Queuing Theory

Why it matters

Queuing Theory is essential in industrial engineering for optimizing processes involving waiting lines, such as production lines, customer service, and logistics. By understanding and managing queues, businesses can improve efficiency, reduce wait times, and enhance customer satisfaction.

Key ideas

• Queue: A line of waiting items or people. In industrial settings, this could be products on an assembly line or customers waiting for service.
• Arrival Rate (λ): The rate at which items or customers arrive at the queue, typically measured in items per unit time.
• Service Rate (μ): The rate at which items or customers are serviced and leave the queue, also measured in items per unit time.
• Queue Discipline: The rule by which items are selected from the queue for service, such as First-In-First-Out (FIFO).
• Utilization (ρ): The fraction of time the service facility is busy, calculated as ρ = λ / μ.
• Little’s Law: Relates the average number of items in the system (L), the average arrival rate (λ), and the average time an item spends in the system (W) as L = λW.

Formulas

• ρ = λ / μ
  • ρ: Utilization (dimensionless)
  • λ: Arrival rate (items/s)
  • μ: Service rate (items/s)
• L = λW
  • L: Average number of items in the system (dimensionless)
  • λ: Arrival rate (items/s)
  • W: Average time an item spends in the system (s)

For Little’s law, L and W must refer to the same stable system boundary and λ to its effective throughput. Queue-only quantities satisfy L_q = λW_q. Arrival and service rates alone do not determine waiting time without distribution and server assumptions. For an M/M/1 model assume Poisson arrivals, exponential independent service times, one server, FIFO, unlimited waiting space and λ < μ. Then W = 1/(μ-λ), L = λ/(μ-λ), and W_q = W-1/μ.

Worked example

Given for an M/M/1 queue in steady state:

• Arrival rate, λ = 5 items/min
• Service rate, μ = 8 items/min

1. Calculate Utilization (ρ):

   Formula: ρ = λ / μ

   Calculation: ρ = 5 / 8 = 0.625

   Utilization is 0.625, meaning the service facility is busy 62.5% of the time.

2. Compute time in system from the M/M/1 model: W = 1/(8-5) = 1/3 min = 20 s.

3. Apply Little’s law: L = 5(1/3) = 1.6667 items.

4. Mean service time = 1/8 min = 7.5 s, so W_q = 20-7.5 = 12.5 s and L_q = 5(12.5/60) = 1.0417 items.

Answer: Utilization 62.5%, mean number in system 1.667, and mean queue wait 12.5 s under the stated M/M/1 assumptions. An arbitrary assumed W should not be substituted as though it follows from λ and μ.

Common mistakes

• Confusing arrival rate (λ) with service rate (μ).
• Misapplying Little’s Law by not ensuring consistent units.
• Ignoring the queue discipline, which can affect the analysis.

For GATE ME

Questions often involve calculating utilization, average number of items in the system, or average waiting time using given arrival and service rates. Practice problems involving Little’s Law and understanding different queue disciplines.

Quick check

1. What is the utilization if λ = 10 items/min and μ = 15 items/min?
2. How does Little’s Law relate L, λ, and W?
3. What is the queue discipline in a typical supermarket checkout?

Answers: 1. 0.6667; 2. L = λW; 3. FIFO