Urban Transportation Planning – Fall 2006
Assignment no 4
„
„
Let us begin by a world tour
thanks to J.Kenworthy and F.
Laube – “The Millennium
Cities. Data base for
Sustainable Transport”
sponsored by the UITP
A follow-up to the 1989
“Cities and Automobile
Dependence” by P.Newman
and J.Kenworthy
1
Urban Transportation Planning – Fall 2006
… And it is complemented by its follow-up
„
A set of 50 cities with evolution of performance between 1995 and 2001, plus mobility policies in 2006
2
Urban Transportation Planning – Fall 2006
Urban Density is critical…
200
URBAN DENSITY
160
Persons/Ha
120
80
40
0
Western Eastern North
Latin
Middle
Europe Europe America America East
Africa
Asian
Other
Affluent Asian
(55.1)
(101.67)
(134.36)
(71.36)
(18.67)
(89.66)
(76.79)
(189.75)
Ocean A
(15.02)
World Regions
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
.
3
Urban Transportation Planning – Fall 2006
The higher the density, the higher…
Modal Choice (% Public TransportCycling-Walking)
Density (Inhabitants/Hectare) vs Modal Choice (% Public Transport
+ Cycling + Walking)
100
All Cities
80
60
40
U.S.A.
Western Europe
20
0
20
40
80
160
320
Density (Inhabitants/Hectare)
… the percentage of sustainable modes
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
4
Urban Transportation Planning – Fall 2006
Job density is also a critical parameter
30
Proportion of Jobs in CBD
25
20
Percent
As Joel Garreau says
(The Edge Cities),
when the president
moves to the
suburbs, he takes
the office along
15
10
5
0
Western Eastern North
Latin
Middle
Europe Europe America America East
Africa
Asian
Other
Affluent Asian
(18.29)
(14.03)
(20.08)
(17.03)
(11.34)
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
(22.6)
(13.32)
World Regions
(25.53)
Ocean A
(15.06)
Urban Transportation Planning – Fall 2006
Number of trips, nearly a constant
The number of
trips result from
the activities
profile
„ But be aware that
non-motorized
trips may go
unaccounted for,
in some surveys
„
4
Total Daily Trips per Capita
Trips/Person
3
2
1
0
Latin
Western Eastern North
Middle
Europe Europe America America East
(2.89)
(2.81)
(3.5)
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
(1.71)
(2.34)
Africa
(1.59)
World Regions
Asian
Other
Affluent Asian
(2.66)
(2.13)
6
Ocean A
(3.86)
Urban Transportation Planning – Fall 2006
Percentage of non-motorized trips
* Percentage of Non Motorised
Modes Over all Trips
40
30
Is this a surprise?
Percent
20
10
0
Latin
Middle
Western Eastern North
Europe Europe America America East
Africa
Asian
Other
Affluent Asian
(31.04)
(42.17)
(29.05)
(26.22)
(8.86)
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
(24.65)
(14.96)
World Regions
(39.9)
Ocean A
7
(15.8)
Urban Transportation Planning – Fall 2006
… not to be taken lightly
From Mean
Streets 2000 by
the Surface
Transportation
Project Policy
(STPP)
Percent of People who are
Overweight
0.50
0.40
0.30
0.20
0
0.05
0.10
0.15
(one-tenth of a mile)
0.20
0.25
0.30
(quarter mile)
Fewer People are Overweight in Places where People Walk More
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
8
Urban Transportation Planning – Fall 2006
Road building
m/1000 Persons
Notice that the
U.S. ratio doubles the one of Western Europe
8000
Length of Road per 1000 People
6000
4000
2000
0
Latin
Middle
Western Eastern North
Europe Europe America America East
(2997.88) (1373.68) (6080.07)
Africa
Asian
Other
Affluent Asian
(1464.06) (1640.14) (1425.53) (2424.76)
(591.06)
Ocean A
(8059.89)
World Regions
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
9
Urban Transportation Planning – Fall 2006
Automobile ownership
600
Passenger Cars per 1000 People
500
… is congestion in
Western Europe
higher than in the
States?
400
Units/1000 People
As the difference is
not as big as the
supply of roads…
300
200
100
0
Latin
Middle
Western Eastern North
Europe Europe America America East
Africa
Asian
Other
Affluent Asian
(411.86)
(102.12)
(217.33)
(279.23)
(567.95)
(188.53)
(185.26)
World Regions
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
(88.3)
10
Ocean A
(575.36)
Urban Transportation Planning – Fall 2006
Trip length by car
20000
Passenger Car Kilometers per Car
14000
If the number of trips are comparable…
Km/Unit
Does the average car
trip length increase
inversely
proportional to the
metropolitan
density?
12000
8000
4000
0
Latin
Western Eastern North
Middle
Europe Europe America America East
(11324.44) (7576.85) (18460.33) (9883.8)
Africa
Asian
Other
Affluent Asian
(14887.83) (11834.34) (12195)
Ocean A
(9963.11) (12904.22)
World Regions
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
11
Urban Transportation Planning – Fall 2006
Transit coverage
Total Length of Public Transport Lines per 1000 People
4000
3000
m/1000 Persons
Notice that the
Western
Europe ratio
more than
doubles the
US ratio
2000
1000
0
Western Eastern North
Latin
Middle
Europe Europe America America East
(3634.16)
(1684.41) (1590.66)
(4227.79)
(690.9)
Africa
Asian
Other
Affluent Asian
(3940.63) (1127.85)
Ocean A
(1327.23) (3204.86)
World Regions
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
12
Urban Transportation Planning – Fall 2006
Parking supply in downtown
800
600
Spaces/1000 Jobs
Again the U.S.
leads clearly
over Western
Europe
Parking Spaces per 1000
CBD Jobs
400
200
0
Western Eastern
Europe Europe
North
Latin
Middle
America America East
Africa
Asian
Other
Affluent Asian
(270)
(500.15)
(195.57)
(120.59)
(74.04)
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
(71.38)
(790.93)
World Regions
(91.25)
Ocean A
(505.24)
13
Urban Transportation Planning – Fall 2006
The role of the automobile
* Percentage of Motorised Private Modes Over all Trips
80
Again, this should
come as no
surprise
Percent
60
40
20
0
Latin
Western Eastern North
Middle
Europe Europe America America East
Africa
Asian
Other
Affluent Asian
(50.24)
(32.88)
(38.62)
(27.96)
(85.83)
(35.51)
(72.93)
Ocean A
(31.95)
(79.12)
14
World Regions
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
Urban Transportation Planning – Fall 2006
Economic efficiency
„
The arguments
go well beyond
environmental
concerns,
quality of life
issues, moral
grounds…
Clear economic
consequences
25
Total Passenger Transport Cost
as Percentage of Metropolitan GDP
20
15
Percent
„
10
5
0
Latin
Middle
Western Eastern North
Europe Europe America America East
Africa
Asian
Other
Affluent Asian
(8.29)
(21.68)
(5.4)
(14.77)
(12.44)
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
(13.97)
(9.27)
(13.65)
Ocean A
(13.46)
World Regions
15
Urban Transportation Planning – Fall 2006
The cost of a balanced system
Modal Choice (% Public Transport + Cycling + Walking) vs
Transport Cost (% GDP)
Transport Cost (% GDP)
20,0
Developed Cities
16,0
12,0
8,0
U.S.A.
4,0
0,0
Western Europe
0
10
20
30
40
50
60
70
80
90
Modal Choice (% Public Transport - Cycling - Walking)
Economic sustainability
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
16
Urban Transportation Planning – Fall 2006
CO emissions per capita
200
* Emissions of CO per Capita
160
The Environmental cost
Kg/Person
120
80
40
0
Latin
Middle
Western Eastern North
Europe Europe America America East
Africa
Asian
Other
Affluent Asian
(70.9)
(73.46)
(18.06)
(71.64)
(183.01)
(70.71)
(181.4)
(53)
Ocean A
(146.19)
World Regions
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
17
Urban Transportation Planning – Fall 2006
The high price of road fatalities
Modal Choice (% Public Transport + Cycling + Walking) vs Fatalities
Fatalities (Per Million Inhabitants)
250
Developed Cities
200
150
100
50
0
U.S.A.
Western Europe
0
10
20
30
40
50
60
70
80
90
Modal Choice (% Public Transport - Cycling - Walking)
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
18
Urban Transportation Planning – Fall 2006
A congestion index
„
„
In spite of the
differences, a similar
congestion ratio
2500000
One reason the
higher trip length in
the States
Is traffic like an
expanding gas?
Total Private Passenger Vehicle Kilometers per Kilometer of Road
2000000
1500000
V.km/km
„
1000000
500000
0
Latin
Middle
Western Eastern North
Europe Europe America America East
Africa Asian
Other
Affluent Asian
(1793686.45) (1783284.03) (1840224.45) (1972405.23) (2060896.34) (849220.9)
Figure by MIT OCW, adapted from the UITP Millennium Cities Database.
World Regions
Ocean A
(1420566.31) (1684813.84) (938709.62)
19
Urban Transportation Planning – Fall 2006
Some Lessons by the Database Authors
1. Transit is the least cost option when density
is higher than 20 people per hectare
2. Growth and income does not necessarily
imply sprawl and automobile reliance
3. In fact, sharp growth in automobile traffic in
cities in developing countries may
compromise economic development
4. In sprawling affluent cities where the auto
dominates, speeds may be high but so are
the total travel times ( while no-car
households may lack accessibility)
20
Urban Transportation Planning – Fall 2006
Lessons by the Database Authors
5. Car ownership while usually has an impact
on transit ridership, may be balanced
through mobility management policies
6. Parking policies are critical to curb the auto
and provide transit priority
7. Transit supply and demand are clearly
correlated
8. Rail modes are more attractive and
competitive, and most cost efficient in
major cities
9. The ratio of transit to automobile speed is
most critical to explain modal split
21
Urban Transportation Planning – Fall 2006
Lessons by the Database Authors
10. Low transit fares, while necessary in social
terms, are not critical to capture motorists
11. Sustainable policy = urban planning +
traffic and parking controls + public
transport with traffic priority
12. In developing countries, transit supply
shortage and high operator profits
See “Millennium Cities Database for Sustainable
Mobility. Analyses and Recommendations”
by Jean Vivier, UITP
22
Urban Transportation Planning – Fall 2006
Our suggestions and lessons
Main issue is quality of life
„ Not auto vs. transit, rather auto versus
transit + non-motorized trips
„ The virtuous cycle: low fares higher
ridership higher frequencies
„ Accessibility varies greatly within a
metropolitan area (walk trips in Beacon
Hill and auto trips in Framingham)
„
23
Urban Transportation Planning – Fall 2006
Our suggestions and lessons
„
„
„
„
Cost based on local GDP obscures the fact
that some costs are local (ie driver wages)
with multiplier effects, while others are
international (ie. fuel, automobiles…)
If you substract the transportations costs
from the average GDP in a city, you may find
that many such cities have a higher GDP than
equivalent US cities
Metropolitan area as an archipelago of
transport arrangements
Let us continue the discussion on how to deal
with the “ocean of red circles”
24
Urban Transportation Planning – Fall 2006
A few tips…
„
„
„
„
„
„
„
Frustration while interpreting someone else
data (or sometimes even yours)
Work in pairs with complementary skills
We are dealing with a complex problem (with
social, economic and technical issues), simple
hypotheses will not suffice (no magic bullets)
Simple linear regression or multi-regression?
“We love our cars in the US” (and in Rome,
Nairobi, New Delhi, Copenhagen ..)
The myth about income and transit use
Do work with this database in the future
25
Urban Transportation Planning – Fall 2006
„
Good luck and enjoy…
26