TRAFFIC ENGINEERING

SAB3843
STATISTICS & TRAFFIC
DATA ANALYSIS

CHE ROS BIN ISMAIL

and
OTHMAN BIN CHE PUAN
 Statistics
• Statistics is the branch of scientific method which
deals with the data obtained by counting or
measuring the properties of population of natural
phenomena.
• This branch of study includes:
– The process of collecting data
– The study of manipulation and arrangement of
figures using mathematical processes, and
– Interpretation of the figures
 Traffic Data Analysis
Understanding of traffic data – types of data, data
presentation and description, validity, basic
statistical distribution of the data, etc.
 Sampling – Population vs. Sample
• ‘Population’ refers to all the measurements that could be
made.
• A ‘sample’ is a subset of measurements selected from the
population.
• Samples are tested in order to make inferences about the
properties of the population. Therefore, it is vital to be clear
which population is of interest.
E.g., when considering the level of car ownership, are we
interested in a particular group of people, or a population
of an area. Hence, information in the sample is used to
make an inference about the population.
 Accuracy of Sampling
• Sampling is necessary because it is usually impracticable to test
the entire population.
• Each and every sample must be selected in a random manner
so that it is representative of the population from which it is
drawn
• A value arrived at by sampling is absolutely accurate only for
the sample itself. For the population which it represents, a
sample can only give an estimate whose accuracy is expressed
in terms of probability.
• Therefore, the greater the size of the sample (i.e. number of
observations) the greater the confidence that can be placed on
the estimate for the population.
 Data Description
• Data is an information which in general has 2 main
characteristics;
a) Qualitative – involves non-numerical data,
e.g. consider “YES” or “NO” as an answer to
questionnaires
b) Quantitative – involves numerical data
 Quantitative Data
Two forms:
i. Discrete data – figures obtained from
counting processes, usually in integer form.
ii. Continuous data – figures obtained from
measurements, can be in any forms.
Two ways of describing data are:
• Numerically
• Graphically
 Numerical Descriptive Measures for
Describing Data
Two most common measures are:
1. Measures of Central Tendency
2. Measures of scatter
 Measures of central tendency
1. Arithmetic Mean (or simply known as ‘mean’)
– Mean of a set of measurements is the sum of the
measurements divided by the total number of
measurements: n

∑( f X )
i =1
i i
X= n

∑f i =1
i

where i = 1,2,3,…….. n
 Measures of central tendency
2. Median
The median of a set of measurements is the middle value
when the measurements are arranged in order of
magnitude. It, therefore, divides a histogram and a
frequency polygon into two equal areas.

E.g., consider this set of data: 1, 3, 4, 7, 8, 9, 10

Median = 50th percentile = P50

3. Mode – is the measurement that occurs most often

Mode = Mean – 3(Mean – Median)
 Measures of Scatter
1) Range – the range of a set of measurements is
defined to be the difference between the largest
and the smallest measurements of the set.

eg. 15, 15, 20, 21, 30, 12, 11, 5, 40, 40, 26

Range = largest – smallest = 40 – 5 = 35

 Measures of Scatter
2) Percentile – the rth percentile of a set of n measurements
arranged in order of magnitude is that value that has r% of
the measurements below it.
100
90 (a) 85th percentile of X = a

80
Cumulative frequency, %

70
60 (b) 50th percentile of X = b
50
40
30
20 (c) 15th percentile of X = c
10
0
(c) (b) (a)
X
 Measures of Scatter
3) Variance (S2) n

∑ i
X 2
− nX 2

S2 = i =1
n −1
or
2
n
 n

∑ fX i ∑ fX i
2

S 2 = i =1 −  i =1 
∑ f  ∑ f 

 
 Measures of Scatter
4) Standard Deviation (SD)

SD is a measure of the average deviation of readings from

their mean.
SD = variance
5) Standard Error (SE)
SD
SE =
n
Graphical Method for Describing Data (typical diagrams)
1) Histogram

Frequency (numbers)

speed class (km/h)

 Graphical Method for Describing Data
2) Cumulative Frequency Curve

100
90
Cumulative frequency, %

80
70
60
50
40
30
20
10
0
Speed (Xi), km/h
Example 1 – Spot speed analysis
Analyse the following spot speed data based on a sample of 172
vehicles traversing a section of sub-urban roadway.
Speed class Frequency
(km/h) fi
20 - 25 1
25 - 30 3
30 - 35 6
35 - 40 13
40 - 45 25
45 - 50 34
50 - 55 31
55 - 60 27
60 - 65 18
65 - 70 9
70 - 75 4
75 - 80 1
Total 172
Solution 1 – tabulate data as follows
Speed class Mid point Frequency Cum. Cum.

v (km/h) vi Fi Freq. Freq. (%) Fi * vi Fi * vi2

20 - 25 22.5 1 1 0.6 22.5 506.25

25 - 30 27.5 3 4 2.3 82.5 2268.75

30 - 35 32.5 6 10 5.8 195.0 6337.5

35 - 40 37.5 13 23 13.4 487.5 18281.25

40 - 45 42.5 25 48 27.9 1062.5 45156.25

45 - 50 47.5 34 82 47.7 1615.0 76712.5

50 - 55 52.5 31 113 65.7 1627.5 85443.75

55 - 60 57.5 27 140 81.4 1552.5 89268.75

60 - 65 62.5 18 158 91.9 1125.0 70312.5

65 - 70 67.5 9 167 97.1 607.5 41006.25

70 - 75 72.5 4 171 99.4 290.0 21025

75 - 80 77.5 1 172 100.0 77.5 6006.25

Total 600 172 8745 462325

Solution 1 – compute basic statistical facts for the data
n

• Mean speed = ∑( f v )
i =1
i i = 8745/172 = 50.84 km/h
v= n

∑f i =1
i

• Std deviation:  n 2

 n 
 ∑ fi vi  ∑ fi vi
2
   2

 462325  8745 
SD =  i =1 −  i =1  =
 −   
 ∑f  ∑f    172  172  
   
   

= 10.16 km/h

Can you compute the variance and standard error for the data?
What can you say about this result?
 Solution 1 – plot the histogram for the data
Examine the plot & answer
these:
• Do you think that the 40

general shape of the plot 35

is a typical of a normal
30
distribution data?

F re q u e n c y (n u m b e rs )
25
• Do you think that the data
20
should follow a normal
distribution curve? 15

10
• Is the expected mean lies
5
somewhere in the middle
of the plot? 0
5

0
-2

-3

-3

-4

-4

-5

-5

-6

-6

-7

-7

-8
• What if it is not so?
20

25

30

35

40

45

50

55

60

65

70

75
speed class (km/h)
 Solution 1 – plot cumulative curve
100

90 (a) 85th percentileof v = a

80

70
Cumulative frequency, %

60
(b) 50th percentile of v = b
50

40

30

20 (c) 15th percentile of v = c

10

0
(c) (b) (a)
0 10 20 30 40 50 60 70 80
Speed, km/h

• Compare the calculated mean & median, should both values are equal? Which
value to report?
• Establish the required speeds at various percentiles, in what way these values will
be used?
Example 2
Evaluate the following traffic data obtained for 7 consecutive
days on a stretch of road section.

Day Traffic volume (veh/day)

Monday 3231
Tuesday 3011
Wednesday 3137
Thursday 3247
Friday 3065
Saturday 3240
Sunday 1530
 Solution 2
Compute the average traffic volume per day:
Day Traffic volume (veh/day)
By definition, the average volume
Monday 3231 of 2923 veh/day can be reported
as the PLH or Purata Lalu Lintas
Tuesday 3011
Harian for the road.
Wednesday 3137
Thursday 3247 But, certain traffic analyst may
Friday 3065 remove the data taken on
Saturday 3240 Sunday because we have 6 data
points with more than 3000 &
Sunday 1530
only 1 data is much lesser. PLH
is not representative.
= 20461/7 The reported PLH would be
Average = total traffic/7
= 2923 veh/day = 18931/6 = 3155 veh/day
Example 3
Two series of one–week traffic counts were carried out on a
stretch of rural road and the data obtained are as follows:
March 2006 October 2006
Day
(veh/day) (veh/day)
Monday 12500 10300
Tuesday 10500 12000
Wednesday 15200 13000
Thursday 13400 14500
Friday 16000 15200
Saturday 10500 8500
Sunday 8000 10200

(a) Determine the ADT and AADT on that particular road section.
(b) State the AADT in PCU/day if the average composition is 45% cars, 20%
medium lorries, 10% buses, 7% heavy lorries & 18% motorcycles.
 References
1. Garber, N.J., Hoel, L.A., TRAFFIC AND HIGHWAY ENGINEERING,4th Edition,
SI Version., Cengage Learning (2010).
2. Currin, T. R., INTRODUCTION TO TRAFFIC ENGINEERING – A Manual for
Data Collection and Analysis, Brooks/Cole (2001).
3. Kadiyali, L.R., TRAFFIC ENGINEERING AND TRANSPORT PLANNING,
Khanna Publishers (1987) .
4. Othman Che Puan. Modul Kuliah Kejuruteraan Lalu Lintas. Published for
Internal Circulation. (2004).
5. Dorina Astana, Othman Che Puan, Che Ros Ismail, TRAFFIC ENGINEERING
NOTES, Published for Internal Circulation. (2011)

25