From transport to accessibility: the new lease of life of an old concept

Transcription

1 Paris 07 /01/ 2015 From transport to accessibility: the new lease of life of an old concept Pr. Yves Crozet Laboratory of Transport Economics (LET) University of Lyon (IEP) - France yves.crozet@let.ish-lyon.cnrs.fr

2 Urban mobility: some new trends Some paradoxical changes in urban mobility policies A public preference for slow modes (tramway, bus, bikes, pedestrian streets..) Voluntary restrictions to car mobility (lower speed, reduced width of roads ) Towards a lower accessibility? A risk for urban attractiveness? The decreasing relevancy of Cost Benefit Analysis (CBA) to define public choice

3 Market as a model Price Consumer Surplus Supply Pe Demand Supplier Surplus Qe 3 Quantities

4 Contents 1 Transport and accessibility: a common dream of unidimensionality (economics and geography) a) Travel demand and accessibility b) Key variables and parameters c) Accessibility and GIS 2 The new lease of life of accessibility: bad and good reasons 4

5 I prix coût Cg mi = cm + voti t m A surplus initial cij2 S = Tij dcij c1ij d1 c1 d2 c2 O variation de surplus T1 T2 D déplacements entre zones i et j

6 Time gains and CBA NPV = BNA j =tn tr j =t p tr I j + R j C j + A j (1 + a) j + K tn (1 + a) 6 tn tr

7 The key ratio A city is a place where the number of opportunities is higher than in other places Travel cost (time + money) is compared to the corresponding amenities What we have to take into account is the ratio Opportunities Travel cost

8 Definition Accessibility at point 1 to a particular type of activity at point 2 is directly proportional to the size of the activity at point 2 and inversely proportional to a function of the distance separating the two points. The total accessibility at point 1 to the activity is the summation of the accessibility to each of the points around point 1 (Hansen 1959, p. 74 ff)

9 The components of accessibility Size of the city Average revenue/cap. Available space Available modes Infrastructure network Average speed/mode Travel time budget Travel Monetary budget Density of jobs Density of population Location of jobs Location of residential a. Quality of Public transit Etc.

10 We can describe accessibility as a function of territorial structure and transport supply. Ai = D j f (cij ) with j Ai = Accessibility to destinations D from point i Dj = Activity destinations at points j cij = Generalized cost (time, price )

11 I Demand Curve? Tij = Oi Wilson (1970) (Hansen, 1959) D k f (cik ) k f (cij ) = exp( βcij ) Accessibility D j f (cij ) Ai = Dk exp( βcik ) k

12 I Surplus change for zone i cij2 Si = T dc ij ij zones j c1 ij Neuburger (1971) ( ( D j exp βcij2 Oi j Si = ln β D j exp βcij1 j = Koenig (1974) ) O = ( ln A ) β i 2 i ln A 1 i )

13 Time gains and accessibility The first age of urban accessibility is dominated by the search of speed and especially road speed for cars. In this age, speed is the main factor of the improvement of urban accessibility. The dream of policy makers is the same than the dream of car drivers: to keep car accessibility during peak hours at the same level than during off peak hours.

14

15

16 Economic theory of urban accessibility Hansen 1959, Koenig 1974 Ai = D j exp j ( βcij ) Parameter Sensitivity to Generalised cost Generalized cost «Attractive Masses» Monetary cost + Travel Time + parameters Housing Jobs Shops, Leisure

17 A GIS TO MODEL ACCESSIBILITY (MOSART) Shops et services PT network Road network Administratives areas Census data Orthophotos

18 MOSART (MOdeling and Simulating Accessibility to networks and Territories) MOSART : an innovative modelling platform for planning sustainable mobility

19

20 MOSART Version 2 : case study

21 Road congestion during morning peak hours

22 MOSART Version 2 : case study

23

24 The second age of urban accessibility emerges when road congestion is recurrent during peak hours. Public transit (PT) becomes the priority to keep, or to improve urban accessibility. Speed is therefore less important than reliability, frequency and comfort of PT. Even some relatively slow modes like bikes are promoted.

25 Potential passenger flow in a dense area per hour for a 3.5 meters width infrastructure

26 Contents 1 Transport and accessibility: a common dream of unidimensionality (economics and geography) 2 The new lease of life of accessibility: bad and good reasons a) Accessibility gains and economic growth b) Accessibility and effective speed c) Accessibility and land use 26

27 Accessibility and economic growth??? The main belief of decision makers is the following Higher speed => time gains => accessibility improvements => more mobility => economic growth Behind the official appraisal methodology, lies a direct relationship between accessibility and productivity 27

28 TOTAL MOBILITY (Data Points: ; Curves: ) Source: Schafer and Victor (2000); economic growth rates based on IPCC IS92a/e scenario 28

29 Where does the growth come from? Aij = ΣDj. exp (- βcij) Within Aij you have a relation between the utility and the cost of trip Finally you have a relationship Si = Ui Ci = (λ * ln Ji40) Ci The surplus depends on the number of jobs accessible in less than 40 More jobs are accessible => economic growth 29

30

31 From a correlation to a causality GDP/ cap Number of accessible Jobs in 40 minutes

32 1 1 + Traffic Generalized + cost GDP + + GDP/capita 32

33 1 1 1 Traffic + Generalised cost GDP GDP/capita

34 1 1 1 Traffic + Generalised cost GDP GDP/capita

35 Energy and Equity (Ivan Illich and J.P. Dupuy 1973) The more you increase speed, the more you reduce equity From the generalized cost to the generalized speed or «effective speed» Ef. Speed = 1 / (1/S) + (k/w))

36 «Effective Speed» and «Social Effective Speed» Average speed = harmonic High speed rail effective average speed n/ [(1/V1)+(1/V2)] 1/ [(1/200)+ (0.15/8)] = Effective speed of bike 40km/h 1/ [(1/V)+(k/W)] High speed rail social 1/ [(1/14)+ (0.001/8) = effective speed / [(1/200)+ (0.5/8)] = Supersonic effective speed 14.8km/h Regional train 1/ [(1/2000)+ (1/6)] = 6km/h 1/ [(1/100)+ (0.30/8)] = 21km/h

37 Social effective speeds

38 Land use and accessibility The third age of urban accessibility appears when public policies are no more focused on transport. When the priorities become land use and density. Initially based on individual transport costs, the concept of accessibility is now used to put the focus on land use as a collective process. Urban policies are no more promoting accessibility for all parts of the city

39 Road congestion during morning peak hours

40

41 Road congestion with the new infrastructure during morning peak hours

42 Accessibility change with the new infrastructure Accessibility gains 42

43 Accessibility change with the new infrastructure Accessibility gains to jobs outside from Grand-Lyon Accessibility gains 43

44 Conclusion The coming challenge of urban mobility is: how public policies on land-use and transport supply can help to keep an acceptable level of accessibility in a context of rising travel costs From How accessibility shapes land use (Hansen 1959) To How land use and environment priorities shape accessibility policies

45 Time gains versus land use: what is at stake? Objective function of the decision maker? Speed and time gains Land use and environment Decision-maker? Benevolent and all-knowing ruler Ignorant or opportunist decision maker Accessibility in Accessibility index CBA, transport and map-based demand and traffic tools growth Cost efficiency analysis, cost of accessibility gains, «social effective speed Accessibility maps to open the black box of transport demand modelling

46 Conclusion (2) Ai = D j exp( βc ij ) j Tell me what accessibility you are ready to promote, and how (mode? speed? reliablity? density?) and for who? Tell we what accessibility you are ready not to promote, and even to reduce And I ll tell you what city you prepare for tomorrow!

47 Density and new apartments per municipality(1991) Sources : SITADEL, INSEE

48 Density and new apartments per municipality (2007) Sources : SITADEL, INSEE

49

50 Road traffic in Paris 1953 and 1969

51 Tramways in Paris Region within 2020

52 Searching the relevant decay function

53 Results Île de France Exponential ( min-1) Power Mix 0,182 2,792 α=94407 (sans unité) β= 1,21 (sans unité) γ=1,08 (min ) -1 Rhône-Alpes 0,159 3,28 α=48122 (sans unité) β= 1,12 (sans unité) γ=1,033 (min ) -1 Limousin 0,035 1,554 α=10195 (sans unité) β= 2,25 (sans unité) γ= 0 (min ) -1 France 0,134 3,97 -

54

55 Searching the best function Exponential national coefficient Ile de Franc e Rhône -Alpes Limou sin Exponential local coefficient Power Local coefficient Mix local coefficient 0,605 0,580 0,590 0,534 0,474 0,457 0,670 0,455 0,340 0,719 0,497 0,419

56 3 Accessibility index: for decision makers or for public debate (A 65 motorway)? A 65 Motorway Southwest of France Open in 2011 Expected traffic v/d Traffic in June !

57 Municipalities at less than 40 minutes Few jobs along the motorway High density of jobs in the main cities

58