327 Profnotes 01 All

Lecture Notes for CE327:
Transportation Systems Engineering
Instructor: Mahdieh Allahviranloo

Lecture Notes for CE327,Spring 2015
Instructor: Mahdieh Allahviranloo; The City College of New York
Contents
1 Introduction to Transportation Engineering ..................................................................................... 4
1.1 Concept of travel distance and speed: ...................................................................................... 4
2 Detectors: ...................................................................................................................................... 10
3 Speed and density: ........................................................................................................................ 11
4 Car following models: .................................................................................................................... 17
5 Probability: .................................................................................................................................... 33
5.1 Previews: ............................................................................................................................... 33
5.2 Binomial PMF: ........................................................................................................................ 37
5.3 Geometric Distribution:.......................................................................................................... 40
5.4 Poisson distribution:............................................................................................................... 41
6 Continuous Distributions ............................................................................................................... 42
6.1 Exponential Distribution ......................................................................................................... 42
6.2 Normal Distribution ............................................................................................................... 43
6.2.1 PDF distribution.............................................................................................................. 43
6.2.2 Standard Normal Distribution ......................................................................................... 43
6.2.3 Solving for the Xth percentile ........................................................................................... 43
7 Queue Theory ................................................................................................................................ 47
8 Shockwave..................................................................................................................................... 51
9 Signal Timing ................................................................................................................................. 51
10 Signal Design.............................................................................................................................. 52
11 Road Geometric design: ............................................................................................................. 66
11.1 Design vehicles classification: ................................................................................................. 66
11.2 Turning radius: ....................................................................................................................... 69
11.3 Cross-sections ........................................................................................................................ 71
11.4 VERTICAL ALIGNMENT............................................................................................................ 71
11.4.1 How to identify the minimum length of the curve? ......................................................... 73
11.4.2 Stopping Sight distance: ................................................................................................. 73
11.4.3 Comfort criteria based on AASHTO: ................................................................................ 74
11.4.4 Example1: Minimum length of a crest vertical curve ....................................................... 74
11.4.5 Example 2: Design of crest vertical curve, ....................................................................... 75
11.5 Horizontal Alignment: ............................................................................................................ 76
11.5.1 Some definitions:............................................................................................................ 76
11.5.2 Circular curve: ................................................................................................................ 77
2
11.5.3 Example: ........................................................................................................................ 79

11.5.4 Transition Curves: ........................................................................................................... 80
11.5.5 Superelevation: .............................................................................................................. 81
12 Pavement design: ...................................................................................................................... 83
12.1 Flexible Pavement .................................................................................................................. 83
12.1.1 Design methods:............................................................................................................. 83
12.1.2 What is HMA? ................................................................................................................ 84
12.1.3 Layered flexible pavement.............................................................................................. 84
12.1.4 Full-Depth Asphalt Pavement ......................................................................................... 86
12.2 Rigid Pavement ...................................................................................................................... 86
12.3 Stresses and Strains in Flexible Pavements ............................................................................. 87
12.3.1 Solutions by charts: ........................................................................................................ 87
12.3.2 Numerical solutions: ....................................................................................................... 93
12.3.3 Two layered systems: ..................................................................................................... 96
12.4 Traffic Loading: ...................................................................................................................... 98
12.4.1 Damage types:.............................................................................................................. 100
3
1 Introduction to Transportation Engineering
What is a Transportation System? “A transportation system may be defined as consisting of the fixed
facilities, the flow entities, and the control system that permit people and goods to overcome the
friction of geographical space efficiently in order to participate in a timely manner in some desired
activity.”
Importance of transportation:
 Safety
 Mobility
 Environment
 Accessibility
 Productivity
 Sustainable economic development
 Quality of life
 Equity
Let’s start with some preliminaries:

1.1 Concept of travel distance and speed:
Travel distance:
𝑥𝑖 (𝑡 + ∆𝑡) − 𝑥𝑖 (𝑡)
Speed:
𝑥𝑖 (𝑡 + ∆𝑡) − 𝑥𝑖 (𝑡)
∆𝑡
Instantaneous speed:
𝑑(𝑥𝑖 (𝑡)) 𝑥𝑖 (𝑡 + ∆𝑡) − 𝑥𝑖 (𝑡)

𝑢𝑖 (𝑡) = = lim
𝑑(𝑡) ∆𝑡→0 ∆𝑡
Acceleration rate:
𝑑(𝑢𝑖 (𝑡)) 𝑑 2 (𝑥𝑖 (𝑡))
𝑢𝑖 (𝑡) = =
𝑑(𝑡) 𝑑(𝑡)2
4
𝑡
𝑥𝑖 (𝑡) = ∫ 𝑣𝑖 (𝑠)𝑑𝑠 + 𝑥𝑖 (0)
0
Headway: (unit : time)

Time between successive passes when vehicles pass a location x on lane. It is measured from the same
reference point of each vehicle.
ℎ𝑖 (𝑥) = 𝑡𝑖+1 (𝑥) − 𝑡𝑖 (𝑥)
t
ti ti+1
5
Gap: (unit :time)

Time between rear bumper of the first vehicle to front bumper of the following vehicle. It is measured in
time.
Spacing (unit: length )

Distance between successive vehicles in traffic streams, measured from the same reference point on
vehicles.
Clearance
Spacing
Spacing = clearance + length of vehicle
Traffic flow: (unit: vehicles/hour)

Flow rate (q) is number of vehicles passing a certain location on the road in one hour.
𝑁
𝑞=
𝑇
Density : (vehicles/mile/lane)
Number of vehicles in section of road.
1 mile
6
Average speed:
Time mean speed:
The arithmetic mean of all individual speeds at the space spot.
𝑇𝑜𝑡𝑎𝑙 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 ∑𝑁 𝑢𝑖 . ∆𝑡 ∑𝑁 𝑢 𝑖
𝑇𝑀𝑆 = = 𝑖=1 = 𝑖=1
𝑇𝑜𝑡𝑎𝑙 𝑇𝑖𝑚𝑒 𝑁. ∆𝑡 𝑁
Space mean speed: (harmonic speed)
𝑁
SMS = 1
∑𝑁
𝑖=1𝑈 (𝑥)
𝑖
TMS>=SMS
Example 1:
If a traveler goes from A to B, a distance of 20 mile at average speed of 80mph and return at 40 mph,
what is the average speed?
80+40
TMS: 𝑢̅ = = 60
2
SMS:
20+20
𝑢̅ = 20 20 = 53 mph
+
80 40
Example 2 :
Given the observed spot speeds below, determine the:
1) Time mean speed
2) Space mean speed
7
Observed Spot
Speed (mph)
25
26
31
28
22
38
40
31
29
30
Time Mean Speed:

𝑁
1
𝑢̅𝑡 = ∑ 𝑣𝑖
𝑁
𝑖=1
1
= (25 + 26 + 31 + 28 + 22 + 38 + 40 + 31 + 29 + 30 = 𝟑𝟎. 𝟎 𝐦𝐩𝐡
10
Space Mean Speed:

1
𝑢𝑠 =
̅̅̅
1 𝑁 1
∑
𝑁 𝑖=1 𝑣𝑖
1
= = 𝟐𝟗. 𝟏𝟑 𝐦𝐩𝐡
1 1 1 1 1 1 1 1 1 1 1
( + + + + + + + + + )
10 25 26 31 28 22 38 40 31 29 30
8
Uniform traffic:
𝑢1 = 𝑢2 = ⋯ = 𝑢𝑎𝑣𝑒
Fundamental law of traffic flow:
Flow = density * speed.
Occupancy:
The fraction of time that a detector i4s covered by vehicles.
𝐿+𝑑
∑𝑁 𝐿+𝑑
𝑁
1
𝑖=1 𝑢𝑖
𝑜𝑐𝑐 = = .∑
𝑇 𝑇 𝑢𝑖
𝑖=1
d
Presence Detector
Input
Signal
9
2 Detectors:
Types of detector outputs:
Presence Detector Output
0000001111111111000001111111111110000011111111100000000111111100000
Sampling Interval
Pulse Detector Output
1
0000001000000000000001000000000000000010000000000000000100000000000
Sampling Interval
10
3 Speed and density:
Greenshield model:
Linear relation between speed and density
𝐾
𝑢 = 𝑢𝑓 (1 − )
𝐾𝑗
𝐾
𝑞 = 𝑢. 𝑘 = 𝑢𝑓 𝐾(1 − )
𝐾𝑗
100
90
80
70
60
Speed (mph)
50
40
30
20
10
0
0 50 100 150 200 250
Density (veh/mile)
11
4000
3500
3000
2500
Flow(Veh/hour)
2000
1500
1000
500
0
0 50 100 150 200 250
Density (veh/mile)
80
70
60
50
Speed (mph)
40
30
20
10
0
0 500 1000 1500 2000 2500 3000 3500 4000
Flow(Veh/hour)
12
𝐾
𝑞 = 𝑢. 𝑘 = 𝑢𝑓 𝐾(1 − )
𝐾𝑗
Question:
Find the value of maximum capacity on the freeway. Take the derivative of flow wrt density. (Kc
represents density at the capacity)
𝐾𝑐
𝑑(𝑢𝑓 𝐾𝑐 (1 − )) 𝑢𝑓 𝐾𝑗
𝑑𝑞 𝐾𝑗
= = 𝑢𝑓 − 2. 𝐾𝑐 = 0 , 𝐾𝑐 =
𝑑𝐾 𝑑𝑘𝑐 𝐾𝑗 2
Replacing Kc in the equation of speed :

𝐾
𝑢 = 𝑢𝑓 (1 − )
𝐾𝑗
𝐾𝑐 𝑢𝑓
𝑢𝑐 = 𝑢𝑓 (1 − )=
𝐾𝑗 2
And,
𝑢𝑓 𝐾𝑗
𝑞𝑐 = 𝑢𝑐 . 𝑘𝑐 =
4
13
Question: Identify congested and uncongested areas in diagrams:
Greenshield Flow - Speed Relationship
4000
qc
3500
3000
Flow (vehicles/hour/lane)
2500
2000
1500
Congested Uncongested
1000 (Unstable) (Stable)
Flow Flow
500
0
0 10 20 30 xc 40 50 60
S70f 80
Speed (mph)
Greenshield Flow - Density Relationship
4000
qc
3500
3000
Flow (vehicles/hour/lane)
2500
2000
1500
Uncongested Congested
(Stable) Flow (Unstable) Flow
1000
500
0
0 50 100 kc 150 200
kj 250
Density (veh/mi/lane)
14
Greenberg model:
𝐾𝑗
𝑢 = 𝐶. 𝑙𝑛( )
𝐾
𝐾𝑗
𝑞 = 𝑢. 𝑘 = 𝐶. 𝑙𝑛( )𝐾
𝐾
150
100
Speed (mph)
50
0
0 50 100 150 200 250
Density (veh/mile)
4000
3500
3000
2500
Flow(Veh/hour)
2000
1500
1000
500
0
0 50 100 150 200 250
Density (veh/mile)
15
150
100
Speed (mph)
50
0
0 500 1000 1500 2000 2500 3000
Flow(Veh/hour)
𝐾𝑗
𝑞 = 𝑢. 𝑘 = 𝐶. 𝑙𝑛( )𝐾
𝐾
𝐾𝑗
𝑑𝑞 𝑑(𝐶. 𝑙𝑛 ( ) 𝐾𝑐 )
𝐾𝑐
=
𝑑𝑘𝑐 𝑑𝑘𝑐
𝑑 (𝐶. (𝐾𝑐 . ln 𝐾𝑗 − 𝐾𝑐 𝑙𝑛𝐾𝑐 ))

= ln 𝐾𝑗 − ln 𝐾𝑐 − 1 = 0, 𝐾𝑐 = 𝑒 (ln(𝐾𝑗 )−1)
𝑑𝑘𝑐
𝐾𝑗
𝐾𝑐 =
𝑒
By replacing Kc in speed and flow equations:
𝐾𝑗
𝑈𝑐 = 𝐶. ln ( ) = 𝐶
𝐾𝑐
𝑞𝑐 = 𝐶. 𝐾𝑐
16
4 Car following models:

How vehicles follow each other on road sections.
California DMV rule:

“A good rule for following another vehicle at a safe distance is to allow yourself at least the
length of a car between your vehicle and the vehicle ahead for every ten miles per hour of speed
at which you are traveling.”
(𝑥̇ ) is the speed of the vehicles.
xn 1 (t ) xn (t )
Ln 1 x (t ) d (t ) Ln xn (t )
n 1
 x (t )(mph)   x n1 (t )(ft/sec) 

d (t )(ft)   n1   Ln 1 (ft)     Ln 1 (ft)
 10mph   14.67ft/sec 
 x (t )(ft/sec) 
d (t )  xn (t )  Ln  xn 1 (t )   n1   Ln 1 (ft)
 14.67ft/sec 
 x (t )(ft/sec) 
xn (t )  Ln  xn 1 (t )   n1   Ln 1 (ft)  0
 14.67ft/sec 
17
Safe braking distance:
As a general rule, the spacing between vehicles should be such that if a sudden deceleration
becomes necessary by the leading vehicle, the following vehicle has ample time to react and stop
safely without colliding with the stopping lead vehicle.
xn 1 (t ) xn (t )
Ln 1 x (t ) d (t ) Ln xn (t )
n 1
t R  Reaction time
d n  Distance required for vehicle n to stop (assuming panic braking)
d n 1  Distance required for vehicle n  1 to stop (assuming comfortable braking)
d *  Safety margin (after stop)
bi  Braking distance of vehicle i (i  n, n  1)
18
For safe stopping:
d n 1  d (t )  d n
or
d (t )  d n 1  d n
or
d (t )  d n 1  d n  d *
dn1  tR  xn1 (t )  bn1
 x (t ) 
2
xn (t )2
d (t )  tR  xn 1 (t )  n 1   d*
2an 1 2an
19
Example: Designing Ramp Metering

Consider the section of freeway shown below:
Loop Station 2 Loop Station 3
Loop Station 1
0.2 mile 1.5 miles
1 mile 2 miles
Based on previous studies, the following flow – density data have been gathered for the three loop
stations shown in the diagram:
Upstream Station 1
9000
8000
7000
6000
Flow (vph)
5000
4000
3000
2000
1000
0
0 100 200 300 400 500 600
Density (vpm)
20
Upstream Station 2
9000
8000
7000
6000
Flow (vph)
5000
4000
3000
2000
1000
0
0 100 200 300 400 500 600
Density (vpm)
Downstream Station 3
6000
5000
4000
Flow (vph)
3000
2000
1000
0
0 50 100 150 200 250 300 350 400
Density (vpm)
21
Peak hour traffic on this particular freeway section lasts from about 7:00 AM until 9:00 AM and is
relatively uniform and stable during that two-hour period.
One morning, the following loop data are transmitted to the Traffic Management Center from each of
the three stations:
Station 1: 30-second Count Data
40
35
30
30-second Count (Vehicles)
25
20
15
10
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
22
Station 1: 30-second Occupancy Data
0.9
0.8
0.7
Occupancy (decimal)
0.6
0.5
0.4
0.3
0.2
0.1
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
40
35
30
25
20
15
10
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
23
0.9
0.8
0.7
Occupancy (decimal)
0.6
0.5
0.4
0.3
0.2
0.1
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
40
35
30
25
20
15
10
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
24
0.9
0.8
0.7
Occupancy (decimal)
0.6
0.5
0.4
0.3
0.2
0.1
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
Questions:
What is the state of the system during the morning peak period?
What is the current ramp metering rate?
What would be the maximum ramp metering rate permissible in order to ensure that, barring any
incidents such as accidents, the speed at all sections downstream from the ramp entry point virtually
never fell below 40 mph during the morning peak period?
Solution:
First, draw approximate lines and/or curves through data, i.e.,
Station1:
25
Upstream Station 1
9000
8000
7000
6000
Flow (vph)
5000
4000
3000
2000
1000
0
0 100 200 300 400 500 600
Density (vpm)
Upstream Station 2
9000
8000
7000
6000
Flow (vph)
5000
4000
3000
2000
1000
0
0 100 200 300 400 500 600
Density (vpm)
26
Downstream Station 3
6000
5000
4000
Flow (vph)
3000
2000
1000
0
0 50 100 150 200 250 300 350 400
Density (vpm)
40
35
30
25
20
15
10
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
q(vph)  (count ) 120

5280  Occupancy
k (vpm) 
( L  LD )
5280  Occupancy

(15  5)
 264  Occupancy
27
0.9
0.8
0.7
Occupancy (decimal)
0.6
0.5
0.4
0.3
0.2
0.1
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
q1  20 120  2400 vph

k1  264  0.16  42 vpm
x1  q1 k1  57 mph
Upstream Station 1
9000
8000
7000
6000
Flow (vph)
5000
4000
3000
2000
1000
0
0 100 200 300 400 500 600
Density (vpm)
For station 2:
28
40
35
30
25
20
15
10
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
q(vph)  (count ) 120

5280  Occupancy
k (vpm) 
( L  LD )
5280  Occupancy

(15  5)
 264  Occupancy
0.9
0.8
0.7
Occupancy (decimal)
0.6
0.5
0.4
0.3
0.2
0.1
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
29
q2  30 120  3600 vph

k2  264  0.22  58 vpm
x2  q2 k2  62 mph
Upstream Station 2
9000
8000
7000
6000
Flow (vph)
5000
4000
3000
2000
1000
0
0 100 200 300 400 500 600
Density (vpm)
30
Station 3:
40
35
30
25
20
15
10
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
0.9
0.8
0.7
Occupancy (decimal)
0.6
0.5
0.4
0.3
0.2
0.1
0
7:00:00 AM 7:10:00 AM 7:20:00 AM 7:30:00 AM 7:40:00 AM 7:50:00 AM 8:00:00 AM
Time
q3  30 120  3600 vph

k3  264  0.28  74 vpm
x3  q3 k3  48.7 mph
31
What would be the maximum ramp metering rate permissible in order to ensure that, barring any
incidents such as accidents, the speed at all sections downstream from the ramp entry point virtually
never fell below 40 mph during the morning peak period?
From station1 loop data, upstream flow rate during peak hour is about 2400 vph. From the flow –
density relationship for the two-lane section downstream, draw a line from origin with slope = 40 mph
Therefore, max flow downstream to keep speed above 40mph is about 4600 vph. Thus, metering rate
should be less than 4600-2400 = 2200 vph, or a metering headway greater than 1.6 sec.
q1  20 120  2400 vph q2  30 120  3600 vph q3  30 120  3600 vph

k1  264  0.16  42 vpm k2  264  0.22  58 vpm k3  264  0.28  74 vpm
x1  q1 k1  57 mph x2  q2 k2  62 mph x3  q3 k3  48.7 mph
Region 1 Region 2 Region 3

Loop Station 2 Loop Station 3
Loop Station 1
0.2 mile 1.5 miles
1 mile 2 miles
MR = 3600 vph - 2400 vph = 1200 vph
32
5 Probability:
Given a set of outcomes, determine the likelihood of an outcome
It is often directly related to plausible or conceivable statistical experiments. Ex: probability that 15 cars
will turn right from 96th street to Broadway in 1 minutes?
If the number of vehicles turning right from 96 th street to Broadway in 100 minutes is given in the
following table then and in only 8 of those numbers is 15, then the probability will be 8%.
data # cars Yes or no (0 -1)
1 8 0
2 10 0
3 15 1
4 15 1
˸ : :
100 11 0
5.1 Previews:
Mean: average value

 Arithmetic
 Harmonic
Median:
Middle value
Mode:
Most frequent value
Pace:
Range containing most observations
Standard Deviation:
Measure of spread of values
Sample Space:
Set of all possible experiments. The probability for an outcome form the sample space to happen is 1.
33
For a dice the sample space is {1,2,3,4,5,6}.

Two types of sample spaces: discrete or continuous sample space
- Discrete sample spaces: can associate an integer number. Outcome of the experiment is integer
number.
- Continuous sample space: can associate only real numbers to outcomes. Sample space can be
finite or infinite.
Event: is a subset of sample space.
Random variable:
A random variable is a variable associated with the probabilistic experiment (or occurrences) that takes
on a numerical value corresponding each outcome of the experiment.
Examples of discrete r.v.:
- Throw a dice (1, 2,3,4,5,6)
- Toss a coin
If P represents the probability of a discrete random variable. Then the followings apply:
o P(A)≥ 𝟎,
o ∑𝒊 𝑷(𝑨𝒊) = 1,
o 𝑷(𝑨) ≤ 𝟏.
PMF
0.005
0.0045
0.004
0.0035
0.003
0.0025
0.002
0.0015
0.001
0.0005
0
1
37
73
217
361
541
685
865
109
145
181
253
289
325
397
433
469
505
577
613
649
721
757
793
829
34
CDF
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 200 400 600 800
Examples of continuous r.v.:

- Next bus arrival time
For continuous r.v.: 𝑓(𝑥)𝑑𝑥, the probability is defined as ‘a random variable X takes value within a
small dx around value x.
𝑥= ∞
𝑓𝑥 (𝑋) ≥ 0, ∫ 𝑓𝑥 (𝑋)𝑑𝑥 = 1
𝑥=−∞
Expected value:
𝐸(𝑥) = ∑ 𝑥𝑖 . 𝑃𝑖 (𝑥𝑖 )
𝑖
Variance:
2
𝜎(𝑥) = ∑ (𝑥 − 𝐸(𝑥)) 𝑓𝑥 (𝑥)𝑑𝑥
𝑎𝑙𝑙 𝑥
𝜎(𝑥) = 𝐸(𝑥 2 ) − 𝐸(𝑥)2
35
Sets operations:
A , B
Mutually Exclusive Events:
If 𝐴 ∩ 𝐵 = ∅
Collectively Exhaustive:
If 𝐴 ∪ 𝐵 = {𝑠𝑎𝑚𝑝𝑙𝑒 𝑆𝑒𝑡}, then A and B are collectively exhaustive sets.
Independent events:
If 𝑃(𝐴 ∩ 𝐵) = 𝑃(𝐴). 𝑃(𝐵)
Example: Probability of events when tossing a coin.
Question:
Is the probability of cars arrival to the intersection independent, in congested traffic? How about
uncongested traffic?
36
Discrete distributions:
- Binomial
- Geometric
- Poisson
Continuous distributions:
- Exponential
- Normal
We first start with the definition of Bernoulli trial.
Bernoulli trial:
A probability trial for experiment with only 2 outcomes, (ex: flipping a coin)
Say success with probability of P and failure with the probability of 1-P.
5.2 Binomial PMF:
Probability of n success in m ‘independent’ trial each time having probability of Bernoulli.

𝑃(𝑠𝑢𝑐𝑐𝑒𝑠𝑠) = 𝑃, 𝑃(𝑓𝑎𝑖𝑙) = 1 − 𝑃
𝑚!
𝑃(𝑛 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑖𝑛 𝑚 𝑡𝑟𝑖𝑎𝑙) = 𝑃𝑛 . (1 − 𝑃)𝑚−𝑛
𝑛! (𝑚 − 𝑛)!
𝐸(𝑛) = 𝑚. 𝑃, 𝜎𝑥2 = 𝑚𝑝(1 − 𝑝)
Example of Binomial PMF:

Probability of one car’s arrival in every second to the intersection is p and the probability of ‘no arrival’
in each second is (1-p). Assuming that the traffic is uncongested and probability of arrival for different
vehicles are independent events. If P =0.2 what is the probability of arrival of 3 cars in 5 seconds?
5
( ) 𝑃3 (1 − 𝑃)2 = 10 ∗ 0.008 ∗ (0.64) = 0.0512
3
37

On average 12% of the data collected by the detectors installed on a section of the road are not
reliable due to measurement errors caused by detectors. What is the probability that for a
network with 10 road sections,
(a)No more than 2 detectors have issues?
(b) At least 2 of the detectors are out of order?
Let X= number of detectors which are out of order. (In this case, "success" means
‘detectors are out of order’!) . Here, m=10, p=0.12, q=0.88.
(a) No rejects. That is, when x=0:

10!
𝑃(𝑋 = 0) = 0.120 . (0.88)10 = 0.2785
10! 0!
when x=1:
10!
𝑃(𝑋 = 1) = 0.121. (0.88)9 = 0.37977
9! 1!
when x=2:
10!
𝑃(𝑋 = 2) = 0.122. (0.88)8 = 0.23304
8! 2!
Probability=P(X≤2)
=0.2785+0.37977+0.23304=0.89131
(b) We could work out all the cases for X=2,3,4,…,10, but it is much easier to proceed as
follows:
Probability of at least 2 rejects =1−P(X≤1)=1−(P(x0)+P(x1))

=1−(0.2785+0.37977)=0.34173
38

Following table shows the number of different vehicle types detected by sensor at a section of road.
Given this table compute the followings:
Truck Types # Vehicles

2-Axle Single Unit 68
5-Axle Semitrailer 35
Total 140
 What is the probability that any truck in the fleet is a 5-Axle Semitrailer?
35
𝑃= = 0.25
140
 If 20 trucks arrive in an hour:
a. What is the probability that exactly 4 will be 5-Axle semitrailers?
20!
𝑃 (𝑋 = 4) = 0.254. (0.75)16 = 0.189
4! 16!
b. What is the probability that 5 or fewer will be 5-Axle semitrailers?
𝑃𝑟𝑜𝑏 (𝑋 ≤ 5) = 𝑃(𝑋 = 0) + 𝑃(𝑋 = 1) + 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5)

= 0.6171
39
5.3 Geometric Distribution:

Probability that the first success happens in the xth trial. (zero success until (x-1) trials then a success
occurs. If P represents the probability of success and (1-P) represents the probability of fail:
𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 = (1 − 𝑃) 𝑥−1 𝑃
1 1−𝑝
𝐸(𝑛) = , 𝑉(𝑛) = 2
𝑝 𝑝
Example of Geometric PMF:

Following table shows the number of different vehicle types detected by sensor at a section of road.
Given this table compute the followings:
Truck Types # Vehicles

5-Axle Semitrailer 35
Total 140
a) What is the probability that the first 5-axle semitrailer to arrive will be the 8th vehicle?
𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 = (1 − 0.25)8−1 ∗ 0.25 = 0.033
b) What is the probability that if 5 vehicles arrive, only the 2nd and 4th are 5-axle semitrailers?
𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 = 0.75 ∗ 0.25 ∗ 0.75 ∗ 0.25 ∗ 0.75 = 0.026
40
5.4 Poisson distribution:

Probability of occurrence of ‘n’ events in ‘t’ time period. If 𝜆 is the arrival rate then this probability can
be written as:
(𝜆𝑡)𝑛 . 𝑒 −𝜆𝑡
𝑃(𝑛 𝑠𝑢𝑐𝑐𝑒𝑠 𝑖𝑛 𝑡 𝑡𝑖𝑚𝑒) =
𝑛!
Example of Poisson PMF:
Vehicles arrive to a toll booth at an average rate of one vehicle every 30 seconds. Find the
probability that during a 5 minute period:
a) Exactly 10 vehicles arrive
b) More than 3 vehicles arrive
Lambda = 2, t = 5
0.14
0.12
0.1
Probability
0.08
0.06
0.04
0.02
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
N
Probability of N event, in 5 minutes where Lambda= 2 vehicle/minute.
a)
𝜆 = 2 𝑣𝑒ℎ/𝑚𝑖𝑛𝑢𝑡𝑒
(2 ∗ 5)10 . 𝑒 −2∗5
𝑃(10 𝑠𝑢𝑐𝑐𝑒𝑠 𝑖𝑛 5 𝑚𝑖𝑛𝑢𝑡𝑒) = = 0.125
10!
b) 𝑃(𝑛 > 3) = 1 − 𝑃(𝑛 = 0) − 𝑃(𝑛 = 1) − 𝑃(𝑛 = 2) − 𝑃(𝑛 = 3) = 0.989
𝐸(𝑁) = 𝑛. 𝑃(𝑛) = 𝜆𝑡
41
6 Continuous Distributions
6.1 Exponential Distribution
Used to determine probability of gaps in a traffic stream. When the arrivals or occurrence of events have
Poisson distribution then the gap between every arrival has Exponential Distribution. This can be called
as the ‘Headway’ distribution.
Recall from Poisson distribution,
(𝜆𝑡)𝑛 . 𝑒 −𝜆𝑡
𝑃(𝑛 𝑠𝑢𝑐𝑐𝑒𝑠 𝑖𝑛 𝑡 𝑡𝑖𝑚𝑒) =
𝑛!
Then
𝑃(0 𝑎𝑟𝑟𝑖𝑣𝑎𝑙 𝑖𝑛 𝑡 𝑡𝑖𝑚𝑒) = 𝑒 −𝜆𝑡

If there is no arrival in time ‘t’ means that interarrival time or Headway is greater than t. then we cans
say
𝑃(ℎ𝑒𝑎𝑑𝑤𝑎𝑦 > 𝑡) = 𝑒 −𝜆𝑡
Then
𝑃(ℎ𝑒𝑎𝑑𝑤𝑎𝑦 < 𝑡) = 1 − 𝑒 −𝜆𝑡
Then we can compute 𝑃(ℎ𝑒𝑎𝑑𝑤𝑎𝑦 = 𝑡) using following equation:
𝑑(1 − 𝑒 −𝜆𝑡 )
𝑃(ℎ𝑒𝑎𝑑𝑤𝑎𝑦 = 𝑡) = = 𝜆𝑒 −𝜆𝑡
𝑑𝑡
𝟏
𝑬(𝒕) =
𝝀
Example of Exponential PMF:

Given a vehicle arrival rate of 2 veh/min, find the probability that the gap between any two vehicles will
exceed 1 minute.
𝜆=2
𝑃(ℎ𝑒𝑎𝑑𝑤𝑎𝑦 > 1) = 𝑒 −2 = 0.135
How many gaps of longer than 1 minute do you expect in a 24 hour period?
42
6.2 Normal Distribution 1

6.2.1 PDF distribution
σ σ σ σ σ σ
68%
95%
98%
6.2.2 Standard Normal Distribution
For Standard Normal distribution Mean = 0 and the variance is 1.

Question:
How to convert a random variable x with mean μ and standard deviation σ to a standard normal?
Define a variable z and use the following conversion equation:
𝑥−𝜇
𝑍=
𝜎
6.2.3 Solving for the Xth percentile
1
The lecture for Normal Distribution was prepared using Dr.Conway’s CE327 class.
43
(100-i)%
i%
Steps:
𝑥 −𝜇
a) Convert 𝑥𝑖 to standard normal distribution 𝑍𝑖 = 𝑖
𝜎
b) Find the area under the graph using following table: (note that the area presented in this table is
the area from the mean to the location of Z.
44
c)
45
Example of Normal Distributions in Design:

85th percentile speed: Speed at or below which 85% of vehicles are expected to travel.
90th percentile stopping sight distance: Distance traveled from initial reaction until full stop that will
not be exceeded by 90% of vehicles.
Numerical Example:
Given the observed speeds for a highway segment
Observation # Speed (x) (mph) Observation # Speed (x) (mph) Observation # Speed (x) (mph)
1 57 8 52 15 61
2 60 9 58 16 48
3 51 10 55 17 55
4 57 11 57 18 50
5 49 12 55 19 55
6 51 13 54 20 59
7 58 14 50
Determine the 85th percentile speed.
46
7 Queue Theory
 Deterministic Queues
 Stochastic Queues (Probability)
 Isolated Queues (grocery checkout line) versus Queuing systems (factory floor assembly line,
where output of one queue is the input to another queue)
Basic idea of stochastic queues:
Say ‘S’ is the service time and ‘a’ is average arrival time,
 If S>a, queue keeps on lengthening
 If S<a, we get a queue with steady probabilities for it to be of any length (1,0,2,1,…)
Notation:
𝐴/𝑆/𝑁 ∶
A: letter denoting arrival distribution
S: letter denoting service distribution
N : number of servers in the queue
Arrival Patterns(A) Service Patterns (S) Queue Disciplines
Uniform Arrivals (D) Uniform Departures (D) First-in-First-Out (FIFO)
Poisson Distribution Exponential Distribution of Service Times (M) Last-In-First-Out (LIFO)

of Arrivals (M)
General Probability General Probability Distribution (G)

Distribution (G)
Question:
If 𝜆 is arrival rate and 𝜇 is service rate for each server, and c’ is the number of severs. Lets define r as
𝜆
𝑟= , in which of these following conditions queue will form?
𝑐𝜇
a) 𝑟 < 1
b) 𝑟 > 1
When 𝑟 < 1 we have a stable condition. However for 𝑟 > 1 the system is unstable.
47
Cumulative numbers
cumulative arrival time a(t)
cumulative service completion s(t)
Time
Total time spend in the system by time T for all the costumers is:
𝑇 𝑇
𝑙(𝑇) = ∫ 𝑎(𝑡)𝑑𝑡 − ∫ 𝑠(𝑡)𝑑𝑡
0 0
Average number of people in the queue:

𝑙(𝑇)
𝑛=
𝑇
Queue for (M/M/1)
Assumes Poisson arrivals and Exponential distribution of service times (M/M/1)
 Average # customers waiting for service in the queue
𝑟2
𝐸[𝐿𝑞 ] =
(1 − 𝑟)
 Average wait time spent in the queue per costumer
𝜆
𝐸[𝑇𝑞 ] =
𝜇(𝜇 − 𝜆)
 Average time spent in the system
1
𝐸[𝑇] =
(𝜇 − 𝜆)
 Probability of exactly n customers in the system
𝑃(𝑛) = (1 − 𝑟)𝑟 𝑛
48
Example2:
Vehicles arrive to a parking garage at a rate of 50 veh/hr Exponentially distributed service rate with
mean of 2 veh/min. Find 1) The average # customers waiting in a queue, 2) The average waiting time
per customer 3) The average time spent in the queuing system.
Solution:
Arrival rate and service rate must be in the same units:
𝑣𝑒ℎ 60 𝑚𝑖𝑛
𝑆𝑒𝑟𝑣𝑖𝑐𝑒 𝑟𝑎𝑡𝑒 (𝜆) = 2 × = 120 𝑣𝑒ℎ/ℎ𝑟
𝑚𝑖𝑛 ℎ𝑟
𝜆 50
𝑟= = = .4167
𝜇 120
𝑟2 (.4167)2
1) 𝐸[𝐿𝑞 ] = (1−𝑟)
= (1−.4167)
= .2976 𝑝𝑒𝑜𝑝𝑙𝑒
𝜆 50
2) 𝐸[𝑇𝑞 ] = = = .00595 ℎ𝑟 = .857 min = 21.4 𝑠𝑒𝑐
𝜇(𝜇−𝜆) 120∗(120−50)
1 1
3) 𝐸[𝑇] = (𝜇−𝜆)
= = .0143 ℎ𝑟 = .679 min = 51.4 𝑠𝑒𝑐
120−50
Example:
Vehicles travling during a peak hour (8 AM to 9 AM) arrive at a rate of 4,800 vehicles/hr. The vehicles
travel in 3 lanes with a capacity of 2,000 vehicles/hr/ln. At 8:15, an accident occurs that blocks one lane.
At 8:45, the accident is cleared.
Find:
a. The maximum queue length
b. The maximum vehicle wait time
c. The total vehicle delay resulting from the incident.
_____________________________________________________________________________________
2
Examples from CE327 Dr.Conways class
49
a. Maximum queue length = 3600-3200 = 400 vehicles (Solved graphically)

or
Maximum queue length = 4800*.5-4000*.5 = 400 vehicles
b. Time spent in queue = horizontal distance from arrival curve to departure curve
Max occurs at point when accident is cleared = 8:45-8:40 = 5 minutes
c. Total delay measured in vehicle-hours = area between the curves (see graphics on next page)
Green Triangle = .5*4000 vehicles*(5/6)hr = 1666.67 vehicle-hrs
Yellow Triangle = .5*2000 vehicles*(1/3)hr = 333.33 vehicle-hrs
Orange Rectangle = 2000 vehicles * (1/3) hr = 666.67 vehicle-hrs
Pink Triangle = .5*2000 vehicles*(1/2) hr = 500 vehicle-hrs
Area Between Curves = 1666.67-(333.33+666.67+500) = 166.67 veh-hrs
50
8 Shockwave
Shockwave notes were posted in pdf format to the blackboard.
9 Signal Timing
These notes were posted in pdf format to the blackboard.
51
10 Signal Design
Notes:
In the absence of traffic, the vehicle can be expected to travel at the free speed of the roadway. We can plot
the trajectory of the vehicle simply by a line in time – space that has a slope equal to the free speed, as
shown here.
Trajectory of vehicle
with free speed
Sf
Sf
f
S
Time
In the presence of other traffic on the roadway, we can expect that the speed of travel will decrease to a
speed that is dependent on the density of traffic. As a result, the vehicle trajectory will look something like
shown here. As you can see, under these conditions it takes the vehicle longer to cover the same distance;
this extra time falls under the general category of delay, and is specifically referred to as the delay due to
traffic or, simply, traffic delay.
52
with reduced speed
due to traffic
x
x
Sf
x
x
x
x
f
S
x
Time
When we add signalization to the intersection, we introduce the possibility of additional delay in the form
of having travel interrupted by having to wait at a red signal until the green signal is given.
53
x
with reduced speed

due to traffic
x
with free speed

x
Sf
x with reduced speed
x
due to traffic and

signal delay
x
x
x
Time
Once given the green indication, the vehicle will continue its travel along the roadway at the prevailing
speed in the presence of traffic.
In order to fully understand delay, we’ll first have to introduce a few concepts that turn out to be useful in
the calculation of intersection delay. The first of these is the notion of saturation flow rate. Let’s start by
representing the green phase for a particular traffic movement on a time-space diagram. Here we designate
the three basic components of the phase for the movement, the green, yellow, and all red indications.
Together, they represent the complete phase length for the movement. Observe the vehicle trajectories
displayed in the figure. When the signal first turns green, the first vehicle in the queue will experience some
delay in reacting to the green indication simply based on the driver’s reaction time. (This time may
approach zero in the case in which the driver anticipates the green.) Once the driver reacts to the green by
depressing the accelerator, the vehicle will begin to accelerate and enter the intersection. The second
vehicle in line also reacts to the green indication, but is constrained in its movement by the vehicle ahead
of it; once the first vehicle begins to move, the second vehicle can begin to accelerate toward the stop line
from its position in the queue. The time between the first vehicle crossing the stop line and the second
vehicle crossing the stop line is called the vehicle headway (expressed in seconds), and designated as h2 .
The time headway between the second vehicle crossing the stop line and the third is designated as h3, and
so forth.
In general, the time headways between the first few vehicles is greater than those between vehicles further
back in the queue since, once the first few vehicles accelerate to a comfortable speed, the rest of the vehicles
in line effectively become a “train” moving at a constant speed by the time they reach the stop line. In fact,
after about the third or fourth vehicle in line, the headways between vehicles approach a constant value,
that we designate as h, and call the “steady state headway.” The steady state headway represents the
54
minimum time difference between successive vehicles crossing the stop line; its reciprocal thus represents
the maximum flow rate (per lane of traffic), called the “saturation flow rate.”
The “startup lost time” is simply the total difference between the actual headways, hi , and the steady state
headway h, of vehicles in the queue.
Typical values for the headways range between 2 – 4 seconds.
The “effective green time” is the time during which flow through the intersection at the saturation flow rate
can actually take place. It includes both (a portion of) the actual green display plus some use of the yellow
period following the green display. The portion of the actual green display that is consumed by the startup
lost time is deducted from the actual green display time.
The “total lost time” is simply the difference between the phase length and the effective green time.
More often than not, the amount of time added to the green by vehicles using a portion of the yellow as an
“extended green” period, just about compensates for the startup lost time. So, in all but unusual cases, the
effective green time can be assumed to approximately equal the actual green display time. In such cases,
the total lost time for any phase will simply be the yellow time plus any all-red clearance time.
h7
h1  h2  h3  h4  h5  h6  h7  h
h6
x
h5
3600
h4 S  Saturation Flow Rate 
h
h3
h2
x
h1 Yellow
“All Red” Clearance
x
tl  Total Lost Time

 t  t g
x
x
h h h h h h
Time
x
Startup Lost Time Effective Green Time

tstart tg
Phase Length t
55
There are several features of the combined arrival and departure patterns under signalization that are of
interest to us: 1) the time that it takes to dissipate the queue once green is displayed, 2) the total stopped
delay incurred by vehicles during any particular cycle of the signal, 3) the average stopped delay per vehicle
during any particular cycle of the signal, and 4) the proportion of vehicles that have to stop during any
particular cycle of the signal. All of these features can be extracted from the geometry of the arrival-
departure patterns
Let C denote the cycle length (in seconds); t g the effective green time for the particular phase in question,
and tr the effective red period during the cycle (under normal conditions, the yellow interval + any all red
clearance for the particular phase +the sum of the phase lengths for all other movements conflicting with
the phase in question). During any particular cycle, the total volume of vehicle arrivals can be grouped
under one of three conditions of vehicle arrival and discharge: 1) those vehicles that arrive on red and are
stopped in the queue when the signal turns green, numbering VRS; 2) those vehicles that arrive on green but
are forced to join the back of the queue (i.e., at the end of the line of ) before proceeding through the
intersection, numbering VGq; and 3) those vehicles that arrive on green after the queue has been dissipated
and proceed through the intersection unimpeded, numbering VGG. Note that only those vehicles included
in categories 1) and 2) experience intersection delay; and, because they form a queue, they are all
discharged at the maximum (saturation) flow rate, S. The vehicles comprising category 3) experience no
intersection delay and are discharged at the same rate as their arrival rate, qA.
Note that the total number of vehicles that join the queue is equal to those that arrive during the red interval,
VRS, plus those that arrive on green but join the back of the queued-up vehicles that are being discharged
during the time tq , or an amount equal to VGq. So, the total number of vehicles that end up in the queue is
qA(tr+tq). These vehicles are discharged at the saturation flow rate during the time interval t q. Using the
figure shown, following equation between Saturation flow rate and green time can be written:
q t
S  tq  q A  (t R  tq )  tq  A R t 
 q
S
A t

 t
R
; 
q
1  q S  (1   )
R A
S  qA q
A
S
56
Maximum vehicle discharge Vehicle arrival pattern
qA
S
Traffic Volume (Vehicles)
Vehicle departure pattern qA  (tr  tq )

S
VGG
tq
VGq
S tr tg
VRS
Time
Discharge Rate
0 S qA
C
Total Delay (D) at each cycle for each phase is computed by the area between arrival and departure graphs.
qA  (tr  tq )
S
tr tq
1 1
DS   q A  (t R  tq ) 2   q A  (t R  tq )  tq
2 2
1
  q A  (t R  tq )  (t R  tq )  tq 
2
1
  q A  (t R  tq )  t R
2
57
1    tR 
DS   q A  t R   tR
2  (1   ) 
1  t  (1   )    t R 
  qA   R   tR
2  (1   ) 
1 q  (t ) 2
  A R
2 (1   )
Lets define:
gi
i 
C
qi 
Xi   i  degree of saturation
 i  Si  i
Total Delay:
qi  C 2  1  i 
2
Di  , having this total delay, average delay per vehicle ‘d’ is computed by:
2  (1  i  X i )
C  1  i 
2
di 
2  (1  i  X i )
Example1
Let’s next look at an example of how we use what we’ve learned about uniform delay to time the signals at
a simple intersection. Consider the following intersection under two-phase signal control, with constant
uniform arrivals. For simplicity, we assume that no turns are allowed, i.e., all traffic is through traffic
(NEMA phases 2, 4, 6, and 8).
1500 vph
6
1
7
8
3
4
5
2
6 8
400 vph
300 vph 4 2
1600 vph
58
For this simple phasing, the corresponding ring diagram looks like this, or equivalently, like this. The basic
feature of a two-phase signal is that the greens for the north and southbound traffic come on simultaneously,
and terminate simultaneously. Likewise for the east-west movements. We’ll assume that the saturation
flow rates for the movements are these. Nominally, these rates would correspond to two lanes of traffic for
the NEMA 2 and 6 movements, and a single lane of traffic for the NEMA 4 and 8 movements.
1 2 3 4
5 6 7 8
Saturation Flow Rates
NEMA Phase Si (vph)
2 4000
4 2000
6 4000
8 2000
Solution:
Recall total delay for direction i in intersection is computed by following equation:
q  C 2  1  i 
2
Di  i
2  (1  i  X i )
gi
i 
C
 i  X i  qi Si
qi
Xi 
 i  Si
59
In this particular example, g 2  g6 and g 4  g8 . So,  2   6  say  26 and  4  8  say  48 .
Let’s first calculate the values of qi Si :
NEMA Phase qi (vph) Si (vph) qi Si
2 1600 4000 0.40

4 300 2000 0.15
6 1500 4000 0.375
8 400 2000 0.20
D   Di  D2  D4  D6  D8
i
1600  C 2  (1   26 )2 300  C 2  (1   48 )2 1500  C 2  (1   26 )2 400  C 2  (1   48 )2

D   
2  (1  0.40) 2  (1  0.15) 2  (1  0.375) 2  (1  0.20)
Simplifying Assumption: No Lost Time
( g2 or g4 )  ( g6 or g8 )  C
( g2 or g6 ) ( g 4 or g8 ) C
i
i   26   48 
C

C
 1
C
 26   48  1   48  1   26
D 1600  (1   26 )2 300  ( 26 )2 1500  (1   26 )2 400  ( 26 )2

   
C2 2  (1  0.40) 2  (1  0.15) 2  (1  0.375) 2  (1  0.20)
D
2
 1333  (1   26 ) 2  176.47  ( 26 ) 2  1200  (1   26 ) 2  250  ( 26 ) 2
C
 2533  (1  2 26   26 )  426.47  26
2 2
 2533  26
 5066of  26 2959.47  26 D , say  26
2 *
To find the value that minimizes , we simply take the
derivative of the above expression with respect to  26 and set it equal
to zero; i.e.,
 C
d D 2
0
D
 2533  5066  26  2959.47 26
2
d 26 C2
60
 26
*
 C
d D 2
 5066  5918.94  26
*
 0   26
*

5066
 0.85
d 26 5918.94
 26
*
  1   26*  1  0.85  0.15

*
48
For 100 sec Cycle, the average delay for vehicles at each NEMA phase of the intersection is computed
by following equation:
C  1   i  C  1   i  100 sec  1   i 
2 2 2
di   
2  (1   i  X i ) 2  (1  gi Si ) 2  (1  gi Si )
100  (1  0.85) 2
d2   1.875 sec
2  (1  0.40)
100  (1  0.85) 2
d6   1.8 sec
2  (1  0.375)
100  (1  0.15) 2
d4   42.5 sec
2  (1  0.15)
100  (1  0.15) 2
d8   45.2 sec
2  (1  0.20)
However, there is a bit of a hidden problem here that has been masked by somewhat blindly using the
formulas that we developed without fully recognizing at least one critical assumption.
If you refer to the geometric development, you will appreciate that in determining the time required to
dissipate the queue, tq , (which later entered our formula for delay) we specifically relied on the presumption
that tq  t g ; otherwise the time required to dissipate the queue would have carried over into the next red
period, and the queue would have continued to build indefinitely, yielding a value tq   .
Such conditions are called “oversaturated;” our formulas hold only for degrees of saturation  1 (or for
“undersaturated” conditions). So, in effect, our optimization is not unconstrained, but rather must satisfy
the constraint that tq  t g or, equivalently, that the degree of saturation, X i , for each phase must be strictly
 1 . Let’s return to our solution; with 26  0.85,48  0.15
61
NEMA Phase qi Si i X i  qi (i  Si )
2 0.40 0.85 0.47

4 0.15 0.15 1.00
6 0.375 0.85 0.44
8 0.20 0.15 1.33
In order to resolve this issue, we should set the green time larger that the time required to relieve the queue:
 q A   (C  t )
 S 0.20  (100  t g ) 20  0.2t g
t g  tq   
g
   t g  20 sec
1  q A  (1  0.20) 0.8
 S 

Then 𝜃8 = 0.2
C  1    C  1  i  100 sec  1  i 
2 2 2
di  100  (1  i0.80)2 
d 2 2  (1  i  X i ) 2 3.33
 (1  sec
g i Si ) 2  (1  gi Si )
2  (1  0.40)
100  (1  0.80) 2
d6   3.2 sec
2  (1  0.375)
100  (1  0.20) 2
d4   37.6 sec
2  (1  0.15)
100  (1  0.20) 2
d8   40.0 sec
2  (1  0.20)
NEMA Phase qi Si i X i  qi (i  Si )
2 0.40 0.80 0.50

4 0.15 0.20 0.75
6 0.375 0.80 0.47
8 0.20 0.20 1.00
Example 2: Setting Max Green

In the following intersection, suppose we wanted to ensure that, with a 100 second cycle length,
qi
95% of the time NEMA Phase 2 would operate at a degree of saturation ( X i  ) not greater than
 i  Si
0.85. if the arrival has Poisson distribution, what would be the max green for this phase?
62
Phasing Diagram
1 2 3 4 1 2 3 4
5 6 7 8
 5 6 7 8
NEMA Phase qi (vph) Si (vph) qi Si
2 1600 4000 0.40

4 300 2000 0.15
6 1500 4000 0.375
8 400 2000 0.20
 t ( t ) i
Pi (t )  e 
i!
Suppose we wanted to ensure that, with a 100 second cycle length, 95% of the time NEMA Phase 2 would
operate at a degree of saturation not greater than 0.85. Assume Poisson arrivals.
( t ) i
Pi (t )  e t 
i!
The mean arrival rate for NEMA Phase 2 is 1600 vph    1600 vph or 0.44 vehicles per second. The
probability of i arrivals per 100 second cycle is
(0.44 100)i (44)i

Pi (100)  e0.44(100)   e44 
i! i!
63
Poision Probability of Exactly i Vehicles Arriving
0.07
0.06
0.05
Probability
0.04
0.03
0.02
0.01
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Number of Vehicles
64
Poision Probability of Vehicles Arriving Being Less Than Number on Abscissa
1.2
1
0.95
0.8
Probability
0.6
0.4
0.2
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Number of Vehicles
From this, we see that we are 95% sure that the total number of vehicles arriving on NEMA 2 during any
particular cycle of length 100 seconds will be less than or equal to 56 vehicles, or an equivalent hourly flow
of 56  3600 100  2016 vph.
gi
i 
C
qi 
Xi   i  degree of saturation
 i  Si  i
Or,
q2  C 2016 100 50.4
X2   
g 2  S2 g 2  4000 g2
For X 2  0.85 ,
50.4
g2   59 sec.
0.85
65
11 Road Geometric design:

Radius of the curves on the roads are designed such that they meet the minimum requirements for safe u-
turns and turns for design vehicles.
11.1 Design vehicles classification:
66
67
68
11.2 Turning radius:
69
Angle of turn Design vehicle Simple curve Simple curve radius with taper
in degree radius (ft) Radius (ft) Offset(ft) Taper L:T
90 P 30 20 2.5 10:1
SU 50 10:1
WB-40 10:1
Intermediate Semitrailer
WB-50 15:1
Intermediate Semitrailer
WB-62 30:1
Interstate Semitrailer
WB-67 30:1
Interstate Semitrailer
WB-100T 15:1
WB-109D 15:1
45 P 50 - - -
SU 75 - - -
WB-40 120 - - -
WB-50 175 120 2 15:1
WB-62 230 145 4 15:1
WB-67 250 145 4.5 15:1
WB-100T 200 115 2.5 15:1
WB-109D - 200 4.5 20:1
70
11.3 Cross-sections
The primary consideration in the design of geometric cross sections for highways, runways, and taxiways
is drainage. Highway cross sections consist of traveled way, shoulders (or parking lanes), and drainage
channels. Following figure illustrates the cross section of Two-lane highway with ditches.
(source: http://www.mhhe.com/engcs/civil/banks/graphics/chap4.pdf )
For high-type roadways (such as freeways), cross-slopes are normally 1.5 to 2.0 percent. Shoulders or
parking lanes slope away from the centerline at 2 to 6 percent. As a general rule, superelevated sections
will be constructed in a single plane (including shoulders) if the rate of superelevation exceeds the normal
cross slope of the shoulder.
Where ditches are used, foreslopes should normally be 1:4 or flatter to prevent damage to vehicles or injury
to occupants when traversed. Backslopes (cut or fill slopes) are discussed in more detail in Chapter 5. These
are normally on the order of 1:2 or 1:1.5 except in rock cuts, where they may sometimes be even vertical.
Use of 1:3 or flatter ditch backslopes facilitates the use of motorized equipment for maintenance, however;
also, composite slopes (that is, two different slopes at different distances from the roadway) are sometimes
used.
Standard lane widths are normally 3.6 m (12 ft.), although narrower lanes are common on older roadways,
and may still be provided in cases where the standard lane width is not economical. Shoulders or parking
lanes for heavily traveled roads are normally 2.4 to 3.6 m (8 to 12 ft.) in width; narrower shoulders are
sometimes used on lightly traveled roads.
11.4 VERTICAL ALIGNMENT
Recommended standards for maximum grades, Percent
Type of terrain Freeways Rural Highways Urban Highways
Level 3-4 3-5 5-8
Rolling 4-5 5-6 6-9
Mountainous 5-6 5-8 8-11
Source: from ‘A policy on Geometric design of highways and streets Vertical curves;
71
Vertical tangents with different grades are joined by vertical curves such as the one shown in Figure 4.7.
Vertical curves are normally parabolas centered about the point of intersection (P.I.) of the vertical tangents
they join. Vertical curves are thus of the form
𝑟𝑥 2
𝑦 = 𝑦0 + 𝑔1 𝑥 +
2
y =elevation of a point on the curve in meter
y0 = elevation of the beginning of the vertical curve (BVC)
g1 = grade just prior to the curve in percent
x =horizontal distance from the BVC to the point on the curve meter
r = rate of change of grade
𝑔2 − 𝑔1
𝑟=
𝐿
English standards:
𝑔1 𝑥 𝑟𝑥 2
𝑦 = 𝑦0 + +
100 200
y =elevation of a point on the curve in ft.

y0 = elevation of the beginning of the vertical curve (BVC), in ft.
g1 = grade just prior to the curve in percent
x =horizontal distance from the BVC to the point on the curve, in ft.
r = rate of change of grade
𝑔2 − 𝑔1
𝑟=
𝐿
72
11.4.1 How to identify the minimum length of the curve?

One of the main factors to find the length of a vertical curve is sight distance. In the following sections the
relevant equations to compute sight distance in metric and english system are explained.
11.4.2 Stopping Sight distance:
𝑉2
𝑆𝑆𝐷 = 1.47 𝑉. 𝑡 + (English system)
30(𝑓±𝐺)
𝑉2
𝑆𝑆𝐷 = 0.278 𝑉. 𝑡 + (metric)
254(𝑓±𝐺)
SSD = required stopping sight distance, m or ft.

V = speed, kph or mph
t = perception-reaction time, sec., typically 2.5 sec. for design
f = coefficient of friction, typically for a poor, wet pavement, f = a/g.
G = grade, decimal.
Depending on the grades of the tangent lines the vertical curve can be crest or sagged curve. If the curve is
crest the minimum length of the curve in order to follow the sight distance is:
𝐴𝑆 2
2, 𝑤ℎ𝑒𝑛 𝑆 ≤ 𝐿
200(√ℎ1 + √ℎ2 )
𝐿𝑚𝑖𝑛 = 2
200(√ℎ1 + √ℎ2 )
{2𝑆 − , 𝑤ℎ𝑒𝑛 𝑆 ≥ 𝐿
𝐴
S = sight distance
L = vertical curve length
A = absolute value of the algebraic difference in grades, in percent, |g1 _ g2|
h1 = height of eye
73
h2 = height of object
For the ‘sag’ curve the minimum distance is computed as follows:
𝐴𝑆 2
, 𝑤ℎ𝑒𝑛 𝑆 ≤ 𝐿
𝐿𝑚𝑖𝑛 = 120 + 3.5𝑆
120 + 3.5𝑆
{ 2𝑆 − 𝐴
, 𝑤ℎ𝑒𝑛 𝑆 ≥ 𝐿
11.4.3 Comfort criteria based on AASHTO:

𝐴𝑉 2
𝐿≥
395
L = length of vertical curve, m
A = g2 _ g1, percent
V=design speed, km/h
𝐴𝑉 2
𝐿≥
46.5
L = length of vertical curve, ft
A = g2 _ g1, percent
V=design speed, mph
11.4.4 Example1: Minimum length of a crest vertical curve

A crest vertical curve is to be designed to join a +3 percent grade with a -3 percent grade at a section of a
2-lane highway. Determine the minimum length of the curve if the design speed of the highway is 60mph
74
and S<L. Assume that the perception-reaction time is 2.5 sec and the deceleration rate for the braking is
11.2 ft/s2.
𝑉2
𝑆𝑆𝐷 = 1.47 𝑉. 𝑡 + 𝑎 (English system)
30( ±𝐺)
𝑔
602
𝑆𝑆𝐷 = 1.47 ∗ 60 ∗ 2.5 + 11.2 = 598.1 ft
30( −0.03)
32.2
ℎ1 = 3.5, 𝐻2 = 2 𝑓𝑡
𝐴𝑆 2
𝐿_ min = = 994.3ft
2158
11.4.5 Example 2: Design of crest vertical curve,

A crest vertical curve joining a +3 percent and -4 percent grade is to be designed for the speed of 75m/h.
minimum length for this speed is 2184 ft. If the tangent intersect at station 345+60.00 at an elevation of 250
ft, determine stations and elevations of the BVC and EVC.
2184
Station of BVC : 34560 – = 33468, so the location of the station will be 334+68.
2
Station of EVC: 33468 + 2184 = 35652, so the location of station at EVC is 356 + 52.
Elevation of BVC = 250 – 0.03*2184/2 = 217.42
75
Station Distance to BVC (x) y0+g1.x rx^2/200 Y (Curve Elevation)

334+68 0 217.24 0 217.24
335+00 32 218.2 -0.01641 218.1836
336+00 132 221.2 -0.27923 220.9208
337+00 232 224.2 -0.86256 223.3374
338+00 332 227.2 -1.76641 225.4336
339+00 432 230.2 -2.99077 227.2092
340+00 532 233.2 -4.53564 228.6644
341+00 632 236.2 -6.40103 229.799
342+00 732 239.2 -8.58692 230.6131
343+00 832 242.2 -11.0933 231.1067
344+00 932 245.2 -13.9203 231.2797
345+00 1032 248.2 -17.0677 231.1323
345+60 1092 250 -19.11 230.89
346+00 1132 251.2 -20.5356 230.6644
347+00 1232 254.2 -24.3241 229.8759
348+00 1332 257.2 -28.4331 228.7669
349+00 1432 260.2 -32.8626 227.3374
350+00 1532 263.2 -37.6126 225.5874
351+00 1632 266.2 -42.6831 223.5169
352+00 1732 269.2 -48.0741 221.1259
353+00 1832 272.2 -53.7856 218.4144
354+00 1932 275.2 -59.8177 215.3823
355+00 2032 278.2 -66.1703 212.0297
356+00 2132 281.2 -72.8433 208.3567
356+52 2184 282.76 -76.44 206.32
11.5 Horizontal Alignment:

11.5.1 Some definitions:
ARC
The arc defines the curve in terms of the angle subtended at the center by a circular arc 100 ft in length.
76
𝑅𝜋𝐷0 5729
100 = ,𝑅 = 0
180 𝐷𝑅
Chord:
50
𝑅=
sin 𝐷𝑐0 /2
The Chord defines the curve in terms of the angle subtended at the center by a chord of 100 ft.
11.5.2 Circular curve:
77
R = radius
PC = the point at which curve begins and is called point of curve
PT = point of tangent, the point at which curve ends
PI = point of intersection
T = tangent length
∆ = deflection angle
M = middle ordinate
E = external distance
2𝜋𝑟∆
𝐿=
360
∆
𝑇 = 𝑅. tan
2
∆
𝑀 = 𝑅 − 𝑅. 𝐶𝑜𝑠 ( )
2
𝑅
𝐸= −𝑅
∆
𝐶𝑜𝑠 ( )
2
∆
𝐶 = 2𝑅. 𝑠𝑖𝑛 ( )
2
78
How to set out the curve:
11.5.3 Example:
The deflection angle of a 4 degree curve is 550 25 . If the PC is locateds at station (238+44.75), determine
the length of the curve and the station of PT. Also determine the deflection angles and chords for setting
out the chord at whole stations from PC.
5729 5729
radius of curve    1432.4 ft
D 4
R
length of curve   1385.42 ft
180
Station PT is located at (238+44.75)+(13+85.42) = 252+30.17
79
Station arc length (Arc length) /L deflection angle deflection angle/2 deflection angle/2 in degree
238+44.75 0 0 0 0 0
239 55.25 0.03988 2.209996012 1.104998 1 618
240 155.25 0.11206 6.209988794 3.104994 3 618
241 255.25 0.18424 10.20998158 5.104991 5 618
242 355.25 0.25642 14.20997436 7.104987 7 618
243 455.25 0.328601 18.20996714 9.104984 9 618
244 555.25 0.400781 22.20995992 11.10498 11 618
245 655.25 0.472961 26.2099527 13.10498 13 618
246 755.25 0.545142 30.20994549 15.10497 15 618
247 855.25 0.617322 34.20993827 17.10497 17 618
248 955.25 0.689502 38.20993105 19.10497 19 618
249 1055.25 0.761682 42.20992383 21.10496 21 618
250 1155.25 0.833863 46.20991661 23.10496 23 618
251 1255.25 0.906043 50.2099094 25.10495 25 618
252 1355.25 0.978223 54.20990218 27.10495 27 618
252+30.17 1385.42 1 55.4167 27.70835 27 4230
11.5.4 Transition Curves:
𝐿5 𝐿9
𝑋= 𝐿− + +⋯
40𝐴4 3456𝐴8
𝐿3 𝐿7 𝐿11
𝑌= − + +⋯
6𝐴2 336𝐴6 42240𝐴10
80
𝐴2 = 𝑅𝐿
11.5.5 Superelevation:
The purpose of superelevation or banking of curves is to counteract the centripetal acceleration produced
as a vehicle rounds a curve.
𝑉2
𝑒= −𝑓
𝑔. 𝑅
81
82
12 Pavement design:
(Most of the notes and examples are based on ‘Pavement Analysis and Design’, 2 nd edition, Yang H Huang,
Pearson publication. )
12.1 Flexible Pavement
Flexible pavements will transmit wheel load stresses to the lower layers by grain-to-grain transfer through
the points of contact in the granular structure.
The wheel load acting on the pavement will be distributed to a wider area, and the stress decreases with the
depth. Taking advantage of this stress distribution characteristic, flexible pavements normally has many
layers. Hence, the design of flexible pavement uses the concept of layered system. Based on this, flexible
pavement may be constructed in a number of layers and the top layer has to be of best quality to sustain
maximum compressive stress, in addition to wear and tear. The lower layers will experience lesser
magnitude of stress and low quality material can be used. Flexible pavements are constructed using
bituminous materials. These can be either in the form of surface treatments (such as bituminous surface
treatments generally found on low volume roads) or, asphalt concrete surface courses (generally used on
high volume roads such as national highways). Flexible pavement layers reflect the deformation of the
lower layers on to the surface layer (e.g., if there is any undulation in sub-grade then it will be transferred
to the surface layer). In the case of flexible pavement, the design is based on overall performance of flexible
pavement, and the stresses produced should be kept well below the allowable stresses of each pavement
layer.
(Source: http://www.civil.iitb.ac.in/tvm/1100_LnTse/401_lnTse/plain/plain.html)
- The first asphalt roadway in US constructed in 1870 at Newark
- The first sheet-asphal pavement, hot mixture of asphalt cement with clean, angular, graded sand
and mineral filler was laid on Pensylvania Ave., Washington D.C.
- As of 2001, there are about 2.5 million miles paved roads in US, 94% are asphalt surfaced.
12.1.1 Design methods:

- Empirical methods with or without a soil strength test
-
83
- Limiting shear failure method: determine the thickness of pavements so that shear failure will not
occur.
- Limiting deflection method: determine the thickness of pavements so that the vertical deflection
will not exceed the allowable limit.
- Regression method based on pavement performance or road test: a good example is AASHTO
method
- Mechanistic-empirical method: lab test to predict field-performance
The following types of construction have been used in flexible pavement:

- Conventional layered flexible pavement,
- Full - depth asphalt pavement, and
- Contained rock asphalt mat (CRAM).
12.1.2 What is HMA?

It is “hot mixture asphalt”, is synonymous with the commonly used “asphalt concrete”. It is an asphalt
aggregate mixture produced at a batch or drum mixing facility that must be mixed, spread, and compacted
at an elevated temperature.
12.1.3 Layered flexible pavement

Typical layers of a conventional flexible pavement includes seal coat, surface course, tack coat, binder
course, prime coat, base course, sub-base course, compacted sub-grade, and natural sub-grade (Figure 2).
Seal Coat:
Seal coat is a thin surface treatment used to water-proof the surface and to provide skid resistance.
Tack Coat:
Tack coat is a very light application of asphalt, usually asphalt emulsion diluted with water. It provides
proper bonding between two layers of binder course and must be thin, uniformly cover the entire surface,
and set very fast.
Prime Coat:
Prime coat is an application of low viscous cutback bitumen to an absorbent surface like granular bases on
which binder layer is placed. It provides bonding between two layers. Unlike tack coat, prime coat
penetrates into the layer below, plugs the voids, and forms a water tight surface.
84
Figure 2: Typical cross section of a flexible pavement
Surface course
Surface course is the layer directly in contact with traffic loads and generally contains superior quality
materials. They are usually constructed with dense graded asphalt concrete(AC). The functions and
requirements of this layer are:
- It provides characteristics such as friction, smoothness, drainage, etc. Also it will prevent the
entrance of excessive quantities of surface water into the underlying base, sub-base and sub-grade,
- It must be tough to resist the distortion under traffic and provide a smooth and skid- resistant riding
surface,
- It must be water proof to protect the entire base and sub-grade from the weakening effect of water.
Binder course
 The reasons to use Binder: 1) HMA is too thick to be compacted in one layer, 2) The binder course
generally consists of aggregates having less asphalt and doesn't require quality as high as the surface
course, so replacing a part of the surface course by the binder course results in more economical
design.
 If Binder is more than 3 in., it is placed in two layers.
 This layer provides the bulk of the asphalt concrete structure.
 Its chief purpose is to distribute load to the base course

Base course
 The base course is the layer of material immediately beneath the surface of binder course and it
provides additional load distribution and contributes to the sub-surface drainage.
 It may be composed of crushed stone, crushed slag, and other untreated or stabilized materials.
Sub-Base course
The sub-base course is the layer of material beneath the base course and the primary functions are to provide
structural support, improve drainage, and reduce the intrusion of fines from the sub-grade in the pavement
structure.
85
If the base course is open graded, then the sub-base course with more fines can serve as a filler between
sub-grade and the base course. A sub-base course is not always needed or used. For example, a pavement
constructed over a high quality, stiff sub-grade may not need the additional features offered by a sub-base
course. In such situations, sub-base course may not be provided.
Sub-grade
The top soil or sub-grade is a layer of natural soil prepared to receive the stresses from the layers above. It
is essential that at no time soil sub-grade is overstressed. It should be compacted to the desirable density,
near the optimum moisture content.
12.1.4 Full-Depth Asphalt Pavement
 Are constructed by placing one or more layers of HMA directly on the subgrade or improved
subgrade.
 The most cost effective and dependable type of asphalt pavement for heavy traffic.
 It is quite popular in areas where local materials are not available.
Asphalt Surface 2 to 4 in.

2 to 20 in.
Asphalt base
Prepared Subgrade
 Advantages:
o Time required for construction is reduced
o They have no permeable granular layers to entrap water and impair performance.
o They provide uniformity in the pavement structure
o They are less effected by moisture or frost
12.2 Rigid Pavement

Rigid pavements have sufficient flexural strength to transmit the wheel load stresses to a wider area below.
A typical cross section of the rigid pavement is shown in Figure 3. Compared to flexible pavement, rigid
pavements are placed either directly on the prepared sub-grade or on a single layer of granular or stabilized
material. Since there is only one layer of material between the concrete and the sub-grade, this layer can be
called as base or sub-base course.
86
Figure 3: Typical Cross section of Rigid pavement

In rigid pavement, load is distributed by the slab action, and the pavement behaves like an elastic plate
resting on a viscous medium (Figure 4). Rigid pavements are constructed by Portland cement concrete
(PCC) and should be analyzed by plate theory instead of layer theory, assuming an elastic plate resting on
viscous foundation. Plate theory is a simplified version of layer theory that assumes the concrete slab as a
medium thick plate which is plane before loading and to remain plane after loading. Bending of the slab
due to wheel load and temperature variation and the resulting tensile and flexural stress.
12.3 Stresses and Strains in Flexible Pavements
12.3.1 Solutions by charts:
87
88
89
90
91
Example
Following figure shows a homogenous half-space subjected to tow circular load, each 10 in. in diameter
and spaces at 20 in. on centers. The pressure on the circular area is 50 psi. The half space has elastic modulus
of 10,000 psi and Poisson ratio 0.5. Determine the vertical stress, strain and deflection at point A, which is
located 10 in below the one circle.
10 in. 10 in.
50 psi 50 psi
E = 10,000 psi
10 in.
A 20 in.
a= 0.5, 5 = 50 psi, z = 10.

Stresses of left load:
r/a= 0, z/a= 2, from the graphs:
𝜎𝑧
= 0.28 , 𝑡ℎ𝑒𝑛 𝜎𝑧 = 0.28 ∗ 50 = 14
𝑞
𝜎𝑟 = 𝜎𝑡 = 0.016 ∗ 50 = 0.8
Stresses of right load:
r/a= 4, z/a= 2, from the graphs:
𝜎𝑧
= 0.0076 , 𝑡ℎ𝑒𝑛 𝜎𝑧 = 0.38
𝑞
𝜎𝑟 = 0.026 ∗ 50 = 1.3
𝜎𝑡 = 0
Then adding these pressures:
𝜎𝑧 = 14 + 0.38 = 14.38
𝜎𝑟 = 0.8 + 1.3 = 2.1
𝜎𝑡 = 0.8
[14.38−0.5(2.1+0.8)]
𝜖𝑧 = = 0.00129
10000
Deflection factor due to the left load is 0.68, and due to the right is 0.21. then deflection at point A:
5
𝑤 = (0.68 + 0.21) ∗ 50 ∗ = 0.022 𝑖𝑛.
10000
92
12.3.2 Numerical solutions:

When the load is applied over a single circular loaded area, the most critical stress, strain, and deflection
occur under the center of the circular area on the axis of symmetry, where 𝜏𝑟𝑧 = 0 and 𝜎𝑟 = 𝜎𝑡 and 𝜎𝑟 are
the principal stresses.
Flexible Plate
The load applied from tire to pavement is similar to flexible plate with a radius 𝑎 and a uniform pressure 𝑞.
The stresses beneath the center of the plate can be determined from:
𝑧3
𝜎𝑧 = 𝑞[1 − ]
(𝑎2 + 𝑧 2 )1.5
𝑞 2(1 + 𝜐)𝑧 𝑧3
𝜎𝑟 = [1 + 2𝜐 − 2 + ]
2 (𝑎 + 𝑧 2 )0.5 (𝑎2 + 𝑧 2 )1.5
1
𝜖𝑧 = [𝜎 − 𝜐(𝜎𝑟 + 𝜎𝑡 )]
𝐸 𝑧
1
𝜖𝑟 = [𝜎𝑟 − 𝜐(𝜎𝑧 + 𝜎𝑡 )]
𝐸
1
𝜖𝑡 = [𝜎 − 𝜐(𝜎𝑟 + 𝜎𝑧 )]
𝐸 𝑡
Vertical deflection:
(1 + 𝜐)𝑞𝑎 𝑎 1 − 2𝜐
𝑤= { 2 + [(𝑎2 + 𝑧 2 )0.5 − 𝑧]}
𝐸 (𝑎 + 𝑧 )
2 0.5 𝑎
Example:
If the Poisson ratio is 0.3, determine the stresses, deflection and strains at point A.
10 in.
50 psi
E = 10,000 psi
10 in.
Solution:
𝑎 = 5 𝑖𝑛. , 𝑞 = 50𝑝𝑠𝑖 , 𝑧 = 10𝑖𝑛.,
1000
Then 𝜎𝑧 = 50[1 − (25+100)1.5 ] = 14.2
10 1000
With 𝜐 = 0.3, 𝜎𝑟 = 25 [1 + 0.6 − 2.6 ∗ (125)0.5 + (125)1.5 ] = −0.25 the negative sign indicates tension.
93
50
𝜖𝑧 = 1.3 ∗ = 0.00144
10 1000
10000 [1 − 0.6 + 0.6 ∗ − ]
(125)0.5 (125)1.5
50
𝜖𝑟 = 1.3 ∗ = −0.00044
10 1000
20000 [1 − 0.6 − 1.4 ∗ + ]
(125)0.5 (125)1.5
50 ∗ 50
𝑤 = 1.3 ∗ = 0.0176
5 0.4
10000 [ + ]
(125)0.5 5[(125)0.5 − 10]
Rigid plates
If the load is applied to the rigid plate, the deflection will be the same at all points on the plate, but the
pressure distribution under the plate is not uniform.
Non- uniform pressure q(r) Uniform pressure q
The pressure distribution under a rigid plate is expressed as:

𝑞𝑎
𝑞(𝑟) =
2(𝑎2 − 𝑟 2 )0.5
q is the average pressure, which is equal to the total load divided by the area.
The smallest pressure is at the center and equal to one half of the average pressure. The pressure at the edge
of the plate is infinity.
94
The deflection of the plate is:

𝜋(1 − 𝜐 2 )𝑞𝑎
𝑤0 =
2𝐸
Example:
A plate loading test using a plate of 12 in. diameter was performed on the surface of the subgrade. A total
load of 8000 was applied to the plate at the deflection of 0.1 in. was measured. Assuming that the subgrade
has Poisson ratio 0.4, determine the modulus of the subgrade.
8000
Average pressure is 𝑞 = = 70.74 𝑝𝑠𝑖
36𝜋
6
𝐸 = 𝜋(1 − 0.16) ∗ 70.74 ∗ = 5600
2 ∗ 0.1
95
12.3.3 Two layered systems:

The vertical stress on the top of the subgrade is an important factor in pavement design. The allowable
stress on a subgrade depends on the strength or modulus of the subgrade. The stress in a two layere system
depend on the modulus ratio 𝐸1 ⁄𝐸2 and the thickness ratio ℎ1 ⁄𝑎. Figure shows the stress when ℎ1 ⁄𝑎 = 1
and 𝜈 = 0.5. (ℎ1 is the thickness of top layer)
The vertical stress on the interface between two layers is given in the following graph
Example:
A circular load having radius 6 in. and uniform pressure 80 psi is applied on a two layer system. The
subgrade has elastic modulus 5000 psi and can support a maximum vertical stress of 8 psi. if the HMA has
elastic modulus 500,000 what is the required thickness of a full-depth pavement? If a thin surface treatment
is applied on a granular base with elastic modulus 25,000 psi, what is the thickness of base course required?
96
Solution:
𝜎𝑐 𝑎
Give E1/E2 = 500,000/5000= 100; and = 0.1, from the figure, = 1.15 , then h1 = 5.2 in., which is the
𝑞 ℎ1
minimum thickness required for full depth.
𝜎𝑐 𝑎
Give E1/E2 = 25,000/5000= 5; and = 0.1, from the figure, = 0.4 , then h1 = 15 in., which is the
𝑞 ℎ1
minimum thickness required for full depth.
97
12.4 Traffic Loading:

Equivalent Single Axle Load (ESAL): 1 ESAL is equivalent to 18 kips (80 KN) load.
Equivalent axle load factor (EALF): factor used to convert the axle load to the ESAL.
How to convert dual tire to ESAL:
The following equation is shows how to compute the total load caused by traffic on the surface of the road.
In this equation, EAFL, is equivalent axle load factor which depends on the vehicle type and the distance
between the axles of the vehicles, etc.
98
 m 
ESAL    pi Fi   ADT 0 T  G  D  L  365 Y 
 i 1 
pi  percentageof total repetition for ithload group
Fi  EALFi , computeby AASHTO
T  truck percentage
A  average number of truck per axle
G  growth factor
D  directional distribution factor
L  lane distribution factor
Y  design period in years
99
12.4.1 Damage types:

The following table shows the damage type in different pavements and the cause for the damage:
Flexible Rigid Loading Environment
Alligator Cracking X X
Bleeding and Pumping X X X X
Block Cracking X X
Blow-Ups X X
Corner Break/Cracks X X
Corrugation/Shoving X X X
Faulting X X X
Fatigue Cracking/Potholes X X X
Joint Reflection Composite X
Joint Seal Damage X X X
Longitudinal Cracking X X X X
Map Cracking X X
Popouts X X
Raveling X X X
Rutting X X
Settlement X X X
Slab Cracking X X X
Spalling X X
Surface Attrition X X X
Surface Polish X X X X
Transverse Cracking X X X X
100

327 Profnotes 01 All

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

327 Profnotes 01 All

Uploaded by

Copyright:

Available Formats

Lecture Notes for CE327:

Transportation Systems Engineering

Instructor: Mahdieh Allahviranloo

11.5.3 Example: ........................................................................................................................ 79

1 Introduction to Transportation Engineering

Let’s start with some preliminaries:

𝑑(𝑥𝑖 (𝑡)) 𝑥𝑖 (𝑡 + ∆𝑡) − 𝑥𝑖 (𝑡)

Headway: (unit : time)

ℎ𝑖 (𝑥) = 𝑡𝑖+1 (𝑥) − 𝑡𝑖 (𝑥)

Gap: (unit :time)

Spacing (unit: length )

Spacing = clearance + length of vehicle

Traffic flow: (unit: vehicles/hour)

Space mean speed: (harmonic speed)

Time Mean Speed:

Space Mean Speed:

Fundamental law of traffic flow:

Flow = density * speed.

The fraction of time that a detector i4s covered by vehicles.

Types of detector outputs:

Presence Detector Output

Pulse Detector Output

3 Speed and density:

Replacing Kc in the equation of speed :

Question: Identify congested and uncongested areas in diagrams:

Greenshield Flow - Speed Relationship

Greenshield Flow - Density Relationship

𝑑 (𝐶. (𝐾𝑐 . ln 𝐾𝑗 − 𝐾𝑐 𝑙𝑛𝐾𝑐 ))

4 Car following models:

California DMV rule:

 x (t )(mph)   x n1 (t )(ft/sec) 

Safe braking distance:

For safe stopping:

dn1  tR  xn1 (t )  bn1

Example: Designing Ramp Metering

Loop Station 2 Loop Station 3

0.2 mile 1.5 miles

Station 1: 30-second Count Data

Station 1: 30-second Occupancy Data

Station 2: 30-second Count Data

Station 2: 30-second Occupancy Data

Station 3: 30-second Count Data

Station 3: 30-second Occupancy Data

Station 1: 30-second Count Data

q(vph)  (count ) 120

Station 1: 30-second Occupancy Data

q1  20 120  2400 vph

Station 2: 30-second Count Data

q(vph)  (count ) 120

Station 2: 30-second Occupancy Data

q2  30 120  3600 vph

Station 3: 30-second Count Data

Station 3: 30-second Occupancy Data

q3  30 120  3600 vph

q1  20 120  2400 vph q2  30 120  3600 vph q3  30 120  3600 vph

Region 1 Region 2 Region 3

0.2 mile 1.5 miles

MR = 3600 vph - 2400 vph = 1200 vph

Mean: average value

For a dice the sample space is {1,2,3,4,5,6}.

Examples of continuous r.v.:

𝜎(𝑥) = 𝐸(𝑥 2 ) − 𝐸(𝑥)2

5.2 Binomial PMF:

Probability of n success in m ‘independent’ trial each time having probability of Bernoulli.