Ac#vity Opportuni#es and Changing Travel Pa9erns: A case of Developing Na#ons
Varun Varghese| Under the supervision of Prof. Arnab Jana
26th September, 2015| Early Bird Session| 14th Behavior Modeling Workshop in TransportaJon Networks |The University of Tokyo
Contents • Pubic transport crowding (A case of Ahmedabad city)
• Measuring crowding • Public percepJon of crowding • Cost of crowding
• AcJvity-‐travel behavior in Mumbai • IntroducJon • Research gaps • Research objecJves • Preliminary findings for Mumbai • Ongoing work
• Regional travel behavior (A case of inter-‐regional trips in India) • DesJnaJon choice • Mode choice
2
Pubic transport crowding (A case of Ahmedabad city)
India Gujarat
Ahmedabad
3
PopulaJon-‐ 5.6 million Area-‐ 464.14 sq.km
Public transport systems and crowding
4
Objec#ve measures of crowding
Average count-‐ 4.31 standing pax/m2 (4.54 for AMTS, 3.62 for BRTS) Higher during morning peak Crowding peak – 8.32 standing pax/m2
5
Subjec#ve measures of crowding
• Average increase of 18.49 %
• Some routes as high as 48%
• PercepJon dependent on Jme of day, no. of choices, type of service, personal characterisJcs
6
Generalized cost
𝐺𝑒𝑛𝑒𝑟𝑎𝑙ized 𝐶𝑜𝑠𝑡 (𝑐𝑜𝑚𝑓𝑜𝑟𝑡)= 𝑝↓𝑖 +𝑤𝑡↓𝑖 𝑇↓𝑚
-‐ 10.00 20.00 30.00 40.00 50.00
Generalized cost vs Out of pocket cost
Gc (र) Pi (र)
7
Adapted from Haywood and Koning (2013)
Cost of crowding
Sketch by: Shubhadeep Sengupta
-‐
10.00
20.00
30.00
40.00
1.52 1.00 1.10 1.20 1.30 1.40 1.50 1.60
GC-‐AMTS (र)
GC-‐AMTS (र)
-‐
10.00
20.00
30.00
40.00
50.00
1.32 1.00 1.10 1.20 1.30 1.40 1.50 1.60
GC-‐BRTS (र)
GC-‐BRTS (र)
8
Ac#vity-‐travel behavior in Mumbai
India
Maharashtra Mumbai
9
PopulaJon-‐ 12.5 million Area-‐ 603 sq.km
Ac#vity-‐travel behavior in Mumbai-‐ An Introduc#on Time
Space
H
W
W F
FL
LS
H
H
W
F
L
S
Home
Work-‐ Mandatory
Social connecJon-‐ Leisure
Cinema-‐ Leisure
Shopping-‐ Maintenance Travel-‐ Mandatory
Travel-‐ Leisure
Travel-‐ Maintenance
AcJvity Jme-‐ Mandatory
AcJvity Jme-‐ Leisure
AcJvity Jme-‐ Maintenance
Adapted from Jme-‐space prism developed by Hagerstrand (1970) 10
0
20
40
60
80
100
120
<5k 5k-‐7.5k 7.5k-‐10k 10k-‐20k >20k
Income wise distribuJon of acJvity purposes
Work Shopping School Social visit Entertainment
Health Personal Return to home Other
Data source-‐ Baker et. al (2005)
Travelling pa9erns in Mumbai: Descrip#ves
11
0
20
40
60
80
100
120
140
160
180
Zone 1 Zone 2 Zone 3 Zone 4 Zone 5 Zone 6
Commute Jme (mins)
<5k 5k-‐7.5k 7.5k-‐10k 10k-‐20k >20k
Travelling pa9erns in Mumbai: Descrip#ves
Data source-‐ Baker et. al (2005) 12
Changing pa9erns
Data source: ITU (2014)
0.00 20.00 40.00 60.00 80.00
ICT in India
Internet Mobile Phone
© yourstory.com
13
Poten#al effect on travel behavior
H
W
W F
FL
LS
H
H W
W
Time
Space
S H
W
W L
LH
Use of a service which provides informaJon
Stays connected online Online JckeJng
Orders grocery home
W
W
LS
H
L
S
Use of ICT Low capability
Change of spaJal
structure
Socioeconomic factors HH size, relaJonships, gender, age, level of educaJon, vehicle ownership, type of housing, marital status, type of occupaJon, HH/individual income.
Spa/al factors Access to job, ameniJes (such as school hospital etc.). Distance to job centre, jobs to populaJon raJo.
ICT use
Use of smartphones, internet connecJvity, types of ICT based services used, in-‐home and out of home ICT use.
14
Research gaps & Idea#on
Discre#onary ac#vi#es like leisure etc.
(including social trips)
ICT use
Use of smartphones, internet connec#vity, types of ICT based services used, in-‐
home and out of home ICT use.
Socio-‐economic Parameters
HH size, rela#onships,
gender, age, level of educa#on, vehicle ownership, type of housing, marital status, type of occupa#on, HH/individual income.
Mandatory Ac#vi#es like
work, school etc.
Travel #me, mode of travel, mul#ple use of #me, travel #me ra#o, cost of travel, comfort, in-‐home #me spent,
subs#tu#on, complementarity, or modifica#on.
Spa#al parameters
Access to job, ameni#es (such as school hospital etc.), and public transport modes. Distance to job centre, jobs
to popula#on ra#o, neighborhood density, land rent, house rent, wage paid by business firms, no. of sub-‐
centres etc
Maintenance Ac#vi#es like daily shopping, health etc.
15
Research Objec#ves For a society with exis/ng disparity, travel contributes to produc/vity and individuals aren’t just looking to reduce travel but are looking for be=er access to different opportuni/es. In such a scenario, the effects of different endogenous and exogenous variables (pertaining to spa/al change and ICT use) on travel will vary based on the purpose of ac/vi/es and the socio-‐economic characteris/cs of individuals.
a) To idenJfy network factors which affect residenJal choice at a local level in developing naJons taking the case study of Mumbai.
b) To evaluate the importance of travel Jme savings for different socio-‐economic groups. c) Comprehensively model travel behavior adopJng both spaJal parameters and parameters
related to ICT use, along with socio-‐economic, individual and household parameters. d) To idenJfy the change in Jme uses for different cohorts (e.g. age, gender etc.) based on the
above menJoned parameters. e) To evaluate the concept of value of access in context of a developing country like India,
succinctly idenJfying the parameters that can improve access to opportuniJes for a larger secJon of society.
f) To suggest policy recommendaJons related to spaJal change and ICT use, which can improve the overall access to acJvity opportuniJes in a city. 16
Real estate pricing and transport network
K/E
H/EH/W
G/N
Selection of wards Creation of price analysis zones
Select best fiited model (least AICc
and high adjusted R2)
Generate local surfaces
If satisfactory
No
Creation of network dataset
Select the model
Select the model
No1)model performance 2)each explanatory
variable 3)model significance4)stationarity5)model bias 6)residual SAC
Satisfies 6 checks
Run GWR
Run OLS(Consider all
variables)
Varghese, V., Sarkar, S., Bardhan, R., Velaga, N. & Jana, A. (2015). Transport network characterisJcs and real-‐estate undercurrents: The effect of spaJal nonstaJonarity on property pricing in Mumbai. Paper accepted for presentaJon at the 3rd Conference of the TransportaJon Research Group of India (CTRG), Kolkata, India. 17
Access to train stations Access to parking Access to bus stands Access to taxi stands
Access to schools Access to shopping malls Access to business nodes
Access to train stations Access to shopping malls Access to business nodes
Access to train stations Access to parking Access to bus stands Access to taxi stands
Access to schools Access to shopping malls Access to business nodes
Effect of transport network characteris#cs on real estate prices-‐ Results
ResidenJal
Commercial
Office
Model parameters
Residential
Distance to nearest parkingDistance to nearest bus stationDistance to nearest taxi standDistance to nearest schoolDistance to nearest mall Distance to nearest business node
CommercialDistance to nearest mall Distance to nearest business node
Office
Distance to nearest train stationDistance to nearest parkingDistance to nearest bus stationDistance to nearest taxi standDistance to nearest schoolDistance to nearest mall Distance to nearest business node
Red color imply negaJve associaJon
18
Effect of transport network characteris#cs on real estate prices-‐ Concluding discussions
Residential Commercial Office
• Real estate market is influenced by speculaJon, which is largely a local phenomenon.
• Distance to train staJons negaJvely influenced residenJal property prices at Bandra-‐Kurla complex and distance to bus stops had a severe negaJve impact on housing prices in Dharavi, Dadar and Andheri.
• Distance to schools also have negaJvely influenced both residenJal and office property prices, while the distance to shopping malls and business districts have emerged as a significant factor in esJmaJng property prices for all different types land use.
• A strong correlaJon between office and residenJal prices hint towards the willingness of people to stay near their workplaces. 19
Travel #me savings or the value of access-‐ case study of Mumbai-‐ Concept
When people are less capable to enjoy “beings and doings” which add value to life, then people a;ach more importance to the produc=on aspects of travel, and gradually put more value on the consump=on aspects of travel with the increase in their capability.
1. Travel #me is significantly related to individual capability and is lower for less capable
2. The variance of travel #me, indica#ng the degree of freedom of movement, has posi#ve associa#on with individual capability
Chikaraishi, M., Jana. A., Bardhan, R., Varghese, V., Fujiwara, A. RevisiJng travel Jme in Mumbai: the value of Jme saving or the value of access? Paper submiqed at the 95th TRB annual conference. 20
Travel #me savings or the value of access-‐ case study of Mumbai: Study area and data collec#on
Type of housing Individuals Households
MIG 48 22
Slum 58 23
SRA 52 28
Total 158 73
21
Travel #me savings or the value of access-‐ case study of Mumbai: Results and discussion
• The mean capability and travel Jme of individuals from MIG housing was found to be significantly higher than those of Slum and SRA.
• The variance in travel Jme increased with the capability of the individual
• Individuals with higher capability had the ability to spend more Jme to discreJonary acJviJes, whereas people in the lower order of capability were constrained to spend more Jme in mandatory acJviJes.
• MIG have more flexible mode choices whereas other groups rely more on walking.
• Travel Jme savings would substanJally unde resJmate the benefi t o f t he infrastructure investment.
• Exploring the concept of value of access w.r.t. spaJal policies and ICT use 22
On going work
0% 20% 40% 60% 80% 100%
Smart phone/Internet use vs housing type
MIG SRA Slum 0% 10% 20% 30% 40% 50% 60% 70%
Smart phone/Internet use vs gender
Male Female
23
On going work
0% 10% 20% 30% 40% 50%
Online JckeJng
e-‐shopping
raJng app (complimentarity)
Social networking apps
Cab services
Games
Transport informaJon
Type of services used
0% 10% 20% 30% 40% 50%
Reading Listening to music
Uses internet Gets bored
Window gazing Sleeping
Playing games EaJng
Working Talking to other passengers
Time use while traveling
24
Inter-‐region travel in India-‐ Des#na#on choice
Upsurge in economic acJviJes
PolarisaJon of faciliJes in urban areas Regional disparity
leading to pilgrimage trips
increased social trips
Trips due to healthcare & other needs
“to analyse regional travel paqerns in India by studying social, health, and pilgrimage trips” This might be significant to (1) Beqer understand travel with respect to the exisJng disparity in infrastructure provisioning and economic opportuniJes (2) Analyse the purpose that prompted the travel acJvity, (3) CharacterisJcs of the trips based on mode used, duraJon of the trip and spaJal locaJon of the desJnaJon, and (4) Policy implicaJons related to long distance travel
MigraJon
SaJsfacJon enhancement
Gaps in infrastructure
Bardhan, R., Varghese, V., & Jana, A. (2015). Analyzing Regional Travel Patterns in India: Disaggregated Analysis of Social, Health and Pilgrimage Trips. Journal of the Eastern Asia Society for Transportation Studies (in press)
Destination within the district
Des/na/on outside the district but within the state
Des/na/on outside the state but within the country
Healthcare trips
Conceptual Framework
26
q Is there any significant difference in travel paqerns with respect to ‘within district’, ‘outside district but within state’ & ‘beyond district’?
q On what factors do long distance travel depend in India?
q “Clustering of opportuniJes/infrastructure might hold the key to longer distance trips in India”
Research methodology
27
NSS data on domesJc tourism
MNL models Outcome variable: Trip desJnaJon
ESDA for social, health & pilgrimage trips Unit of analysis: District
MNL for overnight trips
MNL for sameday trips
Local Moran’s I in the type of trips viz. Social, health and pilgrimage trips.
Policy implicaJons, discussions and future scope
Destination within the district
Healthcare trips
O/N trips (Findings)
28
Des/na/on outside the district but within the state Number of HH members in the trip (-‐) Number of places visited (+) Mode: Bus(+), Train (+), Own (+) Time of year: July to Sept (-‐) Purpose: Social (-‐), Health(-‐), Pilgrimage (+) Type of stay: Hotel (+), Friends (-‐)
Des/na/on outside the state but within the country Number of HH members in the trip (-‐) Number of places visited (+) Mode: Bus(+), Train (+) Time of year: April to June (-‐), July to Sept (-‐) Purpose: Social (-‐), Health(-‐), Pilgrimage (+)
Destination within the district
Des/na/on outside the district but within the state Number of HH members in the trip (+) Number of places visited (+) Mode: Bus(+), Train (+), Own (+) Time of year: July to Sept (-‐) Purpose: -‐ Social (+), Pilgrimage (+) Type of stay: Hotel (+), Friends (+)
Des/na/on outside the state but within the country Number of HH members in the trip (+) Number of places visited (+) Mode: Train (+), Own(-‐) Purpose: Social (-‐), Pilgrimage (+)
Healthcare trips
Sameday trips(Findings)
29
I= 0.297
I= 0.147
I= 0.338
I= 0.158
Clustering and Spa#al Autocorrela#on
I= 0.01923
I= 0.0596
I= 0.097
I= 0.059
I= 0.2617
I= 0.207
I= 0.142
I= 0.082
Social trips
Pilgrimage trips
Health trips
Discussions
• Lack of employment opportuniJes and other allied services such as educaJon, force people from rural areas to migrate to urban areas. • Current healthcare planning in India hierarchically allocates healthcare faciliJes based on certain populaJon thresholds, which creates a lopsided distribuJon of sophisJcated health centres in urban areas. • Study provides the basis for travel induced infrastructure delivery planning. Transport policies reducing impedances towards social and pilgrimage acJviJes, while provisioning of infrastructure to reduce the need for travelling for health acJviJes can subsequently lead to the improvement of quality of life.
31
ありがとうございます
You can reach me at [email protected] or [email protected] or hqps://twiqer.com/varunvarghese
32