Traffic

Studies
A- Review for
Statistics
B- Volume Studies
C- Speed Studies
 Basic Statistics: Review
• What are Statistics?
• The art of abstracting Real World via
sampling and deriving general
Real World “estimates” that describes the Real
World at a certain degree of
confidence. Descriptive
Decision
Sample Date Measures for
& Math. Making
Real World
Data Reduction Model &
(@ deg. Of
Design
confidence)
 Basic Statistics: Review
• When do we need Statistics?
When we can not measure all the data values for the
population.
• Before starting: What do we need to address?
• Sample Size (how many measurements are
sufficient?)
• What Confidence should I have in the results?
• What statistical model distribution (math model)
that better describes the observed data?
• Did a traffic engineering solution affected the
status of the Real World significantly?(before &
after analysis)
 Basic Statistics:
Sample Reduction & Visualization
Frequency Histogram Cumulative Frequency
Percentile %
35 100%
90%
30
80%

Cumulative Frequency%
25 70%
60%
Frequency

20
50%
15
40%

10 30%
20%
5
10%
0 0%
55 65 75 85 95 105 115 125 135 55 65 75 85 95 105 115 125 135

Observations Observations
 Basic Statistics:
Common Statistical Estimators
Measures of Central Tendency:

Mean: Median:
N
- Is the middle value of
x i
x  i 1 all the sample data
N ( i.e. 50% of the data
where : are above this value)
x  sample mean Mode:
Is the value that occurs
xi  observation (i )
most frequently
N  number of Observations
 Basic Statistics:
Common Statistical Estimators
Measures of Dispersion:
Variance: Standard Deviation:
N N
 ( x  x)
i
2
 i
( x  x ) 2

S 2  i 1 S  S2  i 1
( N  1) ( N  1)
where : where :
2
S  sample Variance
S  S tan dard Deviation
x  sample mean
S 2  sample Variance
xi  observation (i )
N  number of Observations
 Basic Statistics:
Common Statistical Estimators
Measures of Dispersion:
Coefficient of Variation: Skewness:
The ratio between the Describes the asymmetry in
standard deviation and the the data sample.
mean.
STD
Cvar 
x (mean  mode)
where : Skewness 
STD
Cvar  Coefficient of Variation
STD  Standard deviation
x  sample mean
 Basic Statistics:
Common Statistical Estimators
MS Excel Functions:
Mean  = average(range array)
Mode  = mode(range array)
Median  = median(range array)
Variance  = var(range array)
Standard Deviation  = stdev( range
array)
Skewness  = skew( range array)
 Basic Statistics:
Standard Error, True Mean & Sample Size
Standard Error:
The standard error (E) in the sample mean ( X ) is function in the
sample size and the standard deviation of the population ( the
sample SD can be used instead):

E
N
where
 - is the standard deviation for the population
The SD for the sample can be used instead
N - is the sample size
 Basic Statistics:
Standard Error, True Mean & Sample
Size
True Mean: m
The standard error (E) for the sample mean ( X ) is assumed to follow a
normal distribution around the true mean ( m ).
Hence:   X 1.00 E (at Degree of Confidence 67%)
  X 1.96 E (at Degree of Confidence 95%)
  X 3.00 E (at Degre of Confidence 99.5%)
where
E - is the standard error of the sample
X - is the sample mean
 Basic Statistics:
Standard Error, True Mean & Sample
Sample Size: Size
For a given allowable error ( err ) and a specific degree of
confidence , the sample size ( N ) can be determined as following:
2
 SD 
N   (at Degre of Confidence 67%)
 err 
2
 SD 
N  (1.96)   (at Degre of Confidence 95%)
 err 
2
 SD 
N  (3.00)   (at Degre of Confidence 99.5%)
 err 
where
SD - is the standard deviation of the sample
err - maximum allwable error in the true mean
Speed Studies
 Speed Studies: Spot Speed
Studies
Spot Speed Studies:
• Is defined as the average speed of vehicles passing a point
on a highway. This is also known as the time mean speed.
• Usually conducted in free flow condition and not during
congestion, where the flow rate is:
750-1000 veh/hr/ln  for freeway
<500 veh/hr/ln  for other types
Speed Definition of Interest:
- Average or time mean speed
- Standard Deviation
- 85th % speed
- Median speed
 Speed Studies: Spot Speed
Studies
Uses of Spot Speed Data:
• To determine speed limit for applications
• To assess speed limit enforcement
• Specific Applications:
 For Level of Service (LOS) Assessment
 For Signal timing: Estimation of Yellow/All Red times.
 To determine appropriate sight distance
 For safety and accidents analysis
 Speed Studies: Measurement Techniques

Manual Method: The Simple Stopwatch Method :

By using stopwatch and defining two reference points
with known distance (d) between the two points.
Then, Speed = d / t (m/s)

40 m
Advantages: Simple
Disadvantages:
High error due to stopwatch
depressing time variations.

Class Example
 Speed Studies: Spot Speed
Studies
Doppler Radar (Speed Gun):
It uses Doppler’s effect for speed measurements.
How Does it work?
• The radar transmits a pack of waves with initial frequency fini and initial
wavelength lini ,
• Due to the motion of the target vehicle, the wavelength of the reflected
1
waves lref will be longer
ref  ini  or*Vshorter
target than the
depnding on initial s direction lini
wavelength
the target'
f ini
Transmitted wave
Radar
lini Target

lref
Reflected wave
 Speed Studies: Spot Speed
Studies
Doppler Radar (Speed Gun):
Advantages:
High Accuracy, but the readings must be corrected for aiming
angle.

Disadvantages:
Difficult to conceal, drivers associate Radar
with police which may cause them to
slow their speeds down and yielding
inaccurate results.
ElDessouki 17

CONEN 442 Transportation Engineering S2024

 Speed Studies: Spot Speed
Studies
Doppler Radar (Speed Gun): Readings Correction
Speed
TrueSpeed 
cos 

ElDessouki 18

CONEN 442 Transportation Engineering S2024

 Spot Speed Studies:
Data Reduction & Analysis
The speed data is analyzed and reported as
following:
A-35Graphical: Frequency HistogramThe
& Accumulative
shown speed %
data has
30
a Bimodal
25
Distribution
Frequencey

20

15

10

0
50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160
Speed (km/hr) MODE = 105 km/hr
 Spot Speed Studies:
Data Reduction & Analysis
The speed data is analyzed and reported as following:
A- Graphical: Accumulative %
100%

90%
85
% 80%

70%
Accumulative %
60%

50%

40%

30%

20%
15 Median
% 10%
0% ElDessouki 20
20 40 60 80 100 120 140 160 180
15th% Speed Speed (km/hr) 85th% Speed
CONEN 442 Transportation Engineering S2024
 Spot Speed Studies:
Data Reduction & Analysis

The speed data is analyzed and reported as following:

B- Quantitative:
1- Mode
2- Median
3- Mean
4- Standard Deviation (SD)
5-85th% & 15th% Speeds
6- Pace(15 km/hr band)
 Spot Speed Studies:
Data Reduction & Analysis
The speed data is analyzed and reported as following:
C- Precision and Confidence Intervals:
Note: Most spot speed data tend to be normally distributed (however, this
might not applicable in the shown example histogram), then:
Standard Error: SD
E
N
where
SD - is the standard deviation for the sample
N - is the sample size

  X 1.96 E (at Degree of Confidence 95%)

True Mean: where
E - is the standard error of the sample
X - is the sample mean
 Spot Speed Studies:
Data Reduction & Analysis

The speed data is analyzed and reported as following:

D- Sample Size for Prescribed Precision with Confidence
Intervals:
If the Prescribed precision was set to be +/- e, the needed
sample should not be less than the following number of
2
observations N @ a Ndegree
 of
 (1.96) 
SDconfidence
 95%:
 
 e 
where
SD - is the standard deviation of the sample
e - the needed precision i.e. the error in the true mean
Speed Studies: Before and After Analysis
Before and After Analysis:
• Usually carried out to evaluate the effectiveness of applying a specific
measure on the prevailing speed in an area or a segment.
• The before and after is basically a comparison testing between two
samples , with the objective of finding that the difference between the
two samples is significant or not.
• Hypothesis Testing False/
True/True
False
Any hypothesis test, has 4 possible outcomes:
1- Test Result: True, and Reality: True
2- Test Result: False, and Reality: False False/True True/False

3- Test Result: False, and Reality: True Error Type II

4- Test Result: True, and Reality: False Error Type I
Error Type I - must be avoided at all expenses
 Speed Studies: Before and After Analysis
The Statistical Testing (Z-test):
First, calculate the Pooled Standard Deviation for before & after
samples: S12 S 22
S  
N1 N2
where
S  Pooled Standard Deviation
S1 & S 2  Standard Deviation before & after resp.
N 1 & N 2  Sample size before & after resp.

Second: Calculate Zd the Standard Normal distribution

approximation X 1  Xthe
Z d  for 2  d
Observed
o difference between the before &
after samples: S
Z d  Normalized Diffirence between before & after
X 1 & X 2  Mean speed before & after resp.
S  pooled Standard Deviation
d o  hypothesized population mean difference
 Speed Studies: Before and After Analysis

The Statistical Testing (Z-test) cont.:

Third, we use the normal distribution curve to find the
probability that a value equal to or less than Zd , assuming
that both samples are normally distributed , then:
A) If Prob.( Z<= Zd ) > 0.95 , that means the observed
reduction in speed is statistically significant.
B) If Prob.( Z<= Zd ) < 0.95 , that means the observed
reduction in speed is statistically insignificant.
For case A, that implies also that there is a 5% chance that
the observed difference in mean speed will be exceeded.
 Speed Studies: Before and After Analysis
Example:
The following is the before and after summary for speed
enforcement project that was deployed with a target of
reducing average speed to 60 mph.
Before After
Mean Speed: 65.3 63 (mph)
SD: 5 6 (mph)
N: 50 60 observation
Solution
S12 S 22 52 6 2
Zd 
65.3  63 0
2.19
S     1.05mph
N1 N 2 50 60 1.05

P(Z<2.19) = 0.9857 = 98.57 % >95%

ElDessouki 27
Then: The observed reduction in speed was statistically
significant CONEN 442 Transportation Engineering S2024
 Speed Studies: Before and After Analysis
Example (cont.):
Now, the question did we reach the target?
6
  X 1.96 E 63.0 1.96 * 63.0 1.52mph
60
That means the true mean is between: (61.48 - 64.52) mph
The Answer is NO, the reduction is not sufficient, and we did
not achieve the 60 mph target