SUSTAINABLE TRANSPORTATION SYSTEMS
Lecture 30: Sustainable Transport Planning and Approaches-I
The Traditional Transport Planning Process
Prof. Bhola Ram Gurjar
Dept. of Civil Engineering
1
Contents
• Goals of Transport Planning
• The Traditional 4-Step Planning Model
– Trip Generation
– Trip Distribution
– Modal split (Mode choice)
– Trip Assignment
• Numerical Problems
• Conclusion
Goals of Transport Planning
These are
the Goals
Source: (PlanRva, 2019 )
3
Planning future Transportation
Source: (Piter Biswas, 2016, https://www.slideshare.net/PiterBiswas/urban-transportation-planning-66010727 )
4
Systems Approach to Transport Planning
Decision to adopt planning
Problem, definition,
formulation of goals
Problems, constraints, potentials, forecasting
Solution generation
Performance assessment Solution analysis
and review
Evaluation of possible
alternatives and choice
Operation
Implementation
Source: (L. R. Kadiyali 2017)
5
The Traditional 4-Step Planning Model
Step 1. Trip Generation
Forecasts the travel demands for a specified area- number of trips, purpose etc. in a study area
Step 2. Trip Distribution
Forecasts the number of trips between Origin and Destinations in each zones of study area
Step 3. Modal Split (Mode choice)
Predicts the mode choice preference of individuals
Step 4. Trip Assignment
Assigns traffic to a predicted model
Source: (L. R. Kadiyali 2017)
6
Key Steps in the Traditional 4-Step Planning Model
Trip Generation
Trip Assignment
Trip Distribution
Modal split
7
What we forecast in a Planning process?
• How many total trips ?
• How long is an average trip between Origin
and destinations?
• What is the Modal split (Mode choice)?
• How many cars per hour per lane?
• How to provide Public transport (number of
trips, frequency etc.)?
8
Stages in Transport Planning
The Traditional Transport
forecasting process are
completed by using
these 5 stages of
planning.
Source: (L. R. Kadiyali 2017)
9
Stages in Transport Planning
Stage 1: Survey and analysis of existing conditions Definition of Survey
area and formulation
Inventory of existing Inventory of existing Land use and of Goals
travel pattern transport facilities economic activities
• Origin and • Street inventory • Zoning
destination surveys • Travel time studies • Land use
•
• Household
interviews •
Bus inventory
Rail inventory
• Population
structure
Stage 1
• Rail and bus surveys • Parking inventory • Household
• Goods vehicle structure
interviews Basic Analysis • Employment
• Coach surveys • Car ownership pattern
• Cordon and screen • Trip generation • Income
line studies • Trip attraction
• Modal selection
• Trip distribution
• Goods vehicles trips
Source: (L. R. Kadiyali 2017) Stage 2
10
Stages in Transport Planning
Stage 2: Forecast, analysis of future conditions and plan synthesis Stage 1:
Planning forecasts
Basic analysis
Network planning Future Trip
• Development of Generation • Projection of
tentative network of population and
transport facilities Trip households
Distribution • Projection of
employment
Stage 2
Modal • Future land use
Assignment
Revise split
network Formulation of
feasible Revise
alternatives planning
forecasts
Stage 3: Evaluation Evaluation
• Preliminary cost estimates
Stage 3 •
•
Cost/Benefit analysis
Other considerations
Source: (L. R. Kadiyali 2017) Stage 4
11
Stages in Transport Planning
Stage 4: Programme adoption and implementation
Stage 3: Evaluation
Programme adoption and Implementation
• Selection of network plans
• Development of Transport policies Stage 4
• Staging
• Organisation
• Implementation
Stage 5: Continuing studies
Continuing study of community
Stage 5 development and updating of plans
Source: (L. R. Kadiyali 2017)
Stage 1
12
Transportation Planning: Factors to Consider
Factors to
consider
Source: (PlanRva, 2019)
13
Transportation Surveys
• First stage in the formulation of a Transportation plan.
• Aim is to collect data on all factors influencing travel patterns.
• Voluminous and time consuming work (Can take upto 2 years)
• Cost involved is very high.
Source: (L. R. Kadiyali 2017)
14
Transportation Surveys (cont’d..) External cordon
Screen line
• External cordon: Imaginary line
representing boundary of study
area. Internal to Internal
• Screen line: located along
physical or natural barriers Internal to External to Internal
• Ex. Rivers, railway lines etc. external
• 4 Basic movements in survey of a Basic movements in a transportation survey
study area
(i) Internal to internal
(ii) Internal to external
(iii) External to internal
(iv) External to external
Source: (L. R. Kadiyali 2017)
15
Transportation Surveys: Trade-Offs
Source: (Anthony J. Richardson et. al, 1995)
16
Types of Transportation Surveys
Home Taxi surveys
interview Road-side
surveys interview surveys
Commercial
vehicle
surveys Public transport surveys
Post card
questionnaire
surveys Registration number & Tag surveys
Source: (L. R. Kadiyali 2017)
17
Trip Generation
• Aim is to calculate the number of trips in
a given area Home based trips
Trip Purpose constitutes nearly 80-
90% of the total trips
in a given area.
Home Work
Home-based trips
Work Shop
Source: (Traffic Engineering and Transport Planning, L. R. Kadiyali)
Non-home based trips
18
Factors governing Trip Generation and Attraction
1. Income
2. Car ownership
3. Family size and composition
4. Land use characteristics
5. Distance of the zone from the town
centre
6. Accessibility to public transport system
and its efficiency
7. Employment opportunities
Source: (L. R. Kadiyali, 2017)
19
Trip Generation: Calculation
Multiple Linear Regression Analysis
Source: (Riza Atiq bin O.K. Rahmat , 2012)
20
Trip Generation: Calculation of Total Number of Trips
Multiple Linear Regression Analysis
Dependent variable: Number of trips
Independent variables: Various independent factors that influence trip generation
Yp= 𝑎1X1 + 𝑎2X2 + 𝑎3X3,………,𝑎𝑛Xn + U
Where, Yp = number of trips for specified purpose p
X1,x2,x3,…,Xn = independent variables
a1,a2,a3…,an = coefficients of the respective independent variables
U = disturbance term, which is a constant, and
representing that portion of Yp not explained by the
independent variables
Source: (L. R. Kadiyali, 2017)
21
To calculate the Total Number of Trips in an area
Numerical example 1
Dependent variable: Y = Average trips per dwelling unit
Independent variables: A = Car ownership, H = Household size, X1 = Social rank index and
X2 = Urbanization index with coefficients 3.404, 0.516, 0.0119 and -0.343 respectively
Disturbance, U = 2.18
Yp= 𝑎1X1 + 𝑎2X2 + 𝑎3X3,………,𝑎𝑛Xn + U
The Average number of trips per dwelling unit can be calculated
by,
Y = 3.404 A + 0.516 H + 0.0119 X1 – 0.343 X2 + 2.18
Source: (L. R. Kadiyali, 2017)
22
Trip Rate
• Defined as the ratio of the total number of trips generated
per day to the total population of considered area
𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑡𝑟𝑖𝑝𝑠 𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑒𝑑 𝑝𝑒𝑟 𝑑𝑎𝑦
𝑇𝑟𝑖𝑝 𝑟𝑎𝑡𝑒 =
𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑎𝑟𝑒𝑎 𝑐𝑜𝑛𝑠𝑖𝑑𝑒𝑟𝑒𝑑
Source: (L. R. Kadiyali 2017)
23
Calculation of Trip Rate
Numerical problem 1
Estimate trip rate for a residential colony with a population of
2744 and 6574 number of trips ?
Solution
𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑡𝑟𝑖𝑝𝑠 𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑒𝑑 𝑝𝑒𝑟 𝑑𝑎𝑦
𝑇𝑟𝑖𝑝 𝑟𝑎𝑡𝑒 =
𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑎𝑟𝑒𝑎 𝑐𝑜𝑛𝑠𝑖𝑑𝑒𝑟𝑒𝑑
6574
𝑇𝑟𝑖𝑝 𝑟𝑎𝑡𝑒 = = 2.39 ≈ 2.4
2744
Source: (L. R. Kadiyali 2017)
24
Example of Trip Generation Calculation for Model
development using Multiple Linear Regression Analysis
Step 2.
• Compute each of the parameters of the potential Regression
equations.
Step 3.
Check the following criteria for Statistical significance.
• R2 value, Intercept value and Regression parameters
Source: (Riza Atiq bin O.K. Rahmat , 2012)
25
Example of Trip Generation Calculation for Model
development using Multiple Linear Regression Analysis
To observe any
relationship between
parameters
R2 = 1 means
Max. correlation
between X and Y
R2 = 0 means No
correlation
Source: (Riza Atiq bin O.K. Rahmat , 2012)
26
Trip Distribution
Main Question addressed is
“Where will the generated trips go
to?”
• Internal or External trips in a city
• Internal trips with Origin and Destination
confined to city limits
• External trips whose Destination is beyond
city or town limits
Source: (Riza Atiq bin O.K. Rahmat , 2012)
27
Main Purpose of Trip Distribution
Trip Production Trip Attraction
560 1250 Trips generated (origin)
750 530 and attracted
1105 430
(destination) should be
545 540
balanced and made equal
450 1200
1040 500
4450 4450
Source: (Riza Atiq bin O.K. Rahmat , 2012)
28
Balancing the O-D Matrix in Trip Distribution
Trip Production Trip Attraction
560 1250
1 2 3 4 5 6
750 530
1 157 67 54 68 151 63 560
1105 430
Trip Production
2 211 89 72 91 202 84 750
545 540
3 310 132 107 134 298 124 1105 450 1200
4 153 65 53 66 147 61 545 1040 500
5 126 54 43 55 121 51 450 4450 4450
6 292 124 100 126 280 117 1050
1250 530 430 540 1200 500 4450
1050 Trip Attraction
𝐸𝑥𝑎𝑚𝑝𝑙𝑒: 1250 × = 292
4450
Source: (Riza Atiq bin O.K. Rahmat , 2012)
29
Methods of Trip Distribution
Methods of Trip There are
Distribution
various methods
to calculate the
Growth Factor
Models
Synthetic Models
trip distribution
for a study area
Uniform factor Average factor
The Gravity model Tanner model
method method
Intervening Competing
Fratar method Furness method
opportunities model opportunities model
Source: (L. R. Kadiyali, 2017)
30
Growth Factor Models
• Assumes that we have a basic trip matrix data
from previous studies or recent surveys.
• Main Goal here is to estimate future matrix.
• Used in earlier studies widely, but now been used
only for smaller studies. Origin-Destination matrix
• Nowadays, Synthetic models are widely used.
Source: (L. R. Kadiyali, 2017)
31
Example of a Uniform Growth factor model
Given a trip matrix for base year as in Table 1. If the traffic growth factor for the
forecast year is 1.2 for the study area, estimate the future Trip matrix?
Total Future trips
increased from 4450
to 5340 trips
Growth
𝐹𝑢𝑡𝑢𝑟𝑒 𝑡𝑟𝑖𝑝 = 𝐵𝑎𝑠𝑒 𝑦𝑒𝑎𝑟 𝑡𝑟𝑖𝑝 ×Growth factor factor from
past data
Source: (L. R. Kadiyali, 2017)
𝐸𝑥𝑎𝑚𝑝𝑙𝑒: 157 × 1.2 =188.4
32
Synthetic model (The Gravity model) 𝑚1𝑚2
𝐹=𝐺
• Widely used in Current studies. 𝐷2
• Based on Newton's law of gravitation 𝑤ℎ𝑒𝑟𝑒,
• “The force of attraction between two bodies is 𝑚1 𝑎𝑛𝑑 𝑚2 = masses
directly proportional to the product of masses of bodies
D = distance between
between the two bodies and inversely
the bodies
proportional to the square of the distance.”
• The Gravity model is the well known synthetic model
used widely.
Source: (L. R. Kadiyali, 2017)
33
The Gravity model
𝑚1𝑚2
Newton's law of gravitation 𝐹=𝐺
𝐷2
Total Number of trips between 𝑃𝑖𝐴𝑗
origin and destination 𝑇𝑖𝑗 = 𝐾
𝑓(𝑅𝑖𝑗)
𝑤ℎ𝑒𝑟𝑒,
Tij= Number of trips between origin I and destination j.
𝑃𝑖 = Production of zone I
𝐴𝑗 = Attraction of zone j
𝑓(𝑅𝑖𝑗) = Separation function of the travel cost between zone i
and zone j
Separation function depends on distance and mode
Source: (L. R. Kadiyali, 2017)
34
Modal Split/ Mode Choice
• The process of separating person-trips by
the mode of travel (Choice of people)
Decision structure of
All Trips
Source: (L. R. Kadiyali, 2017; Riza Atiq bin O.K. Rahmat , 2012)
35
Example: Modelling the Choice of people
To Choose between Walking or to Ride a vehicle?
Example
of Plotting
the share
of trips by
Walking
Source: (Riza Atiq bin O.K. Rahmat , 2012)
36
Example of Mode choice of People in the U.S and Western
Europe People of Western
Europe prefer
Public transport,
Cycling or walking
compared to the
Private vehicles
Source: (MIT lecture Notes, 2002)
37
Factors affecting Mode choice
Characteristics of Trip
• Trip Purpose and Trip length
Qualitative
characteristics
• Comfort, convenience
safety etc.
Factors
affecting Household characteristics
• Income, Car ownership,
Mode choice family size & composition
Network characteristics
• Accessibility ratio, Travel Zonal characteristics
cost ratio, Travel time ratio • Residential density,
distance from CBD etc.
Source: (L. R. Kadiyali, 2017)
38
Modal Split: Calculation
• Measures the satisfaction derived from a mode choice
• Expressed as linear weighted sum of independent variables
U = 𝑎0 + 𝑎1X1 + 𝑎2X2 + 𝑎3X3,………,𝑎𝑛Xn
Where, U = utility derived from choice
Xn = attributes
an = model parameters
Source: (L. R. Kadiyali, 2017)
39
Modal Split: Calculation by Binary Logit Model
• Calculates the probability of selecting a particular mode of transport
𝑒 𝑈𝑘
p(K)=
𝑒 𝑈𝑘
Where, p = probability of selecting mode k
The probability of choosing The probability of choosing
public transport (pt) over automobile (auto) over public
automobile (auto) between transport (pt) between zones i
zones i and j , and j ,
𝑒 𝑈𝑝𝑡 𝑒 𝑈𝑎𝑢𝑡𝑜
pij= pij=
𝑒 𝑈𝑝𝑡 +𝑒 𝑈𝑎𝑢𝑡𝑜 𝑒 𝑈𝑝𝑡 +𝑒 𝑈𝑎𝑢𝑡𝑜
Source: (L. R. Kadiyali, 2017)
40
Modal Split: Numerical example 1
Qn: Suppose, the utility function of public transport (pt) and automobile (auto) are
represented by the utility equation:
U = 𝑎𝑘 − 0.35t1−0.08t2−0.005c Where, 𝑎𝑘 = mode specific variable
t1 = travel time in minutes
and the travel characteristics between two zones are given int2= waiting time in minutes
table below: c = cost (cents)
Variable Auto Public transport
𝑎𝑘 -0.46 0.47
t1 20 30
t2 8 6
c 320 100
Calculate the probability of choosing automobile and public
transport w.r.t other mode.
Source: (Reginald Souleyrette, and Adam Kirk, 2013)
41
Modal Split: Numerical example 1
Solution:
Uauto = – 0.46 – 0.35(20) – 0.08(8) – 0.005(320) = -9.70
Upt = – 0.07 – 0.35(30) – 0.08(6) – 0.005(100) = -11.55
𝑒 𝑈𝑎𝑢𝑡𝑜 𝑒 −9.70
pauto= = = 0.86 𝑜𝑟 86%
𝑒 𝑈𝑝𝑡 +𝑒 𝑈𝑎𝑢𝑡𝑜 𝑒 −9.70 +𝑒 −11.55
𝑒 𝑈𝑝𝑡 𝑒 −11.55
ppt= = = 0.14 𝑜𝑟 14%
𝑒 𝑈𝑝𝑡 +𝑒 𝑈𝑎𝑢𝑡𝑜 𝑒 −9.70 +𝑒 −11.55
Source: (Reginald Souleyrette, and Adam Kirk, 2013)
42
Modal Split: Numerical example 2
Qn: Suppose, the utility function of public transport (pt) and automobile (auto) are
represented by the utility equation:
U = 𝑎𝑘 − 0.35t1−0.08t2−0.005c Where, 𝑎𝑘 = mode specific variable
t1 = travel time in minutes
and the travel characteristics between two zones are given int = waiting time in minutes
2
table below: c = cost (cents)
Variable Auto Public transport Bike
𝑎𝑘 -0.46 -0.07 -0.07
t1 20 30 35
t2 8 6 0
c 320 100 0
Calculate the probability of choosing each modes w.r.t other mode.
Source: (Reginald Souleyrette, and Adam Kirk, 2013)
43
Modal Split: Numerical example 2
Solution:
Uauto = – 0.46 – 0.35(20) – 0.08(8) – 0.005(320) = -9.70
Inference:
Upt = – 0.07 – 0.35(30) – 0.08(6) – 0.005(100) = -11.55 The utility of
automobile stays the
Ubike = – 0.07 – 0.35(35) – 0.08(0) – 0.005(0) = -12.32 same even though its
𝑒 𝑈𝑎𝑢𝑡𝑜 𝑒 −9.70 share of passengers
pauto= = = 0.81 𝑜𝑟 81% reduced
𝑒 𝑈𝑝𝑡 +𝑒 𝑈𝑎𝑢𝑡𝑜 +𝑒 𝑈𝑏𝑖𝑘𝑒 𝑒 −9.70 +𝑒 −11.55 +𝑒 −12.32
𝑒 𝑈𝑝𝑡 𝑒 −11.55
ppt= 𝑒 𝑈𝑝𝑡 +𝑒 𝑈𝑎𝑢𝑡𝑜+𝑒 𝑈𝑏𝑖𝑘𝑒 = = 0.13 𝑜𝑟 13%
𝑒 −9.70 +𝑒 −11.55 +𝑒 −12.32
𝑒 𝑈𝑏𝑖𝑘𝑒 𝑒 −12.32
pbike = = = 0.06 𝑜𝑟 6%
𝑒 𝑈𝑝𝑡 +𝑒 𝑈𝑎𝑢𝑡𝑜 +𝑒 𝑈𝑏𝑖𝑘𝑒 𝑒 −9.70 +𝑒 −11.55 +𝑒 −12.32
Source: (Reginald Souleyrette, and Adam Kirk, 2013)
44
Trip Assignment
• Key step in Urban transport forecasting and planning.
Which Route will the people prefer?
– Shortest route?
– Fastest route?
– Least costly route?
– Scenic route?
• What happens if most cars choose a route?
• How to deal with congestion?
Trip Assignment (cont’d..)
1. Estimates the traffic
volume in each network.
2. Estimates travel costs for
inter zones
3. Identifies bottle necks in
the networks.
Source: (Riza Atiq bin O.K. Rahmat , 2012)
Critiques of the Traditional Planning
Approach
• A tool created for a different goal. i.e. New road
infrastructure.
• The Traditional Planning model “Predict and Accommodate”
i.e. Forecasts based on peak spreading and Induced demand.
• Purely based on Quantitative numbers or Forecasted values.
Basic Approach remains unchanged even
today, but is focused on System Management
Source: (MIT Lecture Notes, 2002)
“The McNamara Fallacy” in Traditional Planning Approach
“Measure whatever can be easily measured”.
implies……………This is OK as far as it goes.
“Disregard that which cannot be measured easily.”
implies ……………This is Artificial and misleading.
“Presume that which cannot be measured easily is
not important.”
implies ……………This is Blindness.
“Presume that which cannot be measured easily does
not exist.”
implies ……………This is Suicide.
Source: (https://expertprogrammanagement.com/2017/07/the-mcnamara-fallacy/)
Conclusion
• The four step Transport forecasting and planning
method ensures selection of the most feasible
transportation option for a selected area considering the
local people needs and cost factors.
• Criticized for “Predict and Accommodate” approach and
“Planning based on “Induced demand”.
System management should be addressed in parallel
to the traditional transportation planning process to
give emphasis on Accessibility rather than Mobility.
References
• Ali Alraouf, (2013). “Introduction to Urban Planning”, Lecture at Qutar University, 2013.
• Anthony J. Richardson, Elizabeth S. ampt and Arnim H. Meyburg, (1995). “Survey Methods for Transport Planning”, Transportation Research Library,
ISBN: 0-646-21439-X, https://trid.trb.org/view/450423.
• Jean-Paul Rodrigue (2020). “The Geography of Transport Systems”, Chapter 9, Transport Planning and Governance, New York: Routledge, 456 pages.
ISBN 978-0-367-36463-2
• Kadiyali, L. R. (2017). ”Traffic engineering and transport planning”, Khanna Publishers.
• Massachusetts Institute of Technology (MIT) Lecture notes, https://ocw.mit.edu/courses/civil-and-environmental-engineering/1-221j-transportation-
systems-fall-2004/lecture-notes/summary.pdf.
• Metrasys, (2012). “Efficient Capacity is Common in Nature”, Transport Planning Guideline and Manual to Best Practice, Metrasys-Sustainable
Mobility for Mega Cities, http://www.metrasys.de/medien/document/Transport_Planning_Guideline.pdf
• Ministry of Urban Development (MOUD), (2016). “Report to the High Powered Committee on Decongesting Traffic in Delhi”, Government of India.
• Richmond Regional Planning District Commission (PlanRVA), (2019). “All Anout Transportation”, Richmond Regional Transportation Planning
Organization Virginia, https://planrva.org/transportation/what-is-transportation-planning/.
• Riza Atiq bin O.K. Rahmat, (2012). “Urban Transport: Transport Modelling”, Lecture Notes, The National University of Malaysia.
• The City Alliance, (2009). “Traffic Intersection Points of Conflict”, Graphic depiction of vehicle and pedestrian conflicts at signalized intersections vs
roundabouts.
Thank You
51
