This action might not be possible to undo. Are you sure you want to continue?

# Chapter 5 1

Exponential Distribution &

Poisson Process

Memorylessness & other exponential

distribution properties; Poisson process;

Nonhomogeneous & compound P.P.’s

Chapter 5 2

Exponential Distribution: Basic Facts

• Density

• CDF

• MGF

• Mean

• Variance

( )

, 0

, 0

0, 0

x

e x

f x

x

λ

λ

λ

−

¦

≥

= >

´

<

¹

( )

1 , 0

0, 0

x

e x

F x

x

λ −

¦

− ≥

=

´

<

¹

( )

tX

t E e

t

λ

φ

λ

(

= =

¸ ¸

−

| |

1

E X

λ

=

| |

2

1

Var X

λ

=

Coefficient of variation

| |

| |

1

Var

E X

X

=

Chapter 5 3

Key Property: Memorylessness

• Reliability: Amount of time a component has been in

service has no effect on the amount of time until it fails

• Inter-event times: Amount of time since the last event

contains no information about the amount of time until the

next event

• Service times: Amount of remaining service time is

independent of the amount of service time elapsed so far

{ } { }

for all , 0 P X s t X t P X s s t > + > = > ≥

Chapter 5 4

Other Useful Properties

Competing Exponentials:

If X

1

and X

2

are independent

exponential r.v.’s with

parameters λ

1

and λ

2

, resp.,

then

(generalizes to any number

of competing r.v.’s)

Minimum of exponentials:

If X

1

, X

2

, …, X

n

are

independent exponential

r.v.’s where X

n

has parameter

λ

i

, then

min(X

1

, X

2

, …, X

n

) is

exponential w/parameter

λ

1

+ λ

2

+ … + λ

n

Sum of n independent exponential r.v.’s with common

parameter λ has a gamma distribution w/parameters (n, λ)

{ }

1

1 2

1 2

P X X

λ

λ λ

< =

+

Chapter 5 5

Counting Process

A stochastic process {N(t), t ≥ 0} is a counting process if N(t)

represents the total number of events that have occurred in

[0, t]

Then {N(t), t ≥ 0} must satisfy:

N(t) ≥ 0

N(t) is an integer for all t

If s < t, then N(s) ≤ N(t)

For s < t, N(t) - N(s) is the number of events that occur in

the interval (s, t].

Chapter 5 6

Stationary & Independent Increments

• A counting process has independent increments if, for any

That is, the numbers of events that occur in nonoverlapping

intervals are independent random variables.

• A counting process has stationary increments if the

distribution if, for any s < t, the distribution of N(t) – N(s)

depends only on the length of the time interval, t – s.

( ) ( ) ( ) ( )

0 , is independent of s t u v N t N s N v N u ≤ < ≤ < − −

Chapter 5 7

Poisson Process Definition 1

A counting process {N(t), t ≥ 0} is a Poisson process with

rate λ, λ > 0, if

N(0) = 0

The process has independent increments

The number of events in any interval of length t follows a

Poisson distribution with mean λt (therefore, it has

stationary increments), i.e.,

( ) ( ) { }

( )

, 0,1,...

!

n

t

e t

P N t s N s n n

n

λ

λ

−

+ − = = =

Chapter 5 8

Poisson Process Definition 2

A function f is said to be o(h) (“Little oh of h”) if

A counting process {N(t), t ≥ 0} is a Poisson process with rate

λ, λ > 0, if

N(0) = 0

The process has stationary and independent increments

Definitions 1 and 2 are equivalent!

( )

0

lim 0

h

f h

h

→

=

( ) { } ( )

1 P N h h o h λ = = +

( ) { } ( )

2 P N h o h ≥ =

Chapter 5 9

Interarrival and Waiting Times

The times between arrivals

T

1

, T

2

, … are independent

exponential r.v.’s with mean

1/λ:

The (total) waiting time until

the nth event has a gamma

dist’n:

t

N(t)

T

2

T

1

T

3

T

4

S

1

S

2

S

3

S

4

{ } ( ) { }

1

0

t

P T t P N t e

λ −

> = = =

{ }

2 1

t

P T t T s e

λ −

> = =

1

n

n i

i

S T

=

=

∑

Chapter 5 10

Other Poisson Process Properties

Poisson Splitting:

Suppose {N(t), t ≥ 0} is a P.P. with rate λ, and suppose that

each time an event occurs, it is classified as type I with

probability p and type II with probability 1-p,

independently of all other events. Let N

1

(t) and N

2

(t),

respectively, be the number of type I and type II events up

to time t.

Then {N

1

(t), t ≥ 0} and {N

2

(t), t ≥ 0} are independent

Poisson processes with respective rates λp and λ(1-p).

Chapter 5 11

Other Poisson Process Properties

Competing Poisson Processes:

Suppose {N

1

(t), t ≥ 0} and {N

2

(t), t ≥ 0} are independent

Poisson processes with respective rates λ

1

and λ

2

.

Let S

n

i

be the time of the nth event of process i, i = 1,2.

{ }

1

1

1 2

1 1

1 2 1 2

1

k n m k

n m

n m

k n

n m

P S S

k

λ λ

λ λ λ λ

+ − −

+ −

=

+ −

| | | | | |

< =

| | |

+ +

\ .\ . \ .

∑

Chapter 5 12

Other Poisson Process Properties

If Y

1

, Y

2

, …, Y

n

are random variables, then Y

(1)

, Y

(2)

, …, Y

(n)

are their order statistics if Y

(k)

is the kth smallest value

among Y

1

, Y

2

, …, Y

n

, k = 1, …, n.

Conditional Distribution of Arrival Times:

Suppose {N(t), t ≥ 0} is a Poisson process with rate λ and

for some time t we know that N(t) = n. Then the arrival

times S

1

, S

2

, …, S

n

have the same conditional distribution

as the order statistics of n independent uniform random

variables on (0, t).

Chapter 5 13

Nonhomogeneous Poisson Process

A counting process {N(t), t ≥ 0} is a nonhomogeneous Poisson

process with intensity function λ(t), t ≥ 0, if:

N(0) = 0

The process has independent increments (not stationary incr.)

Let

Then

( ) ( ) { } ( ) ( )

1 P N t h N t t h o h λ + − = = +

( ) ( ) { } ( )

2 P N t h N t o h + − ≥ =

( ) ( )

0

t

m t y dy λ =

∫

( ) ( ) { }

( ) ( )

( ) ( ) ( )

, 0,1,...

!

n

m s t m s

m s t m s

P N t s N s n e n

n

− + − (

¸ ¸

+ −

+ − = = =

Chapter 5 14

Compound Poisson Process

A counting process {X(t), t ≥ 0} is a compound Poisson process

if:

where {N(t), t ≥ 0} is a Poisson process and {Y

i

, i = 1, 2, …}

are independent, identically distributed r.v.’s that are

independent of {N(t), t ≥ 0}.

By conditioning on N(t), we can obtain:

( )

( )

1

, 0

N t

i

i

X t Y t

=

= ≥

∑

( ) | |

( )

1

2

1

Var

E X t tE Y

X t tE Y

λ

λ

= (

¸ ¸

(

= (

¸ ¸

¸ ¸

x < 0 • MGF φ ( t ) = E e = λ −t 1 • Mean E [ X ] = λ • Variance Var [ X ] = 12 tX Chapter 5 λ Coefficient of variation E[X ] Var [ X ] =1 2 λ . x < 0 1 − e − λ x . x ≥ 0 • CDF F ( x ) = 0. x ≥ 0 .Exponential Distribution: Basic Facts λ e − λ x . λ >0 • Density f ( x ) = 0.

t ≥ 0 • Reliability: Amount of time a component has been in service has no effect on the amount of time until it fails • Inter-event times: Amount of time since the last event contains no information about the amount of time until the next event • Service times: Amount of remaining service time is independent of the amount of service time elapsed so far Chapter 5 3 .Key Property: Memorylessness P { X > s + t X > t} = P { X > s} for all s.

’s) Chapter 5 . ….’s with common parameter λ has a gamma distribution w/parameters (n. then Minimum of exponentials: If X1. resp..Other Useful Properties Sum of n independent exponential r.v. X2 . λ) Competing Exponentials: If X1 and X2 are independent exponential r.’s where Xn has parameter λi. then min(X1. X2 .v.’s with parameters λ1 and λ2. Xn) is exponential w/parameter λ1 + λ2 + … + λn 4 P { X1 < X 2} = λ1 + λ2 λ1 (generalizes to any number of competing r. Xn are independent exponential r.v. ….v.

t ≥ 0} must satisfy: N(t) ≥ 0 N(t) is an integer for all t If s < t. t ≥ 0} is a counting process if N(t) represents the total number of events that have occurred in [0. Chapter 5 5 . t] Then {N(t).Counting Process A stochastic process {N(t). t]. N(t) .N(s) is the number of events that occur in the interval (s. then N(s) ≤ N(t) For s < t.

Stationary & Independent Increments • A counting process has independent increments if. for any 0 ≤ s < t ≤ u < v. t – s. the distribution of N(t) – N(s) depends only on the length of the time interval. N ( t ) − N ( s ) is independent of N ( v ) − N ( u ) That is. Chapter 5 6 . for any s < t. the numbers of events that occur in nonoverlapping intervals are independent random variables. • A counting process has stationary increments if the distribution if.

i. λ > 0.. n = 0.1.. P { N ( t + s ) − N ( s ) = n} = n! Chapter 5 7 . n − λt e ( λt ) . it has stationary increments).Poisson Process Definition 1 A counting process {N(t). if N(0) = 0 The process has independent increments The number of events in any interval of length t follows a Poisson distribution with mean λt (therefore.e. t ≥ 0} is a Poisson process with rate λ...

t ≥ 0} is a Poisson process with rate λ. λ > 0. if N(0) = 0 The process has stationary and independent increments P { N ( h ) = 1} = λ h + o ( h ) P { N ( h ) ≥ 2} = o ( h ) Definitions 1 and 2 are equivalent! Chapter 5 8 .f (h) =0 A function f is said to be o(h) (“Little oh of h”) if lim h →0 h Poisson Process Definition 2 A counting process {N(t).

T2.v.’s with mean 1/λ: P {T1 > t} = P { N ( t ) = 0} = e − λt P {T2 > t T1 = s} = e − λt S1 S2 T1 T2 T3 S3 S4 T4 t The (total) waiting time until the nth event has a gamma n dist’n: S = T n ∑ i =1 i Chapter 5 9 .Interarrival and Waiting Times N(t) The times between arrivals T1. … are independent exponential r.

and suppose that each time an event occurs.Other Poisson Process Properties Poisson Splitting: Suppose {N(t). it is classified as type I with probability p and type II with probability 1-p. t ≥ 0} is a P. be the number of type I and type II events up to time t. Let N1(t) and N2(t). t ≥ 0} are independent Poisson processes with respective rates λp and λ(1-p). t ≥ 0} and {N2(t). independently of all other events. respectively. Chapter 5 10 . Then {N1(t). with rate λ.P.

t ≥ 0} and {N2(t).2. i = 1. t ≥ 0} are independent Poisson processes with respective rates λ1 and λ2. Let Sni be the time of the nth event of process i. P {S < S 1 n 2 m }= ∑ k =n n + m −1 n + m − 1 λ1 λ1 k λ1 + λ2 λ1 + λ2 k n + m −1− k Chapter 5 11 .Other Poisson Process Properties Competing Poisson Processes: Suppose {N1(t).

…. Yn are random variables. t ≥ 0} is a Poisson process with rate λ and for some time t we know that N(t) = n. Y(n) are their order statistics if Y(k) is the kth smallest value among Y1. …. Chapter 5 12 . S2. Y2. Y(2). t). …. Then the arrival times S1. Y2. ….Other Poisson Process Properties If Y1. then Y(1). n. Sn have the same conditional distribution as the order statistics of n independent uniform random variables on (0. …. Yn. Conditional Distribution of Arrival Times: Suppose {N(t). k = 1.

.1. if: N(0) = 0 The process has independent increments (not stationary incr.. t ≥ 0} is a nonhomogeneous Poisson process with intensity function λ(t).Nonhomogeneous Poisson Process A counting process {N(t). t ≥ 0. P { N ( t + s ) − N ( s ) = n} = e n! Chapter 5 13 t .. n = 0.) P { N ( t + h ) − N ( t ) = 1} = λ ( t ) h + o ( h ) P { N ( t + h ) − N ( t ) ≥ 2} = o ( h ) Let m ( t ) = λ ( y )dy ∫0 n Then − m( s + t ) − m( s ) ( m ( s + t ) − m ( s ) ) .

identically distributed r.Compound Poisson Process A counting process {X(t).’s that are independent of {N(t). t ≥ 0} is a compound Poisson process if: N (t ) X ( t ) = ∑ Yi . By conditioning on N(t). t ≥ 0} is a Poisson process and {Yi. i = 1. t ≥ 0}. 2. …} are independent. t ≥ 0 i =1 where {N(t). we can obtain: E X ( t ) = λ tE [Y1 ] 2 Var X ( t ) = λtE Y1 Chapter 5 14 .v.