CLE 406 TRAFFIC ENGINEERING

Measurement of density using occupancy from loop detector data

The speed of an individual vehicle is a function of distance traveled divided by the travel time.
That is,

Where, = Speed of individual vehicle (ft/sec)

=Length of individual vehicle (ft)
=Detection zone length (ft)
=Individual vehicle occupancy time (sec)
The relationship can be converted to a macroscopic level by using average values of vehicle
lengths and vehicle occupancy times. i.e.,

The total time the detector is occupied

detector is,

in time period T, when N vehicle passes over the

Where, = Total occupancy time in time period T (hrs)

=Number of vehicles passed
=Average occupancy time
Now equation (1) can be rearranged as

Substituting equation 3 in 2,

The % occupancy in time T is,

Substituting equation 5 in 6,

Now

P1). A loop detector having a length of 12 ft was observed to have 6 vehicles cross over it in a
period of 148 sec for the following durations: 0.44, 0.48, 0.50, 0.41, 0.49, and 0.55. Estimate q, u
and k. The corresponding lengths of vehicles were 18, 21, 20, 23, 17 and 19 ft.
Solution:

Moving observer method:

Estimating the speed and flow of traffic from a moving vehicle is a useful field method. It
consists of making a series of runs in a test vehicle, recording the number of vehicles that

overtake the test vehicle, the number of vehicles passed by the test vehicle and the travel time for
the test vehicle.
Lets assume that a test vehicle begins its run against traffic at time zero. At the same time, an
independent observer located at point A (point where test vehicle begins its run) begins to count
all the vehicles passing point A till the test vehicle passes. The test vehicle reached the
considered length L at time Ta and again continuing takes Tw time to reach point A. So, total time
is Ta+ Tw.
Total number of vehicles the observer counted during the time (Ta+ Tw) = Ma+ Mw
Where, Ma is the number of vehicles the test vehicle encounters during its run against traffic and
Mw is the number of vehicles that overtakes the test vehicle minus number of vehicles overtaken
by the test vehicle while its journey with traffic. So,

Lets consider 2 cases: First a stationary observer and a moving stream.

Where,

= vehicles that overtake the observer during the observation time T.

Second case is observer is moving and stream is stationary. By traveling a distance L, the
observer would overtake a number of vehicles . So,

Where,

is the observers speed and

is the time it takes the observer to traverse distance L.

Now consider the observer actually moving with the stream. Some vehicles
some vehicles
will be overtaken by the observer.

Dividing by T,

will overtake and

Where, u is the space mean speed.

P1). The data shown were noted in a travel time study on a 2 mile stretch of a highway using the
moving observer method. Determine the travel time and flow in each direction on this section of
the highway.

Solution:
Find the average of 4 runs, i.e, average of travel time, no. of vehicles traveling in opposite
direction, no. of vehicles overtaking test vehicle, no. of vehicles overtaken by test vehicle for
both NB and SB runs.

Average Travel time is

P2). The following table gives the particulars collected for a section of road 0.7 km long during
the course of a moving observer method. Calculate the flow in PCU per hour in both directions
of traffic assuming an equivalency factor of 1 for car, 3 for bus, 2 for trucks. Calculate the
journey speed and running speed.

Solution:
Northbound direction:
Average journey time in NB = 0.771 min.
Average stopped time = 0.095 min.
Average no. of vehicles traveling in opposite direction = 78+(5*3)+(38*2)=169, i.e,
169/6=28.166
Average no. of vehicles overtaking test vehicle = 1.166
Average no. of vehicles overtaken by test vehicle = 0.666

Southbound direction:
Average journey time in SB = 1.055 min.
Average stopped time = 0.09 min.
Average no. of vehicles traveling in opposite direction = 61+(1*3)+(21*2)=106, i.e,
106/6=17.666
Average no. of vehicles overtaking test vehicle = 1.5
Average no. of vehicles overtaken by test vehicle = 0.5

Poisson distribution of vehicle arrivals:

Simeon Denis Poisson was a French mathematician (1781-1840) who developed the Poisson
model, which is having wide application in many areas of civil engg. In traffic engg., using the
Poisson model, we can find out how many vehicles that are likely to arrive during a time interval
of a specified length. The two important assumptions of Poisson process are,
1). The events are random, i.e, the number of events that occur in one time interval is
independent of the number of events that occur in any other time interval.
2). The probability that a single event will occur during a short time interval is proportional to
the length of the time interval. i.e, as the time interval used for the analysis is decreased, the
probability of an event occurring also decreases.

Where,

= Probability of having n vehicles arrive in time t.

is the average arrival rate in vehicles per unit time and t is the duration of time interval.
P1). An observer counts 360 veh/hr at a specific highway location. Assuming that the arrival of
vehicles at this location is Poisson distributed, estimate the probabilities of having 0,1,2,3,4 and
5 or more vehicles arriving over a 20 sec time interval.
Solution:

Similarly, P(2) = 0.271; P(3) = 0.180; P(4) = 0.090;

P2). Traffic data are collected in 60 sec. intervals at a specific highway location as shown below
in Table. Assuming that traffic arrivals are Poisson distributed, what is the probability that six or

more vehicles will arrive in each of the next three 60 sec. time intervals? (i.e., 12.15 to 12.16 pm,
12.16 to 12.17 pm, 12.17 to 12.18 pm)

Solution:
101 vehicles arrived in 15 min. period. So,

and t = 60 sec.

P(2) = 0.027; P(3) = 0.0606; P(4) = 0.102; P(5) = 0.137;

P3). In order to check the validity of the Poissonian arrival, a count of vehicles passing an
observer was taken. The number of vehicles arriving intervals of 20 sec. each was counted and
shown below in table. Check whether the given arrivals are poisson distributed or not?

Solution:
Null hypothesis: The vehicle arrivals follow Poisson distribution
Alternative hypothesis: The vehicle arrivals do not follow Poisson distribution
Average arrival per 20 sec. interval = 687/139=4.942
Where, 687 is the total number of vehicles counted during the survey and 139 is the total number
of 20 sec. intervals.

Multiplying the probability values by 139 (the total number of 20 sec. intervals), the expected
number of intervals with stated number of arrivals can be determined as shown below.
Number of vehicles Observed number of intervals with Expected number of intervals with
stated number of vehicles (O)
stated number of vehicles (E)
0
Nil
1
1
4
5
2
13
12
3
23
20
4
26
25
5
24
24
6
14
20
7
20
14
8
6
9
9
2
5
10
3
2
11
3
1
12
1
0

Note that, any cell value in expected frequency should be greater than or equal to 5 before
applying Chi-square test.

Degrees of freedom = No. of categories - 1

i.e., Degrees of freedom = 9 1 = 8

From chi-square test table,

Since,
Poisson distribution.

accept null hypothesis. So, the vehicle arrivals follow

Application of Poisson distribution in design of left turn (LT) lane:

P1). A LT lane at a place has a storage capacity of 7 cars. The average vehicle arrival rate at the
intersection is 60 cars/min with an average of 10% of the cars wanting to turn left. The cycle
time is 60 sec. If we assume Poisson arrivals, what is the probability that there will be more than
7 cars for which the lane is designed?
Solution:
Average arrival rate to turn left = 0.10 * 60 = 6 cars/min.

Application of Queueing theory in traffic engg:

Formation of traffic queues during congested time periods results in time delay to individual
vehicles and overall performance of the highway gets reduced. Before suggesting any solutions
to reduce queue formation, it is first essential to develop a clear understanding of the operational
characteristics of the queue. That is, time at which the queue dissipates, longest delay to an

individual vehicle, longest queue length, etc. The three important components of a queueing
system are,
1). Assumptions about arrival characteristics
2). Assumptions about service characteristics
3). Number of service channels available
The arrival can be either deterministic (uniform arrival, i.e., equal time headway between
vehicles) or random/Poisson arrival. Similarly the service time also can follow deterministic or
random/Poisson pattern. If we assume, both arrival and service as deterministic with one service
channel, then the model is denoted as D/D/1. If arrival is deterministic and service time is a
Poisson process, then the model is denoted as D/M/1. If arrival is random and service time is
deterministic, then the model is denoted as M/D/1. If both arrival and service time follow
Poisson distribution, then the model is denoted as M/M/1.
P1). Vehicles arrive to a recreational park. There is a single gate at which all vehicles must stop
and purchase tickets. The park opens at 8 am at which vehicles begin to arrive at a rate of 480
veh/hr. After 20 min, the arrival flow declines to 120 veh/hr and continues at that level for the
remainder of the day. If the time spend to purchase the ticket coupons is 15 sec./veh, describe the
operational characteristics of the queue.
Solution:
Given, arrival rate

Service time,

Formulate mathematical function for total number of vehicle arrivals and similarly total number
of vehicle departures.
Total number of vehicle arrivals =
Total number of vehicle arrivals =
Total number of vehicle departures =
Now, calculate the total number of vehicle arrivals and similarly total number of vehicle
departures at various times using the above equations as shown below.

Now, plot a graph using the above data points thus showing the arrival and departure pattern as
shown below.

From the graph, it can be clearly seen that, the time at which the queue dissipates = 60 min., the
point at which both arrival and departure curves meets. The longest delay to an individual
vehicle is where the horizontal distance between the arrival and departure curves is maximum. In
the given problem, it is 20 min (i.e., 160th vehicle arrived at 8.20 am and departed only at 8.40
am and experienced the max. delay of 20 min (i.e, 40-20 in x-axis)). Similarly longest queue
length is where the vertical distance between the arrival and departure curves is maximum. In
this case, it is 80 vehicles (i.e., at t=20 min, 160 vehicles arrived and only 80 vehicles departed.
So, 160-80=80 vehicles).

Highway capacity and Level of service

Definition of capacity:
Maximum hourly rate at which persons or vehicles can reasonably be expected to traverse a
point or uniform section of a lane or roadway during a given time period under prevailing
roadway, traffic and control conditions
Definition of level of service (LOS):
LOS represents a qualitative ranking of the traffic conditions experienced by the users of a
facility under prevailing roadway, traffic and control conditions. There are 6 types of LOS
categories, namely LOS A to F as shown below.
LOS A: Free flow conditions (Individual vehicles are unaffected by the presence of other
vehicles in the traffic stream).
LOS B: Speeds at or near free flow speeds. But presence of other vehicles in traffic stream
begins noticeable;
LOS C: Freedom to maneuver is noticeably restricted
LOS D: Speeds begin to decline with increasing flow.
LOS E: Operating conditions are at or near the roadways capacity. Driver will experience
physical and psychological discomfort.
LOS F: Breakdown in vehicle flow; Queue forms as arrival rate is greater than departure rate.
By determining capacity and LOS, improvements and changes in geometric features, junction
improvements, traffic management measures and improvements in traffic control devices can be
planned effectively.
Highway Capacity Manual (HCM) has defined two categories of highway facilities.
1) Uninterrupted flow facilities
a. Freeways
b. Multilane highways
c. Two-lane highways
2) Interrupted flow facilities
a. Urban streets with signalized or unsignalized intersections

HCM (2010) has defined freeway, multilane highways and two-lane highways as given below.

Freeways - Divided highway facility having 2 or more lanes with full access controlled.

Multilane highways 4 or 6 lanes with physical medians in between but sometimes it

may be undivided also. Side friction with vehicles entering and leaving the road and the
existence of opposite vehicles are also prevalent.

Two-lane highways - Undivided highway with only 2 lanes. Lane changing and passing
is possible only in the face of the oncoming traffic in the opposing lane.

Capacity and LOS of freeways:

In order to find LOS of freeways and to check whether the road reaches its capacity or not, the
following steps are usually carried out according to HCM 2010.
Step 1). Determine FFS for current traffic conditions
FFS for current traffic conditions can be determined using the equation given below.

The base or ideal conditions of a freeway are given below.

12 ft minimum lane widths
6 ft min. right shoulder clearance between the edge of the travel lane and objects (utility
poles, walls, etc.) that influence driver behavior
2 ft min. median lateral clearance
Only passenger cars in the stream
5 or more lanes in each direction
2 mi. or greater interchange spacing
Level terrain (no grades greater than 2%)
Driver population of mostly familiar users
The following tables can be used to find the adjustment factors.

Step 2). Determine the 15 min. passenger car equivalent (PCE) flow rate

Where,

ET and ER can be determined from the following table.

Step-3: Determine LOS using Figure 6.2

Draw an imaginary line corresponding to calculated FFS (from step-1) in Fig.6.2. Draw a
vertical line from horizontal axis for the calculated vp value (from step-2). The point where this
vertical line intersects the FFS line is LOS of freeway. Or calculate density using the following
equation. Then either using Fig 6.2 or Table 6.1, the LOS can be calculated.

Where, D is the density in pc/mi/lane; S is the average traffic stream speed (y-axis in Fig. 6.2)

Fig. 6.2 LOS criteria for basic freeway (Source: HCM)

P1). A six lane urban freeway (3 lanes in each direction) is on rolling terrain with 11 ft lanes,
obstructions 2 ft from the right edge of the traveled pavement, 1.5 interchanges per mile. The
traffic stream consists of regular commuters. A directional weekday peak hour volume of 2200
vehicles is observed with 700 vehicles arriving in the most congested 15 min. period. If the
traffic stream has 15% large trucks and buses with no recreational vehicles, determine the LOS.
Solution:

FFS = 70 - 1.9 - 1.6 - 3 5 = 58.5 mi/hr

PHF = 2200/(700*4) = 0.786

Now draw an imaginary line corresponding to calculated FFS of 58.5 mi/hr in Fig.6.2. Draw a
vertical line from horizontal axis for the calculated vp value of 1144 pc/hr/lane. The point where
this vertical line intersects the FFS line is LOS of freeway. In this case, LOS is C. Or calculate
density using the following equation.

Since, 18 < 19.6 < 26, the LOS is C (either using Fig 6.2 or Table 6.1).

P2). For problem-1, how many vehicles can be added before the roadway reaches its capacity
assuming that proportion of vehicle types and PHF remains constant?
Solution:

We need to find out the hourly volume (V) when vp reaches its capacity flow. From table 6.1,
find the max. service flow rate for FFS of 58.5 mph.
For 55 mph, vp = 2250 (at LOS E, i.e, at capacity);
For 60 mph, vp = 2300;
So, for 58.5, vp = 2285;
So,
Thus, 4397-2200 = 2197 veh/hr can be added before the roadway reaches its capacity.