Transportation Planning and Engineering

Lecture 13

Transportation Planning Process

(Data Collection Methods - Cont’d)
Sample Sizes
2
 cvt  (cvt / p) 2
n    n
 p  1  [(cvt / p) 2 / N ]
Infinite population Finite population
n = sample size
cv = coefficient of variance
p = accounts for assumed error
t = t-statistics value
N = population size

Value of ‘t’ in above equations depends on nature of t-test (one tail test of
two tail test). Or whether the null hypothesis is defined to identify difference
in one direction or two. (a null hypothesis to say that ‘x’ is not less than y
would require a one-tailed test, while a null hypothesis to say ‘x’ is not equal
to y would require a two tailed test.)
2
Examples
1. Suppose that it is desired to estimate the average daily
flow of traffic pass a site and to be 95% confident that
the answer is within 10% of the correct value. It is
suspected that the flow varies quite markedly from day
to day. For how many days will it be necessary to
collect flow data?
2. Suppose that it is desired to estimate the average
household trip rate in a large city and to have 95%
confidence that answer obtained is within 10% of the
true mean. How many households need to be
surveyed? If instead of a large city the survey is to be
carried out in a small village with only 100 households
then how many households need to be surveyed?
3
Secondary Sources
 Use of existing data ‘secondary sources’ can
save time and money – Inventory
 Desirable when a survey can cause disruption to
the travelling public
 But, under some critical circumstance it is
necessary to conduct a survey.
 Sources of secondary data:
1. Published database
2. Previous local area surveys and
3. Data produced as by product of control or
management system.
4
Secondary Sources
 Published tabulations
 Collected by government agencies as part of regular monitoring
 Useful for establishing trends on issues such as car ownership,
accidents, seasonal tends in traffic flow.
 Need to focus on variable definitions and sample sizes.
 Detailed local data
 May be part of local monitoring program
 Need to focus on variable definitions and sample sizes.
 Proper documentation of data
 Use of GIS should reduce the documentation problem.
 Data produced as by product of control or management
system.
 Traffic flow and congestion captured during the operation of fully
automated urban traffic flow systems, usage of car parks with computer
controlled entry an exist barriers, flow of vehicles at toll points.

5
 Inputs
Urban Transportation •Transportation system characteristics
Planning or Model System •Land use-activity system characteristics

 Set of models that is commonly UTMS

used to predict the flows on the links
of a particular transportation Trip generation (how many trips?)
network, as a function of the land-
use-activity system that generates Trip distribution (where do they go?)
travel, is generally known as the
UTMS. Mode Choice (By what mode?)
 Often referred to as the four-step,
sequential model because it Trip assignment (By what route?)
comprises four sub-models that
employed in a sequential process.
 Trip generation Outputs
 Trip distribution Traffic flows on network links
 Mode choice (modal split) •Quantity (volume)
•Quality (Speed)
 Trip assignment
6
Urban Transportation Planning or Model System
Trip Generation Aj
Pi
 It is concerned with predicating the number of trips
produced by and attracted to each traffic analysis zone.
 Trip generation models address the question of how i j
many trips are made to and from each zone of the study Trip generation
area.
Tip Distribution
Trip Distribution
 Concerned with predicting where trips the go. Trip
distribution models links the origin and destination of the i Tij
j
trips generated by the trip generation models.
Modal Split Tij, car
 Address the question how the various trips are made? Modal split
These models predicts the proportion of trips by each i j
mode of travel between each origin and destination. Tij,
Trip Assignment transit
i
 Concerned with predicting the route(s) used by the trips Route taken
by car rider
from a given origin to a given destination by a particular
mode. Transit route
j
Trip assignment
7
Approaches of Transport Modeling
 Unit of analysis
 Aggregate model: trip behaviors are aggregated by spatial zones and
such macro behaviors are modeled using aggregate variables
 Disaggregate models: individual trip behaviors (micro behaviors) are
modeled using individual variables
 Description of Phenomenon
 Stochastic model: trip behaviors are estimated by probabilistic
approach (%)
 Deterministic model: trip behaviors are described/ estimated by
deterministic approach. (by 1 or 0)
 Description of behavior decision mechanism
 Simultaneous model: behaviors are decided by considering different
factors at the same time. (travel time, cost and comfort)
 Sequential model: behaviors are decided by considering different
factors sequentially.

8
Trip Rate and Trip Production
 Future total number of trips is estimated by trip
rate ‘R’.
 Present ‘R’ and Future ‘R’ are assumed to be
same
Number of trip present
R present 
Population present
Number of trip future
R future 
Population future
Number of trip future  R future( R present) xPopulatio n future

9
1. Trip generation
 Number of trips produced by (= trip generation) or
attracted to (= trip attraction) a given piece of land or
each zone depends.
 Land use type
 Intensity and socioeconomics characteristics

Zone i Zone j

Total trip generation Total trip attraction

10
1. Trip generation – cont’d
 Models of estimating future trip attraction / generation
 Disaggregate type models
 Analytical unit: person, household
 Whether a person (household) have a trip or not (0 or 1) – DM
 Probability of trip generation (0 to 1) – PM or SM
 Trip frequency rate of a person by each trip purpose
 Characteristics of unit
 Household income, car ownership, trip purpose, occupation, etc.
 Aggregate type model
 Analytical unit: zone
 Number of trip generation / attraction
 Characteristics of unit
 Number of residential population
 Number of working population
 Zonal floor space by business type
 Distance of zone from city center
 Accessibility to the work place
11
 1. Trip generation – cont’d
 Models of estimating future trip attraction / generation
 Estimation by personal and zonal attributes
 Constrained by the given future total tip production of the total area.

Generations Attractions (Destinations - j) Trip

(Origins-i) 1 2 3 4 …………………………………n Generation
T ij

1 T11 T12 T13 T14 ……………………... T1n O1

i

2 T21 T22 T23 T24 …………………...... T2n O2

3 T31 T32 T33 T34 ……………………... T3n O3
4 T41 T42 T43 T44 …………………...... T4n O4
. . . . . …………………….. . .
. . . . . …………………….. . .
. . . . …………………….. .
. . . . . …………………….. . .
. . . . . …………………….. . .
n Tn1 Tn2 Tn3 Tn4 …………………….. Tnn On
Trip Attraction T
j
ij
D1 D2 D3 D4 ………………….. Dn ∑Tij=12T

Control total =Trip production

1. Trip generation – cont’d
1. Growth Factor Model
 Relations across time
 The growth factor Fi depends on the explanatory variable
such as population (P) of the zone , average house hold
income (I) , average vehicle ownership (V).

Ti  Fi ti
Ti  future no. of trips
ti  present no. of trips
Fi  growth factor
where the subscript " d" denotes the
design year and the subscript "c" denotes
F is difficult to estimate the current year.

13
1. Trip generation – cont’d
Example
 Given that a zone has 275 household with car and 275 household without
car and the average trip generation rates for each groups is respectively 5.0
and 2.5 trips per day. Assuming that in the future, all household will have a
car, find the growth factor and future trips from that zone, assuming that the
population and income remains constant.

 The above example also shows the limitation of growth factor method. If we
think intuitively, the trip rate will remain same in the future. Therefore the
number of trips in the future will be 550households 5 trips per day = 2750
trips per day .

14
1. Trip generation – cont’d
2. Linear Multiple Regression Model

Yi = a0 + a1X1i + a2X2i + ……………. + akXki + Ei

i= zone, K= no. of explanatory variables

Yi = no. of trips generated from / attracted in zone I

Xk = personal or zonal attributes

 Obtaining zonal total

 In the case of household model, total number is derived as
sum of total number of household trips
 Control Total
 Total number of generated trips from all zones must be equal
to the total number of attracted trips to each zone.
 The difference must be assigned to each zone.

15
1. Trip generation – cont’d
Model verification
important indices of regression model
 Goodness of fit in each variables
 T-test, checked by t-value (or the P-value)
 More than 95% significant
 Consistency of the sign conditions (±)

 Determination of the model

 F-test, checked by F-value, (or the P-value)
 More than 95% significant
 The coefficient of determination: R2
 0 to 1.0 (app. More than 0.7)

16
QUESTIONS !!!!