Queuing Theory

ISE-183101: Penelitian Operasional III

Program Studi Sarjana Teknik Industri
Universitas Katolik Parahnyangan
 Reference

Winston, W.L. & Goldberg, J.B. (2004). Operations research: applications and algorithms (Vol 4).
BelmonteCalif Calif:Thomson/Brooks/Cole.

Chapter 20.1 – 20.2

Slide design by SlidesCarnival

Outline

Section 20.1 Section 20.2

Discuss some terminology that Review some distributions that
is often used to describe are needed to describe queuing
queues. model.

Section 20.3 Section 20.4 -20.11

Introduce the idea of birth- Examine several models of
death process queuing systems.
20.1 Some Queuing Terminology
The Input or Arrival Process

◉ The input process is usually called the arrival process.

◉ Arrivals are called customers.
◉ We assume that no more than one arrival can occur at a given instant.
◉ We assume that the arrival process is unaffected by the number of
customer present in the system.
◉ The arrival process may depend on the number of customers present:
○ when arrivals are drawn from a small population (finite source)
○ when the rate at which customers arrive at the facility decreases when the
facility becomes too crowded (the customer has balked).
The Output or Service Process

◉ The output process is often called the service process.

◉ We usually specify a probability distribution - the service time
distribution - which governs a customer’s service time.
◉ We assume that the service time distribution is independent of the
number of customers present.
◉ We study two arrangements of servers: servers in parallel and servers
in series
Parallel vs Series

◉ Servers are in parallel if all ◉ Servers are in series if a

servers provide the same type customer must pass through
of service and a customer need several servers before
only pass through one server to completing service.
complete service.
Examples of Queuing Systems
Queue Disciple

The queue discipline describes the method used to determine the

order in which customers are served.
 FCFS = First come first serve
 LCFS = Last come first serve
 SIRO = Service in random order
 Priority queuing discipline classifies each arrival into one of several
categories. Each category is then given a priority level, and within each
priority level, customers enter service on an FCFS basis.
20.2 Modelling Arrival & Service Processes
Modeling the Arrival Process

Assume at most one arrival can occur at a given instant of time.

t1 t2 t3

T1 T2
for example:
t1 = 3 ; t2 = 8 ; t3 = 15
Ti = ti+1 - ti interarrival time
T1 = 8 – 3 = 5
T2 = 15 – 8 = 7
Modeling the Arrival Process (2)

We assume that the 𝑇𝑖 ’s are independent, continuous random variables

described by the random variable A.
A negative interarrival time is impossible.
We also assume stationary interarrival times (often unrealistic).
We may often approximate reality by breaking the time of day into
segments (each segment then has the stationary assumption).
◉ Morning rush hour
◉ Midday segment
◉ Afternoon rush hour
Modeling the Arrival Process (3)

◉ Let 𝑎 𝑡 is density function of 𝑨

1
◉ 𝜆 = the arrival rate (units of arrivals per hour)  𝜆 = mean
interarrival time

◉ The most common choice for 𝑨 is the exponential distribution.

Why exponential distribution?

No-memory Property
◉ It implies that if we want to know the
probability distribution of the time until the
next arrival, then it does not matter how long
it has been since the last arrival.
◉ If we know that at least 𝑡 time units have
elapsed since the last arrival occurred, then the
distribution of the remaining time until the next
arrival ℎ doesn’t depend on 𝑡.
This means that to predict future arrival patterns, we need not
keep track of how long it has been since the last arrival.
Example
Relation Between Poisson and
Exponential Distribution

◉ 𝑵 is the number of arrivals

◉ A discrete random variable 𝑵 has a Poisson distribution with
parameter 𝜆 if, for 𝑛 = 0, 1, 2, . . . ,
◉ If 𝑵 is a Poisson random variable, it can be shown that
𝑬(𝑵) = 𝑽𝒂𝒓 𝑵 = 𝝀
Relation Between Poisson and
Exponential Distribution

◉ Define 𝑵𝑡 to be the number of arrivals to occur during any time interval of

length 𝑡.

◉ 𝑵𝒕 is Poisson with parameter 𝜆𝑡, 𝐸(𝑁𝑡) =𝑉𝑎𝑟 𝑁𝑡 =𝜆𝑡.

◉ An average of 𝜆𝑡 arrivals occur during a time interval of length 𝑡, so 𝜆 may be
thought of as the average number of arrivals per unit time, or the arrival rate.
 What assumption are required for interarrival times to be exponential?

1. Arrivals defined on non-overlapping time intervals are independent.

2. For small 𝑡 (and any value of 𝑡), the probability of one arrival occurring
between times 𝑡 and 𝑡 + ∆𝑡 is 𝜆∆𝑡 + 𝑜(∆𝑡), where 𝑜(∆𝑡) refers to any
quantity satisfying:
𝑜 ∆t
lim =0
∆t→0 ∆𝑡
Also, the probability of no arrival during the interval between 𝑡 and 𝑡 + ∆𝑡 is 1 −
𝜆∆𝑡 + 𝑜 ∆𝑡 , and the probability of more than one interval occurring between 𝑡
and 𝑡 + ∆𝑡 is 𝑜 ∆𝑡
Example 1

The number of boxes of pizza ordered per hour follows a

Poisson distribution, with an average of 30 pizzas per hour
being ordered.
1. Find the probability that exactly 60 pizzas are ordered
between 10 P.M. and 12 midnight.
2. Find the mean and standard deviation of the number of
pizzas ordered between 9 P.M. and 1A.M.
3. Find the probability that the time between two
consecutive orders is between 1 and 3 minutes.
 Example 1 (2)
1. The number of pizzas ordered between 10 P.M. and 12 midnight will follow a Poisson
distribution with parameter: 2(30) = 60.
𝑒 −60 6060
𝑃 𝑁𝑡 = 60 = = 0.051
60!
2. Let 𝑁𝑡 : number of pizza ordered between 9 PM and 1 AM.
𝜆 = 30
𝑡=4
𝐸 𝑁𝑡 = 𝑉𝑎𝑟 𝑁𝑡 = 𝜆𝑡 = 120

3. Let 𝑿 be the time (in minutes) between successive pizza orders. The mean number of
orders per minute is exponential with parameter or rate 30/60 =0.5 pizza per minute.
Thus, the pdf of the time between orders is 0.5𝑒 −0.5𝑡
Example 2

The time between arrivals of buses follow an exponential distribution

with a mean of 60 minutes
a. What is the probability that exactly four buses will arrive
during the next 2 hours?
b. That at least two buses will arrive during the next 2 hours?
c. That no buses will arrive during the next 2 hours?
d. A bus has just arrived. What is the probability that it will be
between 30 and 90 minutes before the next bus arrives
Example 2 (2)

a. Number of buses arriving in next two hours is Poisson with mean 2.

𝑒 −2 24
P(4 buses in next two hours) = = 0,09
4!

b. P(at least 2 buses) = 1 – P(0 buses) – P(1 bus) = 1 − 𝑒 −2 − 2𝑒 −2 = 0.59

𝑒 −2 20
c. P(0 buses) = = 𝑒 −2 = 0.14
0!

d. Arrival rate = 𝜆 = 1 bus/hour. Thus, the pdf of time between bus is 𝑒 −𝑡

P (0,5 ≤ A ≤ 1,5) = ∫ e-t dt = e-0,5 – e-1,5 = 0,38

23
 Erlang Distribution

If interarrival times do not appear to be exponential, they are often modeled by an

Erlang distribution. Erlang density has the following pdf :

Rate parameter = R and Shape parameter = k

 Erlang Distribution (2)
 For k = 1, the Erlang distribution is an exponential distribution with parameter R.

 As k increases, the Erlang distribution behaves more and more like a normal distribution.
For extremely large values of k, it approaches a random variable with zero variance
(constant interarrival time).

 The Erlang distribution has the same distribution as the random variable A1 + A2 +...+ Ak,
where each Ai is an exponential random variable with parameter k𝜆, and the Ai’s are
independent random variables.

 The interarrival process is equivalent to a customer going through k phases (each of

which has the no-memory property) before arriving. The shape parameter is often
referred to as the number of phases of the Erlang distribution.
 Modeling the Service Process

Assume that the service times of different customers are independent random variables and
that each customer’s service time is governed by a random variable S having a density
function s(t).

1
= mean service time for customer (hours per customer)
𝜇
𝜇 = servicerate(customersper hour)
Service times can be accurately modeled as exponential random variables.
Modeling the Service Process (2)

 What is the probability that the customer

who is waiting will be the last of the four
customers to complete service?
 One of customers 1–3 (say, customer 3)
will be the first to complete service. Then
customer 4 will enter service.
 By the no-memory property, customer 4’s
service time has the same distribution as
the remaining service times of customers 1
and 2 = 1/3.
 Modeling the Service Process (3)

The actual service time may be inconsistent with the no memory property
 Erlang distribution can be closely fitted to observed service times.

In certain situations, interarrival or service times may be modeled as

having zero variance; in this case, interarrival or service times are
considered to be deterministic.
 The Kendall-Lee Notation for
Queuing Systems

Each queuing system is described by six characteristics:

1/2/3/4/5/6
1. The nature of the arrival process
𝑀 = Interarrival times are independent, identically distributed (iid)
random variables having an exponential distribution.
𝐷 = Interarrival times are iid and deterministic.
𝐸𝑘 = Interarrival times are iid Erlangs with shape parameter k.
𝐺𝐼 = Interarrival times are iid and governed by some general
distribution
 The Kendall-Lee Notation for
Queuing Systems (2)

2. The nature of the service process

M = Service times are iid and exponentially distributed.
D = Service times are iid and deterministic.
Ek = Service times are iid Erlangs with shape parameter k.
G = Service times are iid and follow some general distribution.
3. The number of parallel servers
4. The queue discipline
◉ FCFS = First come, first served
◉ LCFS = Last come, first served
◉ SIRO = Service in random order
◉ GD = General queue discipline
 The Kendall-Lee Notation for
Queuing Systems (3)

5. The maximum allowable number of customers in the system (including

customers who are waiting and customers who are in service).
6. The size of the population from which customers are drawn.

In many important models 4/5/6 is GD/∞/∞. If this is the case, then 4/5/6 is
often omitted.
Example: M/E2/8/FCFS/10/∞
Represents: A health clinic with 8 doctors, exponential interarrival times,
two-phase Erlang service times, an FCFS queue discipline, and a total
capacity of 10 patients.
 Waiting is so unusual that many of us can't
stand in a queue for 30 seconds without
getting out our phones to check for messages
or to Google something
-Julian Baggini-