,

INDIAN INSTITUTE OF TECHNOLOGY, DELHI

HAUS KHAS, NEW DELHI-110016

CIVIL ENGINEERING DEPARTMENT

Course Number: CEL 742
Traffic Engineering

REPORT
Lab 2: Vehicle Arrival Models: Count

Submitted To:
Dr. Kalaga Ramchandra Rao
(Associate Professor)
Civil Engineering Department, IIT Delhi

Submitted By:
Mohit Kohli
2015CEP2096
M Tech Transportation Engineering, Semester-I (2016-2017)
 LAB-2: Vehicle Arrival Models: Count

Objectives: To determine the probability of how many vehicles will arrive at a left turn pocket at a
signalized signalized intersection during peak periods.

Need for the Study: Modelling arrival of vehicle at section of road is an important step. The vehicle
arrival is a random process, thus this study estimates how vehicles arrive at a section. Vehicle arrival shall
be characterized statistically as several times vehicles comes together, while other time, they arrive
sparsely.

General Overview:
Poisson distribution:
It is used to model the random process, the number of vehicles arriving in a given time period. It can
represent several traffic flow situations where the flow exhibits random behavior. Let h1, h2, ... etc
indicate the headways, then as mentioned earlier, they take some real values. Hence, these headways or
inter arrival time can be modelled using some continuous distribution. Also, let t1, t2, t3 and t4 are four
equal time intervals, then the number of vehicles arrived in each of these interval is an integer value.

Fig : Illustration of Vehicle arrival modeling

The probability density function of the Poisson distribution is given as:

Normal Distribution:
Normal or Gaussian distribution is a very common continuous probability distribution. Normal
distributions are important in statistics and are often used in the civil engineering to represent real-
valued random variables whose distributions are not known. it states that averages of random
variables independently drawn from independent distributions converge in distribution to the normal, that
is, become normally distributed when the number of random variables is sufficiently large. The probability
density of the normal distribution is:

Where:

is mean or expectation of the distribution (and also its median and mode).
is standard deviation
is variance
Study Components:
Poisson distribution is used to determine the expected number of cars that arrive at a left turn pocket in
unit length of time. The spacing between the vehicles is the distance between the front bumper of one car
to the successive car. This spacing phenomenon is studies using Poisson distribution. Using the Chi-
Square test, the goodness of fit can be tested. The other two distributions generally used are normal
distribution and binomial distributions.

Some important terms:

Mean : is the average of all numbers and is sometimes called the arithmetic mean.
Median : is the middle number in a sequence of numbers.
Variance : is the expectation of the squared deviation of a random variable from its mean, and
it informally measures how far a set of (random) numbers are spread out from their mean.
Standard Deviation : is a measure of the spread of scores within a set of data.

Field Work and Data Collection:

A busy intersection having an exclusive left turn is selected. The number of vehicles arriving
during the selected interval is recorded in the data sheet. The data is used to perform Poisson distribution
and normal distribution. The intersection between Aurbindo marg and Gamal Abdel Nasser Marg is
selected as shown in the map below:

Fig : Site for Left Turn Count

 WORK SHEET: LAB 2
DATE: 22-08-2015 DAY: MONDAY WEATHER: SUNNY

STUDENT NAME: MOHIT KOHLI ENTRY NUMBER: 2015CEP2096

STARTING TIME: 12.00PM ENDING TIME: 1PM

Result:
Arrival Pattern

No of Vehicles
Sl. No. Time
2 Wheeler Car Auto/LCV Bus HV Total
1 12:00:00 PM 7 7 4 0 0 18
2 12:01:00 PM 7 5 4 0 0 16
3 12:02:00 PM 3 6 2 0 0 11
4 12:03:00 PM 4 0 2 0 0 6
5 12:04:00 PM 0 0 1 0 0 1
6 12:05:00 PM 9 11 4 2 0 26
7 12:06:00 PM 1 6 3 0 0 10
8 12:07:00 PM 0 3 1 0 0 4
9 12:08:00 PM 6 7 1 0 0 14
10 12:09:00 PM 1 2 0 0 0 3
11 12:10:00 PM 7 2 3 2 0 14
12 12:11:00 PM 3 2 0 0 0 5
13 12:12:00 PM 3 0 0 0 0 3
14 12:13:00 PM 1 0 1 0 0 2
15 12:14:00 PM 1 4 0 0 0 5
16 12:15:00 PM 2 10 8 1 0 21
17 12:16:00 PM 5 8 1 1 0 15
18 12:17:00 PM 2 5 1 0 0 8
19 12:18:00 PM 1 0 0 0 0 1
20 12:19:00 PM 2 1 0 0 0 3
21 12:20:00 PM 7 13 1 0 0 21
22 12:21:00 PM 1 9 3 0 0 13
23 12:22:00 PM 5 3 2 0 0 10
24 12:23:00 PM 0 0 0 0 0 0
25 12:24:00 PM 0 2 1 0 0 3
26 12:25:00 PM 1 5 2 0 0 8
27 12:26:00 PM 1 5 1 0 0 7
28 12:27:00 PM 2 2 1 0 0 5
29 12:28:00 PM 6 3 0 0 0 9
30 12:29:00 PM 3 1 3 0 0 7
31 12:30:00 PM 2 5 2 0 0 9
32 12:31:00 PM 4 7 0 0 0 11
33 12:32:00 PM 3 4 2 0 0 9
34 12:33:00 PM 7 3 0 0 0 10
 35 12:34:00 PM 5 2 1 0 0 8
36 12:35:00 PM 6 0 3 0 0 9
37 12:36:00 PM 8 2 0 0 0 10
38 12:37:00 PM 7 2 2 0 0 11
39 12:38:00 PM 3 4 2 0 0 9
40 12:39:00 PM 5 4 0 1 0 10
41 12:40:00 PM 0 1 4 0 0 5
42 12:41:00 PM 5 2 3 0 0 10
43 12:42:00 PM 6 3 4 0 0 13
44 12:43:00 PM 2 4 0 0 0 6
45 12:44:00 PM 0 3 2 0 0 5
46 12:45:00 PM 4 2 3 0 0 9
47 12:46:00 PM 6 3 0 0 0 9
48 12:47:00 PM 3 4 2 1 0 10
49 12:48:00 PM 7 2 0 0 0 9
50 12:49:00 PM 5 4 1 0 0 10
51 12:50:00 PM 2 3 4 0 0 9
52 12:51:00 PM 0 2 3 0 0 5
53 12:52:00 PM 6 3 3 0 0 12
54 12:53:00 PM 5 4 0 0 0 9
55 12:54:00 PM 3 1 4 2 0 10
56 12:55:00 PM 4 2 2 0 0 8
57 12:56:00 PM 5 3 0 0 0 8
58 12:57:00 PM 2 3 0 0 0 5
59 12:58:00 PM 4 2 4 0 0 10
60 12:59:00 PM 5 3 0 0 0 8
Total 215 209 101 10 0 535
AVG 8.916667
 Poisson distribution:

No of Observed Total Estimated Theoretical

P(x) Chi Square Test
Arrivals frequency Vehicles Probability Frequency
0 1 0 0.00013413 0.000134127 0.008047597 122.2687393
1 2 2 0.00119596 0.001195962 0.071757742 51.8148718
2 1 2 0.005332 0.005331999 0.319919933 1.44570203
3 4 12 0.01584789 0.015847886 0.950873135 9.777513207
4 1 4 0.03532758 0.035327578 2.119654697 0.591429653
5 7 35 0.06300085 0.063000848 3.780050877 2.742839368
6 2 12 0.09362626 0.09362626 5.617575608 2.32962655
7 2 14 0.11926202 0.119262022 7.155721311 3.714714573
8 6 48 0.13292746 0.132927462 7.975647711 0.489387699
9 10 90 0.13169665 0.131696652 7.901799121 0.557144881
10 10 100 0.11742951 0.117429515 7.045770883 1.238682015
11 3 33 0.09518908 0.095189076 5.711344579 1.287155647
12 1 12 0.07073077 0.070730772 4.243846319 2.479481619
13 2 26 0.04851406 0.048514055 2.910843309 0.285015525
14 2 28 0.03089883 0.030898833 1.853929965 0.011508771
15 1 15 0.01836764 0.018367639 1.102058368 0.009451324
16 1 16 0.01023613 0.010236132 0.614167945 0.24238708
17 0 0 0.00536895 0.005368952 0.322137108 0.322137108
18 1 18 0.00265962 0.00265962 0.159577179 4.426137403
19 0 0 0.00124815 0.001248155 0.07488929 0.07488929
20 0 0 0.00055647 0.000556469 0.033388142 0.033388142
21 2 42 0.00023628 0.000236279 0.014176711 278.1670684
22 0 0 9.5764E-05 9.57644E-05 0.005745864 0.005745864
23 0 0 3.7126E-05 3.71261E-05 0.002227563 0.002227563
24 0 0 1.3793E-05 1.37934E-05 0.000827602 0.000827602
25 0 0 4.9196E-06 4.91963E-06 0.000295178 0.000295178
26 1 26 1.6872E-06 1.68718E-06 0.000101231 9876.408367
TOTAL 60 535 0.999939583 59.99637497 10360.72673
 Graph : Probability values of vehicle arrivals computed using Poisson distribution

Graph : Cumulative probability values of vehicle arrivals computed using Poisson distribution

Max. Value 26 Mean 8.916666667

Min Value 0 Median 9
Std.
Range 26 Deviation 4.917253149
Bin Size 4.08
 Normal Distribution:

Cumulative Chi-
Cumulative Normal Estimated
ARRIVAL Frequency Estimated Square
Frequency Frequency Frequency
Frequency Test
0 4 9 9 0.1586841 9.5210469 9.5210469 0.02851
4 8 17 26 0.4260583 25.563499 16.042452 0.05715
8 12 24 50 0.7346848 44.081085 18.517586 1.62315
12 16 6 56 0.9251379 55.508272 11.427187 2.57757
16 20 1 57 0.9879009 59.274055 3.765783 2.03133
20 24 2 59 0.9989205 59.935228 0.6611724 2.71103
24 28 1 60 0.999948 59.996878 0.06165 14.2822
Total 60 59.996878 23.311

RECOMMENDATIONS AND CONCLUSION

1. Chi-square test fails for Poisson distribution and hence normal distribution is performed.

2. Null hypothesis is expected as the observed value is less than critical value from the 2 table.

Reference:

1. D.May, Adolf Traffic Flow Fundamentals.

2. The Chi Square Statistics http://math.hws.edu/javamath/ryan/ChiSquare.html

3. Normal Distribution Statistics https://www.mathsisfun.com/data/standard-normal-

distribution.html