0.4286) \approx 0.1524 \text{ boards} L q ​ = 2 ! ( 1 − ρ ) 2 r 2 ρ ​ P 0 ​ = 2 ( 0.60 ) 2 ( 0.64 ) ( 0.40 ) ​ ( 7 3 ​ ) = 0.72 0.256 ​ ( 7 3 ​ ) = ( 0.3556 ) ( 0.4286 ) ≈ 0.1524 boards

Calculate total boards in system L B L_B L B ​ : L B = L q + λ μ = 0.1524 + 0.80 = 0.9524 boards L_B = L_q + \frac{\lambda}{\mu} = 0.1524 + 0.80 = 0.9524 \text{ boards} L B ​ = L q ​ + μ λ ​ = 0.1524 + 0.80 = 0.9524 boards

Total hourly cost for Alternative B: TC B = 2 ( $ 40.00 ) + 0.9524 ( $ 60.00 ) = $ 80.00 + $ 57.14 = $ 137.14 / hour \text{TC}_B = 2(\$40.00) + 0.9524(\$60.00) = \$80.00 + \$57.14 = \$137.14 / \text{hour} TC B ​ = 2 ( $40.00 ) + 0.9524 ( $60.00 ) = $80.00 + $57.14 = $137.14/ hour

Economic Comparison : Hourly Cost Savings = TC A − TC B = $ 280.00 − $ 137.14 = $ 142.86 / hour \text{Hourly Cost Savings} = \text{TC}_A - \text{TC}_B = \$280.00 - \$137.14 = \$142.86 / \text{hour} Hourly Cost Savings = TC A ​ − TC B ​ = $280.00 − $137.14 = $142.86/ hour

Adding the second technician reduces total expected cost by over 51 % 51\% 51% , as the massive drop in WIP waiting cost (saving $ 182.86 \text{\textdollar}182.86 $ 182.86 /hr) far outweighs the additional $ 40.00 \text{\textdollar}40.00 $ 40.00 /hr labor cost.

9. NCEES Reference Handbook Tips & Realistic Exam Traps

Rate vs. Mean Time Trap : The most common error on the FE exam is entering mean service time directly as μ \mu μ . If a question states that service takes an average of 12 12 12 minutes, μ ≠ 12 \mu \neq 12 μ  = 12 . You must convert time to rate: μ = 60 / 12 = 5 \mu = 60 / 12 = 5 μ = 60/12 = 5 units per hour. Always verify that λ \lambda λ and μ \mu μ share identical time dimensions (e.g., both per hour or both per minute).

Queue vs. System Distinction : Carefully read whether the exam asks for:

Time in queue ( W q W_q W q ​ ) vs. total time in system ( W W W ). Note that W = W q + 1 / μ W = W_q + 1/\mu W = W q ​ + 1/ μ .

Entities in queue ( L q L_q L q ​ ) vs. entities in system ( L L L ). Note that L = L q + λ / μ L = L_q + \lambda/\mu L = L q ​ + λ / μ .

Stability Violation Trap : If a question provides λ = 15 \lambda = 15 λ = 15 arrivals/hr and a single server with μ = 12 \mu = 12 μ = 12 services/hr, ρ = 15 / 12 = 1.25 > 1 \rho = 15/12 = 1.25 > 1 ρ = 15/12 = 1.25 > 1 . Attempting to plug these numbers into the M / M / 1 M/M/1 M / M /1 formula L = λ / ( μ − λ ) L = \lambda / (\mu - \lambda) L = λ / ( μ − λ ) produces 15 / ( 12 − 15 ) = − 5 15 / (12 - 15) = -5 15/ ( 12 − 15 ) = − 5 customers, which is physically impossible. When ρ ≥ 1 \rho \ge 1 ρ ≥ 1 , the queue is unstable and has no steady-state solution.

Units in Little's Law : When computing L = λ W L = \lambda W L = λW , ensure the time units in W W W match the rate units in λ \lambda λ . If λ = 10 \lambda = 10 λ = 10 parts/hour and W = 15 W = 15 W = 15 minutes, convert W W W to 0.25 0.25 0.25 hours before multiplying ( L = 10 × 0.25 = 2.5 L = 10 \times 0.25 = 2.5 L = 10 × 0.25 = 2.5 parts).