C

Course
O
Outline
tli
Introduction to Transportation
Highway Users and their Performance
Geometric Design
Pavement Design
Speed Studies - Project
Traffic
Queuing
Intersections
Level of Service in Highways and Intersections
Department of Civil and Environmental Engineering

Lecture # 11

Previous class

Basic Concepts
a.
b
b.
c.
d
d.
e.

Flow Rate
S
Spacing
i
Headway
Speed 2 types
Density

Relationships
Graphs

Department of Civil and Environmental Engineering

Lecture # 11

Relationships

n
q=
t

t =

hi

q=

i=1

n
n

i=1

1
us =
1 n 1

n i =1 (l t i )
Department of Civil and Environmental Engineering

n q
k= =
l u
Lecture # 11

q = uk

Consider a linear relationship

p between speed
p
and density:
y

k
u = u f 1
k
j

u
q = k j u

uf

Department of Civil and Environmental Engineering

Lecture # 11

MODELS OF TRAFFIC FLOW

Traffic
T ffi is
i rarely
l uniformly
if
l di
distributed
t ib t d
equal time between arriving vehicles or headways?

Must make some assumption for arrival patterns

(distribution)

Department of Civil and Environmental Engineering

Lecture # 11

Poisson Model
Approximation of non-uniform flow

(
t ) e
P ( n) =
n

n!

Where:
P(n) = probability of having n vehicles arrive in time t,
t
t = duration of the time interval over which vehicles
are counted,
= average vehicle flow or arrival rate in vehicles
per unit time, and
e = base
b
off the
h naturall llogarithm
i h ((e = 2.718).
2 718)
Department of Civil and Environmental Engineering

Lecture # 11

Poisson Distribution Example

Assume:
mean = variance (sd^2)
= 360 veh/h = 0.1 veh/s
t = 20 sec
n t

( t ) e
P ( n) =
n!

Department of Civil and Environmental Engineering

Lecture # 11

Poisson Ideas
Probability of exactly 4 vehicles arriving
P(n=4)

Probability
P b bilit off lless th
than 4 vehicles
hi l arriving
i i
P(n<4) = P(0) + P(1) + P(2) + P(3)

Probability of 4 or more vehicles arriving

P(n4) = 1 P(n<4) = 1 - P(0) + P(1) + P(2) + P(3)

A
Amount
t off time
ti
between
b t
arrival
i l off successive
i vehicles
hi l

(
t )
P(0 ) = P(h t ) =

e
0!

Department of Civil and Environmental Engineering

Lecture # 11

= e t = e qt 3600
9

Poisson Model
The assumption of Poisson vehicle arrivals also
implies a distribution of the time intervals between
the arrivals of successive vehicles (time headway).
To show this,
this note that the average arrival rate as:

q
=
3600
Where:
= average vehicle arrival rate in veh/s,
q = flow in veh/h, and 3600 = number of seconds per hour.
Department of Civil and Environmental Engineering

Lecture # 11

10

Poisson Model
Substituting into P(n) equation:

P(n ) =

(qt

3600 ) e
n!
n

- qt 3600

The probability of having no vehicles arrive in a

time interval of length t (P(0)) is equivalent to the
probability of a vehicle headway, h, being greater
than or equal to the time interval t.
t

P(0 ) = P (h t ) = e
Department of Civil and Environmental Engineering

- qqt 3600

Lecture # 11

negative exponential or
simply called
exponential distribution
11

Change of mean and distribution shape

0.25

Mean = 0
0.2
2 vehicles/minute
Probabilitty of Occurrance

0.20

Mean = 0.5 vehicles/minute

0.15

0.10

0.05

0.00
0

10

11

12

13

14

15

16

17

18

Arrivals in 15 minutes
Department of Civil and Environmental Engineering

Lecture # 11

12

19

20

Change of mean and inter-arrival

inter arrival times

P
Probability
of Exceda
ance

1.0
0.9

Mean = 0.2 vehicles/minute

0.8

Mean = 0.5 vehicles/minute

07
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0

10

12

14

16

18

Time Between Arrivals (minutes)

Department of Civil and Environmental Engineering

Lecture # 11

13

20

Limitations of the Poisson Model

M
Mean (average
(
number
b off cars per time
ti
period)
i d)
must be equal to the variance (variance over all
time period)
Otherwise use an alternative model (negative
binomial, etc.).

Department of Civil and Environmental Engineering

Lecture #

14

Poisson Distribution
Lets grab 6 hrs of 5 min aggregated counts from a station in the
Portland freeway network (I-5) for one lane
Does it match what we expect?
mean = 84.57 veh/5 min/ln (1,014 veh/hr/ln)
standard deviation = 9.35 veh
variance = 87
87.44
44 veh^2
veh 2
Poisson Model

Observed Data

0.05

0.05
6

0.04
0.04
P(n) in t=5 min

Frequency

5
4
3

0.03
0.03
0.02
0.02

0.01
0.01

119

112

105

98

91

84

77

70

63

56

49

42

35

28

21

0
120

114

108

96

102

90

84

78

72

66

60

54

48

42

36

30

24

18

12

14

0.00
0

N vehicles

How do we know if a distribution is a good representation of reality?

Any objective test?
Department of Civil and Environmental Engineering

Lecture # 11

15

Problem 5.8.
An observer has determined that the time
headways between successive vehicles
vehicles, on a
section of highway, are exponentially distributed
and
a
d that
a 60% o
of the
e headways
ead ays be
between
ee vehicles
e c es
are 13 seconds or greater. If the observer decides
to count traffic in 30-second time intervals,
estimate the probability of the observer counting
exactly four vehicles in an interval.

Department of Civil and Environmental Engineering

Lecture # 11

16

Queuing delays significance

Queuing delays can account for up to 90%
or more of a drivers total trip travel time.
Examples
E
l off queuing:
i
Traffic Signals
Toll booths
Traffic incidents (accidents and vehicle
disablements)

Department of Civil and Environmental Engineering

Lecture # 11

17

Queueing Theory
Objects passing through point with restriction on
maximum rate of passage
Input + storage area (queue) + restriction +
output
Customers,, arrivals,, arrival process,
p
, server,,
service mechanism, departures, discipline (FIFO)
Input

Storage

Output

Restriction
Department of Civil and Environmental Engineering

Lecture # 11

18

Queueing Theory: Study of Congestion

Phenomena
Applications: Airplane waiting for takeoff
takeoff, toll
gate, wait for elevator, taxi stand, ships at a port,
grocery
g
y store, telecommunications, circuits
Interested in: maximum queue length, typical
queueing times . SERVICE LEVEL !

Input

Storage

Output

Restriction
Department of Civil and Environmental Engineering

Lecture # 11

19

Queueing Theory: common assumptions

Arrival patterns (, in vehicles per unit time):
equal time intervals (derived from the assumption of
uniform, deterministic arrivals) and
exponentially
p
y distributed time intervals ((derived from the
assumption of Poisson-distributed arrivals).

Departure patterns (, in vehicles per unit time)

time),
equal time intervals (derived from the assumption of
uniform, deterministic arrivals)) and
exponentially distributed time intervals (derived from the
assumption of Poisson-distributed arrivals).

Department of Civil and Environmental Engineering

Lecture # 11

20

Queueing Theory: Some definitions

D/D/1

Deterministic arrivals
Deterministic departures
1 channel departures
Graphical solution easiest

G/G/m

General arrivals
G
General
l departures
d
t
Multi-channel departures
SIMULATION

M/D/1

Exponential arrivals
Deterministic departures
1 channel departures
Mathematical solution

M/M/1

Exponential arrivals
Exponential departures
1 channel departures
p
Mathematical solution

Department of Civil and Environmental Engineering

Lecture # 11

21

Queuing discipline
first-in-first-out (FIFO), indicating that the first
vehicle to arrive is the first vehicle to depart
depart, and
last
last-in-first-out
in first out (LIFO)
(LIFO), indicating that the last
vehicle to arrive is the first to depart.
For virtually all traffic-oriented queues, the FIFO
queuing discipline is the more appropriate of the
two.

Department of Civil and Environmental Engineering

Lecture # 11

22

D/D/1 Queuing
d
deterministic
t
i i ti arrivals
i l and
dd
departures
t
with
ith one
departure channel (D/D/1 queue)
D/D/1 queue lends itself to a graphical or
mathematical solution.
solution

Department of Civil and Environmental Engineering

Lecture # 11

23

Queueing Theory : Conservation Principle

Customers dont disappear
Arrival times of customers completely
p
y characterizes
arrival process.
Time/accumulation axes
Uniform arrivals/departures
N(x,t)
3

A=l(t)

D=m(t)

2
1
t1
Department of Civil and Environmental Engineering

t2

Time, t @ x

t3
Lecture # 11

24

Queueing Theory: Departure Process

Observer records times of departure for
corresponding
p
g objects
j
to construct D(t).
()

N(x,t)

A(t)
4
3
2
1

D(t)
t1 t2 t3

Department of Civil and Environmental Engineering

t1 t4

t2 t3
Lecture # 11

t4

Time, t @ x
25

Queueing Theory: Analysis

If system empty at t=0:
Vertical distance is queue length at time t: Q(t)=A(t)-D(t)
For FIFO horizontal distance is waiting time for jth customer.

N(x,t)

Uniform arrivals/departures

A=l(t)

Q(t)

D=m(t)

Wj

1
t1
Department of Civil and Environmental Engineering

t2

Time, t @ x

t3
Lecture # 11

26

Queueing Theory: Analysis

Horizontal strip of unit height
height, width Wj
N(x,t)

A(t)

W2
4
3
2
1

D(t)
t1 t2 t3

Department of Civil and Environmental Engineering

t1 t4

t2 t3
Lecture # 11

t4

Time, t @ x
27

Queueing Theory: Analysis

Add up horizontal stripstotal delay
Total time spent in system by some number
of vehicles (horizontal strips)
N(x,t)
Total Delay=Area
A(t)
4
3
2
1

D(t)
t1 t2 t3

Department of Civil and Environmental Engineering

t1 t4

t2 t3
Lecture # 11

t4

Time, t @ x
28

Queueing Theory
Total delay = W
Average time in queue: w = W/n
Average number in queue: Q = W/T

N(x,t)

A(t)
4
3
2
1

D(t)
t1 t2 t3

Department of Civil and Environmental Engineering

t1 t4

t2 t3
Lecture # 11

t4

Time, t @ x
29

EXAMPLE 5.7
Vehicles arrive at an entrance to a recreational
park There is a single gate (at which all vehicles
park.
must stop), where a park attendant distributes a
free
ee b
brochure.
oc u e The
e pa
park ope
opens
sa
at 8
8:00
00 A.M.,, at
a
which time vehicles begin to arrive at a rate of 480
veh/h. After 20 minutes, the arrival flow rate
declines to 120 veh/h and continues at that level
for the remainder of the day. If the time required to
distribute the brochure is 15 seconds
seconds, and
assuming D/D/1 queuing, describe the operational
characteristics of the queue
queue.
Department of Civil and Environmental Engineering

Lecture # 11

30

EXAMPLE 5
5.7
7 - SOLUTION
Begin by putting arrival and departure rates into
common units of vehicles per minute.

480 veh/h
=
= 8 veh/min for t 20 min
60 min/h
120 veh/h
h/h
=
= 2 veh/min for t > 20 min
60 min/h
60 s/min
/ i
=
= 4 veh/min for all t
15 s/veh
Department of Civil and Environmental Engineering

Lecture # 11

31

EXAMPLE 5
5.7
7 - SOLUTION
Begin by putting arrival and departure rates into
common units of vehicles per minute.

8t

for t 20 min

160 + 2(t 20)

for t > 20 min

the number of vehicle departures is:

4t

f all
for
ll t

Department of Civil and Environmental Engineering

Lecture # 11

32

EXAMPLE 5.7 - SOLUTION

A l(t) 2(t)
A=l(t)=2(t)
Equation of line =
160 + 2(t-20)

80 veh

A=l(t)=8(t)

Department of Civil and Environmental Engineering

D=m(t)=4(t)

Lecture # 11

33

EXAMPLE 5.7 - SOLUTION

When the arrival curve is above the departure
curve a queue condition will exist
curve,
exist.
The point at which the arrival curve meets the
departure curve is the moment when the queue
dissipates (no more queue exists).
The point of queue dissipation can be determined
by equating appropriate arrival and departure
equations, that is

160 + 2(t 20) = 4t

Solving for t gives t = 60 minutes.
Department of Civil and Environmental Engineering

Lecture # 11

34

EXAMPLE 5.7 - SOLUTION

Thus the queue that began to form at 8:00 A.M. will
dissipate 60 minutes later (9:00 A.M.),
) at which
time 240 vehicles will have arrived and departed (4
veh/min
e /
60 min).
)
Individual vehicle delay:
Under FIFO q
queuing
g discipline,
p
, the delay
y of an individual
vehicle is given by the horizontal distance between
arrival and departure curves.
So,
S b
by iinspection
ti off Fi
Fig. 5
5.7,
7 th
the 160th vehicle
hi l tto arrive
i
will have the longest delay of 20 minutes (the longest
horizontal distance between arrival and departure
curves)
Department of Civil and Environmental Engineering

Lecture # 11

35

 The total length of the queue is given by the

vertical distance between arrival and departure
curves at that time.
The longest queue (longest vertical distance
between arrival and departure curves) will occur at
t = 20 minutes and is 80 vehicles long
g
Total vehicle delay, defined as the summation of
the delays of each individual vehicle, is given by
the total area between arrival and departure curves

Department of Civil and Environmental Engineering

Lecture # 11

36

 In this example, the areas between arrival and

departure curves can be determined by summing
triangular areas, giving total delay, Dt, as
1
1
Dt = (80 20) + (80 40)
2
2
= 2400 vehh - min
i

Because 240 vehicles encounter queuing-delay

(as previously determined), the average delay per
vehicle is 10 minutes (2400 veh-min/240 veh),
and
d th
the average queue length
l
th is
i 40 vehicles
hi l
(2400 veh-min/60 min).
Department of Civil and Environmental Engineering

Lecture # 11

37

Problem 5.14.
Vehicles begin to arrive at a parking lot at 6:00
A.M. at a rate 8 per minute
minute. Due to an accident
on the access highway, no vehicles arrive from
6:20
6
0 to 6
6:30
30 A.M. From
o 6:30
6 30 A.M. o
on,, vehicles
e c es
arrive at a rate of 2 per minute. The parking lot
attendant processes incoming vehicles (collects
parking fees) at a rate of 4 per minute
throughout the day. Assuming D/D/1 queuing,
determine total vehicle delay
delay.

Department of Civil and Environmental Engineering

Lecture # 11

38

EXAMPLE 5.8
After observing arrivals and departures at a highway
toll booth over a 60-minute time period,
period the
observer notes that the arrival and departure rates
((or service rates)) are deterministic but,, instead of
being uniform, they change over time according to a
known function. The arrival rate is given by the
f
function
(t)
( ) = 2.2 + 0.17t 0.0032t2 and the
departure rate is given by (t) = 1.2 + 0.07t, where t
is in minutes after the beginning of the observation
period and (t) and (t) are in vehicles per minute.
Determine the total vehicle delay at the toll booth
and the longest queue assuming D/D/1 queuing.
Department of Civil and Environmental Engineering

Lecture # 11

39

M/D/1 Queuing
exponentially distributed times between the
arrivals of successive vehicles (Poisson arrivals)
deterministic departures
departures, and
one departure
d
t
channel
h
l
Obvious
O
example
Traffic Signals

Department of Civil and Environmental Engineering

Lecture # 11

40

M/D/1
Basic relationship:

Where:
= traffic intensity, and is unit-less,
= average arrival rate in vehicles per unit time,
and
= average departure rate in vehicles per unit
ti
time.
Department of Civil and Environmental Engineering

Lecture # 11

41

M/D/1
assuming that <1, for an M/D/1:
2
average length of queue in vehicles: Q =
2(1 )

average waiting time in the queue (for each

vehicle):
hi l )
w=
2 (1 )
average time
ti
spentt in
i th
the system
t
(summation
(
ti off
average queue waiting time + average departure
2
time (service time)):
t=
2 (1 )
Department of Civil and Environmental Engineering

Lecture # 11

42

M/D/1
NOTE !
that
th t the
th traffic
t ffi intensity
i t
it is
i less
l
th
than one (( <1),
<1) th
the
D/D/1 queue will predict
NO queue formation
formation.

Models with random arrivals or departures, such as

the M/D/1 queuing model
model, will predict
queue formations !

2
Q=
2(1 )

Department of Civil and Environmental Engineering

w=

2 (1 )

Lecture # 11

2
t=
2 (1 )

43

M/M/1 Queuing
exponentially distributed times between the arrivals
of successive vehicles (Poisson arrivals)
exponentially distributed departure time patterns in
addition to exponentially distributed arrival times
one departure channel
Traffic applications:
Toll booth where some arriving drivers have the correct
toll and can be p
processed q
quickly,
y, and others may
y not
have the correct toll, thus producing a distribution of
departures about some mean departure rate.

Department of Civil and Environmental Engineering

Lecture # 11

44

M/M/1
= arrival rate
= departure (service) rate
Average length of queue

2
Q=
(1 )

Average time waiting in queue

Average time spent in system

Department of Civil and Environmental Engineering

< 1.0

Lecture # 11

w =

1
t=

45

M/M/N Queuing
Applications:
M/M/N queuing
i iis a reasonable
bl assumption
ti att ttollll
booths on turnpikes or at toll bridges where there
is often more than one departure channel available
(more than one toll booth open).
M/M/N queuing is also frequently encountered in
non-traffic but transportation applications such as
security checks at airports, vessel queueing at
ports/airports and so on.
Other non-transportation :
checkout lines at retail stores, call centers, etc.
Department of Civil and Environmental Engineering

Lecture # 11

46

M/M/N Multi
M/M/N
Multi-channel
channel

N < 1.0

Average length of queue

P0 N +1
1
Q=

2
N ! N (1 N )
Average
g time waiting
g in q
queue
+Q 1
w=

Average time spent in system

t=

Department of Civil and Environmental Engineering

Lecture # 11

+Q

47

M/M/N
Probability of having no vehicles
1
P0 = N 1 nc

N
+

N !(1 N )
nc = 0 nc !
Probability of having n vehicles
n
n
P0

P0
for n N
Pn = n N
Pn =
for n N
N N!
n!
Probability of being in a queue

Pn > N
Department of Civil and Environmental Engineering

Lecture # 11

P0 N +1
=
N ! N (1 N )
48

Problem 5.35
Vehicles leave an airport parking facility (arrive at
parking fee collection booths) at a rate of 500
veh/h (the time between arrivals is exponentially
distributed).
d
s bu ed) The
e pa
parking
g facility
ac y has
as a po
policy
cy that
a
the average time a patron spends in a queue while
waiting to pay for parking is not to exceed 5
seconds. If the time required to pay for parking is
exponentially distributed with a mean of 15
seconds what is the fewest number of payment
seconds,
processing booths that must be open to keep the
average time spent in a queue less than 5
seconds.
Department of Civil and Environmental Engineering

Lecture # 11

49