PROJECT EXEMPLE

ANALYSIS OF VEHICLES ARRIVAL TIME AT A POINT

2. PRIMARY DATA PROCESSING
Objectives:
• Understanding statistical instruments to analyze experimental data.
• Acquiring new abilities to use the Excel programme;
• Acquiring new competencies to interpret experimental data.
STEPS IN THE ANALYSIS OF VEHICLES ARRIVAL TIME AT A POINT.
1. Establishing the types of vehicles and identifying the coefficients of equivalence.
Traffic streams are made up of all types of vehicles, which lead to the heterogeneous aspect
of traffic streams. When conducting traffic analyses it is compulsory to collect data on special
data sheets, for various types of vehicles, correlating these vehicles with the standard or
unitary vehicle, i.e. the passenger car and the other vehicles in the class (see the table below)
for which the coefficient of equivalence is 1,0.
The value of the coefficients of equivalence is given in STAS 7348-86, and the new standard
SR 7348:2001 as follows:
Table E.1: Types of vehicles and coefficients of equivalence
Type of vehicle Coefficient of
equivalence
Motorcycle with/without sidecar; scooters; light motorcycle; motor tricycle 0,5
Passenger cars with or without trailers; station wagons; minibuses; micro-
1,0
buses; trucks, vans;
Trucks with laden weight < 3 tones, with or without trailer; trucks with
laden weight > 3 tones, without trailer; buses without trailer; tour buses; 2,0
agricultural vehicles without trailer;
Trucks with laden weight > 3 tones with trailer; buses with one trailer;
3,5
tractors with trailer or semi-trailer;
In case trucks, tractors or agricultural vehicles with two or more trailers, for
1,5
each trailer it is added:
Articulated bus 4,0
Trolleybus 3,0
Motor tramway 4,0
Articulated tramway 6,0
For each tramway trailer 2,0
*) When establishing the coefficients of equivalence it was the passenger car that was
taken for the standard vehicle Vt.
Turning the data regarding the number of individual vehicles into standard vehicles.
The values obtained, expressed in standard vehicles are summed up on each line and
represent the number of vehicles arriving at the analyzed location every 10 second.
By grouping the values recorded in column (10) for time headways of ∆t=10 seconds, there
will be obtained the values corresponding to xi for ∆t=20 seconds, and ∆t=30 seconds.
Table E.2
Item Passenger 1 Buses 2 Trolleybuse 3 Trucks 2 t=10 t=20 t=30
cars s <3t s s s
(1) (2) (3) (4) (5) (6) (7) (8) (39 (10) (11) (12)
1 I 1 II 4 - 0 - 0 5
10
2 II 2 - 0 I 3 - 5 5 17
3 II 2 I 2 I 3 - 0 7
 PROJECT EXEMPLE

The processing of experimental data recorded on the data sheet is based on the
random character of traffic streams. Thus, the random variable X is defined.
The random variables used in traffic engineering are of two types:
• Discrete random variable (V.A.D.), for instance, vehicles arriving at a certain
location;
• Continuous random variables (V.A.C.), for instance, vehicle speed recorded at a
certain point along the roadway or space headways between vehicles in a traffic
stream, etc.
Table E.3
pag /xi 0 1 2 3 4 5 6 7 8
1 3 10 11 16 6 1 1 0 0
2 6 10 6 11 11 6 3 1 0
3 3 9 16 9 9 4 2 1 1
4 7 14 6 7 11 4 3 2 0
5 3 5 9 13 10 8 2 4 0
6 0 6 15 10 11 8 3 1 0
7 3 3 15 7 8 4 2 0 0
Ni N0=25 N1=57 N2=78 N3=73 N4=66 N5=35 N6=16 N7=9 N8=1
A discrete random variable may be described through a range of values, as follows:

⎛x x2 . . . xn ⎞
X = ⎜⎜ 1 ⎟,
⎝ p1 p2 . . . pn ⎟⎠
where xi is the value of the random variable argument, X, and pi is the probability of
argument occurrence or fi, which represents the occurrence relative frequency, on condition
n
that 0≥pi = fi ≤1, for any i value and ∑p
i =1
i =1.

or

⎛x x2 . . . xn ⎞
X = ⎜⎜ 1 ⎟⎟ ,
⎝ N1 N 2 . . . N n ⎠
n
with Ni≥0, natural numbers, for any i value and ∑ Ni =N.
i =1

In order to determine the occurrence probabilities (relative frequencies) it is necessary to

record the absolute empirical frequencies, Ni in the central table for ∆t=10 seconds, ∆t=20
seconds, and ∆t=30 seconds.
To make the recording process easier, the following steps are to be followed:
• a new Excel sheet is opened to record the values from the traffic data sheets for each
time headway (∆t=10 seconds, ∆t=20 seconds, and ∆t=30 seconds) and for each day
or hour, on one sheet.
• there are selected the time intervals of ∆t=10 seconds, ∆t=20 seconds, and ∆t=30
seconds establishing in this way the absolute frequencies Ni, occurrence of xi vehicles
of time interval. What is, in this case, the total number of observations
effectuated?
 PROJECT EXEMPLE

The table 4 comprises the results of recordings done on ……. at hour…… at time interval
of ∆t=10 seconds.
Table E.4:
xi Ni fi f∑> f∑< x=xi*fi s²=(xi-x)²*fi P(X=x) Ni Ni²/Ni
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
0 25 0,069444 0,069444 1 0 0,576222 0,056031 20,17126 30,98468
1 57 0,158333 0,227778 0,930556 0,158333 0,559944 0,161562 58,16232 55,86091
2 78 0,216667 0,444444 0,772222 0,433333 0,167999 0,232837 83,82146 72,58285
3 73 0,202778 0,647222 0,555556 0,608333 0,002893 0,223619 80,50302 66,19628
4 66 0,183333 0,830556 0,352778 0,733333 0,229745 0,161014 57,96489 75,14894
5 35 0,097222 0,927778 0,169444 0,486111 0,436727 0,092713 33,37662 36,70233
6 16 0,044444 0,972222 0,072222 0,266667 0,432486 0,04447 16,00933 15,99068
7 10 0,027778 1 0,027778 0,194444 0,471384 0,018276 6,579478 15,19877
∑ 360 1 1 2,880556 2,8774 356,5884 368,6654
n
3600 sec
From table it is noticed that ∑N
i =0
i =
10 sec
= 360 int ervals

On the basis of absolute frequencies the relative frequencies, fi may be determined.

Ni
fi =
N
which represents the occurrence probability of each headway. Thus, N0, represents the
occurrence probability of the headway of zero vehicles, N1, represents the occurrence
probability of the headway of one vehicle, N2, represents the occurrence probability of the
headway of two vehicles, etc.
we deduce,
n
∑ fi =1
i =0

In column 3, there are calculated the occurrence probabilities fi of headways of 10 seconds

when xi vehicles arrive.
Table E.5
Input data Empirical data processing Theoretical data processing
xi Ni fi f∑> f∑< x=xi*fi s²=(xi-x)²*fi P(X=x) Ni Ni²/Ni
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
0 25 0,069444 0,069444 1 0 0,576222 0,056031 20,17126 30,98468
1 57 0,158333 0,227778 0,930556 0,158333 0,559944 0,161562 58,16232 55,86091
2 78 0,216667 0,444444 0,772222 0,433333 0,167999 0,232837 83,82146 72,58285
3 73 0,202778 0,647222 0,555556 0,608333 0,002893 0,223619 80,50302 66,19628
4 66 0,183333 0,830556 0,352778 0,733333 0,229745 0,161014 57,96489 75,14894
5 35 0,097222 0,927778 0,169444 0,486111 0,436727 0,092713 33,37662 36,70233
6 16 0,044444 0,972222 0,072222 0,266667 0,432486 0,04447 16,00933 15,99068
7 10 0,027778 1 0,027778 0,194444 0,471384 0,018276 6,579478 15,19877
∑ 360 1 1 2,880556 2,8774 356,5884 368,6654

The determination of relative frequencies cumulated in increasing order, fΣ> and in

decreasing order, fΣ>, may be very useful for traffic data analysis. The calculus is presented
in column 4 and in column 5.
 PROJECT EXEMPLE

Table E.6
Input data Empirical data processing Theoretical data processing
xi Ni fi f∑> f∑< x=xi*fi s²=(xi-x)²*fi P(X=x) Ni Ni²/Ni
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
0 25 0,069444 0,069444 1 0 0,576222 0,056031 20,17126 30,98468
1 57 0,158333 0,227778 0,930556 0,158333 0,559944 0,161562 58,16232 55,86091
2 78 0,216667 0,444444 0,772222 0,433333 0,167999 0,232837 83,82146 72,58285
3 73 0,202778 0,647222 0,555556 0,608333 0,002893 0,223619 80,50302 66,19628
4 66 0,183333 0,830556 0,352778 0,733333 0,229745 0,161014 57,96489 75,14894
5 35 0,097222 0,927778 0,169444 0,486111 0,436727 0,092713 33,37662 36,70233
6 16 0,044444 0,972222 0,072222 0,266667 0,432486 0,04447 16,00933 15,99068
7 10 0,027778 1 0,027778 0,194444 0,471384 0,018276 6,579478 15,19877
∑ 360 1- 2,880556 2,8774 356,5884 368,6654

0,9

0,8

0,7

0,6
f>
0,5
f

f<
0,4

0,3

0,2

0,1

0
0 1 2 3 4 5 6 7 8
xi

Figure E.1: Graphs of cumulated frequencies

In relation to these two columns there may be designed the graphs of cumulated
frequencies according to the number of vehicles arriving within the time interval of 10
seconds. The graph may be represented through histograms.
Calculus of mean random variable.
n
Column 6 comprises the data necessary to the mean x = ∑ fi ⋅ x i = 2,88 vehicles arrive on
i =0
average within the interval of 10 seconds.
Table E.7
Input data Empirical data processing Theoretical data processing
xi Ni fi f∑> f∑< x=xi*fi s²=(xi-x)²*fi P(X=x) Ni Ni²/Ni
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
0 25 0,069444 0,069444 1 0 0,576222 0,056031 20,17126 30,98468
1 57 0,158333 0,227778 0,930556 0,158333 0,559944 0,161562 58,16232 55,86091
2 78 0,216667 0,444444 0,772222 0,433333 0,167999 0,232837 83,82146 72,58285
3 73 0,202778 0,647222 0,555556 0,608333 0,002893 0,223619 80,50302 66,19628
4 66 0,183333 0,830556 0,352778 0,733333 0,229745 0,161014 57,96489 75,14894
5 35 0,097222 0,927778 0,169444 0,486111 0,436727 0,092713 33,37662 36,70233
6 16 0,044444 0,972222 0,072222 0,266667 0,432486 0,04447 16,00933 15,99068
7 10 0,027778 1 0,027778 0,194444 0,471384 0,018276 6,579478 15,19877
∑ 360 1- 2,880556 2,8774 356,5884 368,6654
 PROJECT EXEMPLE

Table E.8.
Input data Empirical data processing Theoretical data processing
xi Ni fi f∑> f∑< x=xi*fi s²=(xi-x)²*fi P(X=x) Ni Ni²/Ni
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
0 25 0,069444 0,069444 1 0 0,576222 0,056031 20,17126 30,98468
1 57 0,158333 0,227778 0,930556 0,158333 0,559944 0,161562 58,16232 55,86091
2 78 0,216667 0,444444 0,772222 0,433333 0,167999 0,232837 83,82146 72,58285
3 73 0,202778 0,647222 0,555556 0,608333 0,002893 0,223619 80,50302 66,19628
4 66 0,183333 0,830556 0,352778 0,733333 0,229745 0,161014 57,96489 75,14894
5 35 0,097222 0,927778 0,169444 0,486111 0,436727 0,092713 33,37662 36,70233
6 16 0,044444 0,972222 0,072222 0,266667 0,432486 0,04447 16,00933 15,99068
7 10 0,027778 1 0,027778 0,194444 0,471384 0,018276 6,579478 15,19877
∑ 360 1- 2,880556 2,8774 356,5884 368,6654
Calculus of random variable dispersion, D2 or s2 or σ2. In our example noted with s2.
Comparing the mean with the dispersion.
Mean = 2,88055>Dispersion = 2,8774
Choosing the distribution model (random distribution)
Since Mean>Dispersion the binomial distribution is adopted.
Determining the probabilities that describe the proposed model (binomial distribution–column
8).
Table E.9.
Input data Empirical data processing Theoretical data processing
xi Ni fi f∑> f∑< x=xi*fi s²=(xi-x)²*fi P(X=x) Ni Ni²/Ni
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
0 25 0,069444 0,069444 1 0 0,576222 0,056031 20,17126 30,98468
1 57 0,158333 0,227778 0,930556 0,158333 0,559944 0,161562 58,16232 55,86091
2 78 0,216667 0,444444 0,772222 0,433333 0,167999 0,232837 83,82146 72,58285
3 73 0,202778 0,647222 0,555556 0,608333 0,002893 0,223619 80,50302 66,19628
4 66 0,183333 0,830556 0,352778 0,733333 0,229745 0,161014 57,96489 75,14894
5 35 0,097222 0,927778 0,169444 0,486111 0,436727 0,092713 33,37662 36,70233
6 16 0,044444 0,972222 0,072222 0,266667 0,432486 0,04447 16,00933 15,99068
7 10 0,027778 1 0,027778 0,194444 0,471384 0,018276 6,579478 15,19877
∑ 360 1- 2,880556 2,8774 0,990523 356,5884 368,6654

Table E.10.
Input data Empirical data processing Theoretical data processing
xi Ni fi f∑> f∑< x=xi*fi s²=(xi-x)²*fi P(X=x) Ni’ Ni²/Ni
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
0 25 0,069444 0,069444 1 0 0,576222 0,056031 20,17126 30,98468
1 57 0,158333 0,227778 0,930556 0,158333 0,559944 0,161562 58,16232 55,86091
2 78 0,216667 0,444444 0,772222 0,433333 0,167999 0,232837 83,82146 72,58285
3 73 0,202778 0,647222 0,555556 0,608333 0,002893 0,223619 80,50302 66,19628
4 66 0,183333 0,830556 0,352778 0,733333 0,229745 0,161014 57,96489 75,14894
5 35 0,097222 0,927778 0,169444 0,486111 0,436727 0,092713 33,37662 36,70233
6 16 0,044444 0,972222 0,072222 0,266667 0,432486 0,04447 16,00933 15,99068
7 10 0,027778 1 0,027778 0,194444 0,471384 0,018276 6,579478 15,19877
∑ 360 1- 2,880556 2,8774 0,990523 356,5884 368,6654
 PROJECT EXEMPLE

The model, i.e. binomial distribution, is validated with the testing criterion noted „ Chi-
square Criterion” or „ χ2 criterion”.
For this very aim there is determined the absolute theoretical frequency in column (9) in
the above table, calculated with the relation:

N i' = P ( X = x ) ⋅ N
To verify the accuracy of calculations, i.e. whether the sum of probabilities calculated in
the preceding column reaches the value 1, the sum of absolute theoretical frequencies will
reach the value corresponding to N, the total number of observations
The next step is to determine the χ2 value given by the sum of squared empirical
frequencies deviations as compared to the theoretical frequencies, with the following equation:

χ 2
=∑
n(N i − N i' )
2

=∑
n N i2
−N .
i =0 N i' i =0 N i'

N i2
In column (10) there are determined only the ratios , which, summed up in the last line,
N i'
will allow the determination of the χ2criterion value.
Table E.11.
Input data Empirical data processing Theoretical data processing
xi Ni fi f∑> f∑< x=xi*fi s²=(xi-x)²*fi P(X=x) Ni’ Ni²/Ni’
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
0 25 0,069444 0,069444 1 0 0,576222 0,056031 20,17126 30,98468
1 57 0,158333 0,227778 0,930556 0,158333 0,559944 0,161562 58,16232 55,86091
2 78 0,216667 0,444444 0,772222 0,433333 0,167999 0,232837 83,82146 72,58285
3 73 0,202778 0,647222 0,555556 0,608333 0,002893 0,223619 80,50302 66,19628
4 66 0,183333 0,830556 0,352778 0,733333 0,229745 0,161014 57,96489 75,14894
5 35 0,097222 0,927778 0,169444 0,486111 0,436727 0,092713 33,37662 36,70233
6 16 0,044444 0,972222 0,072222 0,266667 0,432486 0,04447 16,00933 15,99068
7 10 0,027778 1 0,027778 0,194444 0,471384 0,018276 6,579478 15,19877
∑ 360 1- 2,880556 2,8774 0,990523 356,5884 368,6654

Figure E.2: Graphical representation of χ2criterion

The criterion χ2 is graphically represented in the following figure, which, as noticed,

depends on the ν value of the degrees of freedom.
The graph is characterized by the following characteristics:
 PROJECT EXEMPLE

• The area under the curve is equal with the unit;

• The starting value is χ2=0.

Figure E.3: Graphical representation of χ2 variation for δ = 0,05 and α = 0,95.

The curve is not symmetrical; when ν (the number of observations) increases, the curve
becomes similar with the normal one. It is to be stated that the number of degrees of
freedomν = n − l − 1 , where n – is the number of grouping intervals (number of classes Si),
and l – number of parameters of the theoretical distribution. For example, if the distribution is
binominal or Poisson, distributions that have a single parameter, then l = 1 and
ν = n − 1 − 1 = n − 2 degrees of freedom. Yet, if the law is ordinary, with two parameters, then
l = 2 and ν = n − 2 − 1 = n − 3 degrees of freedom. Therefore, the number of degrees of
freedom is a statistical notion closely connected with the amount of information available in
the research conducted.
When χ2 test is applied it is adequate that N≥50, Ni≥5 and 10 ≤ n ≤ 20. When they have
less than five elements the extreme classes merge with the adjacent classes and the number
υ decreases correspondingly (ν = n − 3 ). If only one class merge with the adjacent one, then
ν = n − 2 . Provided all classes have more than five elements, thenν = n − 1 .
The test efficiency increases when each class comprises approximately the same amount
of data. The test also has a great discrimination power in case of symmetrical distributions
and provides good results when the normality of normal empirical distributions is verified.
The application of χ2 criterion for a certain safety threshold δ, is reduced to be verified if:
P ( χ 2 < χ 02 ) = 1 − δ ,
where χ 02 , is the value adopted from the table below for the number of degrees of freedom
ν = n − 3 , where n is the number of random variables components or the number of grouping
intervals of data observed.
If χ 2 < χ 02 , it is considered that between the empirical and theoretical distribution there is
concordance with the confidence coefficient α = 1 − δ . The table E.12 presents the χ2
variation according to ν for δ = 0,05 and α = 0,95.
Graphs
• Representation of the relative frequency fi and of the probability function P(X=x).
• Representation of absolute experimental and theoretical frequencies under the form of
histograms.
• Comparison of graphs corresponding to different traffic lanes for which data recording
was conducted simultaneously.
 PROJECT EXEMPLE

• Comparison of distributions obtained for the three time headways, for each recording.
The traffic values obtained allow at the same time to draw up graphs necessary for
comparisons making reference to:
• importance of various categories of vehicles for each traffic lane, in case of a
recording;
• importance of various categories of vehicles for the same traffic lane, for recordings
conducted on different days and at different hours;
• analysis of arrival times of vehicles designated to public transportation (for example
the frequency of buses or trolleybuses).
Table E.12: χ2 values for a certain safety threshold δ.
ν χ 02,995 χ 02,99 χ 02,975 χ 02,95 χ 02,05 χ 02, 025 χ 02, 01 χ 02, 005 ν

1 0,000 0,000 0,001 0,004 3,841 5,024 6,635 7,879 1

2 0,010 0,020 0,051 0,103 5,991 7,378 9,210 10,597 2
3 0,072 0,115 0,216 0,352 7,815 9,348 11,345 12,838 3
4 0,207 0,297 0,484 0,711 9,488 11,143 13,277 14,860 4
5 0,412 0,554 0,831 1,145 11,070 12,832 15,086 16,750 5

6 0,676 0,872 1,237 1,635 12,592 14,449 16,812 18,548 6

7 0,989 1,239 1,690 2,167 14,067 16,013 18,475 20,278 7
8 1,344 1,646 2,180 2,733 15,507 17,535 20,090 21,955 8
9 1,735 2,088 2,700 3,325 16,919 19,023 21,666 23,589 9
10 2,156 2,558 3,247 3,940 18,307 20,483 23,209 25,188 10

11 2,609 3,053 3,816 4,575 19,675 21,920 24,725 26,757 11

12 3,074 3,571 4,404 5,226 21,026 23,337 26,217 28,300 12
13 3,565 4,107 5,009 5,892 22,362 24,736 27,688 29,819 13
14 4,075 4,660 5,629 6,571 23,685 26,119 29,141 31,319 14
15 4,601 5,229 6,262 7,261 24,996 27,488 30,578 32,801 15

16 5,142 5,812 6,908 7,962 26,296 28,845 32,000 34,267 16

17 5,697 6,408 7,564 8,672 27,587 30,191 33,409 35,718 17
18 6,265 7,015 8,231 9,390 28,869 31,526 34,805 37,156 18
19 6,844 7,633 8,907 10,117 30,144 32,852 36,191 38,582 19
20 7,434 8,260 9,591 10,851 31,410 34,170 37,566 39,997 20

21 8,034 8,897 10,283 11,591 32,671 35,479 38,932 41,401 21

22 8,643 9,542 10,982 12,338 33,924 36,781 40,289 42,796 22
23 9,260 10,196 11,689 13,091 35,172 38,076 41,638 44,181 23
24 9,886 10,856 12,401 13,848 36,415 39,364 42,980 45,558 24
25 10,520 11,524 13,120 14,611 37,652 40,646 44,314 46,928 25

26 11,160 12,198 13,844 15,379 38,885 41,923 45,642 48,290 26

27 11,808 12,879 14,573 16,151 40,113 43,194 46,963 49,645 27
28 12,461 13,565 15,308 16,928 41,337 44,461 48,278 50,993 28
29 13,121 14,256 16,047 17,708 42,557 45,722 49,588 52,336 29
30 13,787 14,953 16,791 18,493 43,773 46,979 50,892 53,672 30

Interpreting graphs and conclusions with regard to vehicles arrival times at a point.
Students will come up with personal conclusions, according to their capacity to analyze
and synthesize the data collected; students will partake in team discussions and will interpret
the results so as to draw up a common report.