Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Transcript of HKSTS 2012
Why Shanghai? Past Studies Location-based Convenience Sampling
Survey Implementation Dependent Variables(3 models)
Development of Discrete Choice Model
Result Interpretation Built Environment vs. Vehicle Use
Weekday vs. Weekend The Fundamental What is "Motorization"? DEFINITION Motorization is the transition from non-motorized to motorized transportation modes. Motorized modes are powered by fuel (gasoline, electricity, LPG, etc.) instead of human-power. Example #1 Example #2 Example #3 MOTORIZATION IN CHINA Background Information Large (developing) country: 1.3 billion people Motor Vehicle Ownership: Motor Vehicle Fleet Size: China had approximately the same number of cars per capita in 2003 (18 vehicles/1,000 people) as the United States had in 1910 (Schipper, et al. 2004)
China (300%; 1999~2002) China (1000%; 1980~2002) vs. US. (36%; 1980~2002) CHINA IS BIG , SO GROWTH DOES COUNT ! How to explain car-buying craze (2002~)? Price-cut (due to business competition)
Lower tariffs (was 100% on imported car)
In 2002, 238 administrative fees were cancelled
Car loans (1.2 billion USD, in 2002)
Growth of personal income (GDP/capita) “At the national level, income alone typically explains more than 90 percent of the variation in motorization levels … The growth of national motor vehicle fleets parallels that of income: a 1 percent increase in income is associated with a 1 percent increase in motor vehicles, and this relationship has been relatively stable for the past 30 years” Personal Cars and China. National Academy Press, 2003 factors of motorization (at National Level): In Addition, Income or GDP/capita (+)
Population Density (-) WHY SHANGHAI ? Shanghai What do we know about Shanghai? Major port, Commercial center, “Door of China”
Population: 23 million (registered + “floating population”) 
Density: Ave. 2,116 (people/km2); Highest 49,854 (people/km2); Lowest 610 (people/km2) 
Urbanization: 50% of population is clustered in 5% urban area No single city can represent China - Why Shanghai? Reason #1 Shanghai is... Wealthy Shanghai:
over $ 4000 GDP/capita Source: China Statistic Year Book (2002), Exchange Rate: 1 USD = 8.28 (RMB) Reason #2 Shanghai is... Diverse Mobility Alternatives: Walk and Bicycle (non-motorized)
Public Transportation (subway, light rail, bus)
Auxiliary Power Vehicle (2 or 3-wheeler powered by NG, electricity)
Motorcycle (2 or 3-wheeler powered by gasoline)
Car (private, shared, taxi, rented) Reason #3 Shanghai is... A Statistical Outlier A scattered pattern
Car-restricting policies: Quota Control of the License Plate: The auction system How much does it COST to own a car? Shanghai v.s. U.S.A. Vehicle Registration Fee: $12 (USD)
Vehicle License Fee: As high as $4,300 (USD) Through the auction system (hold by Shanghai Vehicle Management Bureau). A cheap Geely Sedan (1L, no AC, no radio) is only $3600
Lots of complaint, government increased the quota. (e.g. to 6,000 licenses, June 2004) 2005 2005 Vehicle Registration Fee: $40 (USD)
Vehicle License Fee: $114 (USD) The "Car Debate" in Shanghai Issue #1 Issue #2 Issue #3 Issue #4 Regarding Issue #1 " Quick Fix " Regarding Issue #2 What is the FUNDAMENTAL ISSUE? Registered in nearby provinces but used in Shanghai. Big resistance from car companies/dealers. Many local people employed in car companies. (purchase/parking subsidy), (more knowledge about car desire a car life) The policy (learn from Singapore and Sydney) of promoting public transportation and restricting car use is NOT consistent with Central Government in Beijing. (Auto as Pillar Industry “Household Car” concept) Non-Shanghai-licensed car can not use certain expressways during peak hour (manually enforced by patrol) New policy considered – Separate car “use” from “purchase” (e.g. no license control, but congestion pricing) Registered in nearby provinces but used in Shanghai. Quick Fix Big resistance from car companies/dealers. Quick Fix Introduction Literature Review Any Others? Personal Cars and China. National Academy Press, 2003 1970-1996 Research Methodology: Past Survey Studies Location-based Convenience Sampling: “… Conducting a random household survey in China is logistically and institutionally difficult. As a result, targeted intercept surveys were conducted at locations that contain a representative sample of urban two-wheel vehicle users, specifically centralized parking facilities of major activity centers and trip generators throughout the urban area.” ~ Christopher, Cherry R. (2007) Electric Two-Wheelers in China: Analysis of Environmental, Safety, and Mobility Impacts, Doctoral Dissertation, University of California, Berkeley Vehicle Purchase Behavior: Choice Model Motorization = What is it? Vehicle Purchase Behavior: China Studies Not many studies using vehicle choice model Vehicle attributes: Study #1: Four level-one cities, 2004 Study #2: Hong Kong, 2001 401 car-owning residents
Exogenous attributes: “Deterrence factors of driving” (Traffic Congestion is #1) However, no coefficients, no statistical significance Choice Model = Why? and How important certain reason is? Random Utility Theory: Utility of decision maker n toward alternative i (qualitative description of phenomenon) (quantitative and statistical) (g = market segment) Research Methodology MOTORization Pathway VEHICLE PURCHASE/USE BEHAVIOR Conclusion Motorization: Facts and Forecasts Safety is #1
Price is #7 Car Purchase Criteria (Shanghai, Beijing, Guangzhou, Shenzhen);
Source: Mercer Consulting Group, 2004 Source: S. Cullinane, K. Cullinane / Transportation Research Part D 8 (2003) 129–138 Location-based Convenience Sampling Non probability-based (non random) sampling: Location-based sampling:
a location-based variant of convenience sampling Unable to generate a random sample for the population (i.e. all Shanghai residents)
Not necessary. Since we don’t use data for “description” but for “explanation” Vehicle Choice: Car dealership Car Owners
Mobility Characteristics: Income , Transit Access Survey Design [Pilot Survey] Survey Design [Final Survey] Goal: Pretest the final survey + Getting basic sense of the geographical and socioeconomic context Sampling: Location-based (Convenience) Sample: Ferry Motorcycle or APV riders ; Wal-Mart “Salary” class
Convenience Sample: Peer network Distribution: On-street, Email Questionnaire: vehicle purchase/use, demographic information, related to present, past and future (mostly open-ended question) 10 Survey Locations Survey Implementation: Period: October 14th ~ November 20th, 2005
Response Rate: 63.5% (122/164)
Reward: A Tongji University sports cap. ($1 USD)
Team Selection and Training: 2 graduate students from Auto Automobile Marketing and Management School, Tongji University.
Able to communicate in both Mandarin and Shanghainese
The “Every 5th-Person” rule Lessons Learned 1. About Questionnaire: “Dream life” is vague for most people 11% of respondents declined to answer at all; another 54% provided either vague or partial answers.
Many people in Shanghai may not be ready yet to imagine a future very different from their past and present, raising difficulties for studying potential hypothetical topics (e.g. alternative fuel vehicle). 2. About Implementation:
Trust, convenience, and comfort are keys to success Lowest response rate (40.7%) in Renmin Square subway station.
The bookstore inside subway station to add convenience and comfort 3. “Getting to know the place” is also important The implementation of on-street interviewing in the pilot phase helped us finalize the location selection for the final phase. Goal: Test the motorization hypothesis + provide the data for the vehicle purchase choice model Sampling: Location-based (Convenience) Sample: Ford dealership+Driving school Car owners; Households (selected by income x transit access)
Convenience Sample: Peer network
On-line (random) Sample Distribution: On-street, Dealership, Household, Email, Internet + 41,754 cell phone messages Questionnaire: vehicle purchase/use, utility comparison, personality/lifestyle, exogenous environment (Liker-scale type question) 24 Survey Locations Survey Implementation: Period: : July 28th ~ October 27th, 2006
Response Rate: 65.8% (1024/1555);
Internet (with cell phone text message): 0.18% (76/41,754)
Reward: Taxi coupon ($2.5 USD) or i-Pod
Team Selection and Training: Graduate/undergrad students from school of transportation engineering, Tongji University.
Understand the topic
Be serious and with passion
Training session for different scenarios
The “Every 5th-Person” rule Lessons Learned The necessary complexity affects the willingness of participation Anonymity/confidentiality, Authorization, and Study topic are top
three factors affecting people’s motivation to participate [Final Survey Part 6: 1,024 respondents ; %] 2. About Implementation: 1. About Questionnaire:
A complicated and long survey is challenging 1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Pop Quiz !! 40 41 42 43 44 45 47 48 49 50 52 53 54 1 8 9 10 12 13 15 17 18 20 21 22 24 25 27 28 29 30 31 33 34 35 37 38 42 43 45 47 48 49 51 52 54 2 2 3 39 4 40 5 7 22 23 26 32 41 44 50 51 6 11 14 16 19 36 46 46 53 Motorization Pathway Question Exact Patterns (311 patterns generated from 992 cases) How Good Are Your Eyes? Walk Bicycle Public
Transportation Taxi or
Rented Car The most frequent pathway, containing 6.5 percent of the sample, is Number of Stages More “complicated pathways” than “simple pathways” TREE DIAgram Analysis Motorization Forward (Consider first 2 ~ 4 stages) Motorization Backward (Consider first 2 ~ 4 stages) Summary of Tree Diagrams Tree Diagram – a way of data reduction Partial vs. Complete Look of Motorization Pathway Hypothesis Test Hypothetical Motorization Direction Hypothesis Test Against the 3 Rules? [Backward View] People might perceive “cost” differently from what we hypothesized "Soft" Accept Hypothesis! About half of the sample Dependent Variables: Purchase & Use 3 models were developed: Most Expensive Vehicle Owned
Most Frequently Used Travel Means (weekday)
Most Frequently Used Travel Means (weekend) Explanatory Variables Factor of Exogenous Environment Factor of Lifestyle Data Cleaning, Imputation, and Consolidation Model Development Flow Chart Model #1: Most Expensive Vehicle Owned Model #2: Most Frequently Used Travel Means (weekday) Model #3: Most Frequently Used Travel Means (weekend) FINDING & Policy Implications Built Environment and Vehicle Use (weekday/weekend) Among two major types of built environment variables (i.e., car parking availability, and distance to transit), “parking availability” has direct but negative impact on the use of public transportation (weekday model). However, the effect of “distance to transit” is more latent, with some positive impact on use of motorcycles, but no direct influence on the use of public transportation (weekend model). By considering the built environment with many other variables (e.g., income, status-seeking lifestyle) in the same model, as noticed, not as many built environment variables can be identified as expected. That is, the effect of built environment might be suppressed by other demographic and lifestyle variables. Policy Implications from Weekday and Weekend Vehicle Use Models This research considered the difference between weekday and weekend vehicle use behavior. In weekday, people are usually in busy commuting mode. Therefore, some guaranteed parking (especially at work) would possibly motivate their car use; or, at least discourage the use of other modes, such as public transportation.
In weekend, people are usually at the leisure mode with no hurry. Some weekend-drivers would have to use their cars to reach certain places (e.g., for entertainment or shopping) without considering much about the parking availability, as long as they feel “starting the engine” is convenient.
PARKING POLICY (as a transportation management scheme) might only work for weekday commuters! Sustainable Transportation –: Global Perspective with Local Care California: California produces roughly 1.4 percent of the world's, and 6.2 percent of the total U.S., greenhouse gases.
Requirements of Assembly Bill 32: Statewide GHG emissions cap for 2020, based on 1990 emissions by January 1, 2008.
Adopt a plan by January 1, 2009 indicating how emission reductions will be achieved from significant GHG sources via regulations, market mechanisms and other actions Shanghai: Taipei: GPS-Dispatched Taxi System: to add "functional utility"!
Taipei (population = 2.6 million with # of Taxi = 30,000) vs. Shanghai (population = 17.4 million; # of Taxi = 50,000)
NOT hard to get taxi in Shanghai anymore?
Concept of TOD, Hub ~ People are "UNIQUE" !!! Publications related to this Research: PAPER SUBMITTED (IN REVIEW, 2010) JOURNAL PUBLICATIONS CONFERENCE PROCEEDINGS & PRESENTATIONS Ni, J., Deakin, E., Cherry, C., Conducting Transportation Survey Research in China: Review of Experiences and Best Practices, Journal of Transport Policy
Ni, J., Sperling, D., Turrentine T., Kurani K., Ma, J., Vehicle Purchase Behavior in China: Case Study of Shanghai, Special issue of TRANSPORTATION Journal dedicated to “The Motorization of Asia: Implications for the Future” (SCI journal) Ni, J., Kurani, K., Sperling, D., Motorization in China: Case Study of Shanghai, Transportation Research Record: Journal of the Transportation Research Board, 2010 (accepted, SCI journal)
Ni, J., Ma, J., Vehicle Purchase Behavior Research in Shanghai (), Automobile & Parts Technology, No.1, OAP, pp. 42 – 43, [DOI] ISSN:1006-0162.0.2007-02-016, 2007 (published, in Chinese) Ni, J., Kurani, K., Sperling, D., Motorization in China: Case Study of Shanghai, Presentation of Transportation Research Board Annual Meeting, Washington D.C., 2010
Ni, J., Kurani, K., Sperling, D., Fu, B., Chen X.H., How To Do Survey Research in China? – Case Study of Shanghai, Proceedings of 7th Asia-Pacific Transportation Development Conference & 20th ICTPA Annual Meeting, Los Angeles, May 27th, 2007
Ni, J., Vehicle Purchase Behavior in China, Proceedings of Tong Zhou Transportation Conference, Tongji University, Shanghai, October 30th, 2006
Ni, J., Vehicle Purchase Behavior in China – Case Study of Shanghai, Presentation of 6th Asia-Pacific Transportation Development Conference & 19th ICTPA Annual Meeting, Hong Kong, June 1st, 2006
Ni, J., Conducting Survey Research in China: Case Study of Vehicle Purchase Behavior in Shanghai, Proceedings of 12th Annual University of California Transportation Center (UCTC) Student Conference, University of California, Berkeley, February 10th, 2006 Finally.. My Thank List ~ USA ASIA UC-Davis: Prof. Dan Sperling, Dr. Ken Kurani, Prof. Pat Mokhtarian, Dr. Tom Turrentine, Prof. Joan Ogden, Dr. Jonathan Weinert
UC Berkeley: Prof. Elizabeth Deakin Department of Transportation, Tongji University: Prof. Chen XiaoHong
Automotive Marketing and Management Institute, Tongji University: Prof. Ma Jun, Prof. Jerry Chen
Ford, China: Mr. Cheng Mei-Wei (President and CEO)
National Taiwan University Discrete Choice Model: Vehicle Purchase and Use FUTURE RESEARCH Segmented Model  Location-based Segmentation (home, work, etc.)
 Comparison between Models
 More Flexibility in Utility Functions (coefficients, # of variables, etc.) More "Macro" Consideration Any Questions? Let's Start with 1 Simple Question... ~ How did you come to this seminar? HA! That's Vehicle Use Behavior! Is it actually that simple? Individual Desire, if not well managed,
could collectively cause impacts/disasters. Is this what we want? Prof. Becky Loo, U. of Hong Kong
Dr. Jason Ni, THI Consultants Something beyond Textbook ~ your bring-home message... Small Piece vs. BIG Picture
Sustability and Equity
Passion and Belief From my Advisors... It also applies to HKU.. and my best wish to all of you. Don't forget.. the LAST Lecture.. Technological Development and Transport.... see you next week ~ December 21.. BUILT ENVIRONMENT: 3Ds & 5Ds ‘‘3D’’ model – Density, Diversity, and Design – first advanced by Cervero and Kockelman (1997)
2 additional ‘‘Ds’’ are also commonly considered: Distance to Transit and Destination Accessibility (Cervero, 2009) A study of mode choice in Montgomery County, Maryland (Cervero, 2002) reveals that development intensities and mixtures of land use significantly influence decisions to drive-alone, share a ride, or patronize transit.
Another study about the effect of transit-oriented development (TOD) using the heavy rail systems in New York City and Hong Kong (Loo et al., 2010): the results show that a combination of variables in different dimensions, including land use, station characteristics, socio-economic and demographic characteristics and inter-modal competition were important in accounting for the variability of rail transit ridership.
Station characteristics appeared to be the most important dimension in affecting average weekday railway patronage.
Interestingly, car ownership is both significant and positively associated with railway patronage --- higher car ownership may be associated with more pick-ups, drop-offs and park-and-ride activities to the transit stations for the longer transit trip legs. LOCATION/PLACE MATTERS ! Among all variables related to the built environment, “owning (car) parking space at work” negatively influenced the weekday use of public transportation, bicycle, motorized two-wheeler and motorcycle with similar magnitude.
This result is not surprising but could be interesting to compare with previous research related to TOD (Loo et al., 2010), which indicated that the car ownership is significant and positively associated with railway patronage.
In that sense, the ownership but restricted car use (e.g., limited parking), at least at the working place, would positively affect the use of public transportation. “Distance from Home to Subway” is the only significant (for motorized two-wheeler, motorcycle) variable related to the built environment.
Its positive sign suggests that the further people’s home are away from the main subway line; they are more likely to use motorized two-wheelers or motorcycles on weekend. 6. FUTURE RESEARCH Time Series...