博士論文発表会のご案内 (Doctoral Thesis Final Presentation)
当研究室博士課程３年Yu Xiao君の博士論文発表会を下記のとおり執り行います．
ご感心ある方でご出席頂ける方は，福田(fukuda[at]plan.cv.titech.ac.jp)までご連絡いただけますと幸いです．
発表者：Yu Xiao
論文題目： Valuing Travel Time Reliability: Individual, System and Dynamic
Perspectives
主査：福田大輔准教授
副査（予定）：朝倉康夫教授，屋井鉄雄教授，室町泰徳准教授，花岡伸也准教授，文世一教授（京都大学大学院経済学研究科）
日時：平成28年1月5日(火) 午前10時45分-12時15分
場所：東京工業大学 緑が丘5号館（創造プロジェクト館）1F大会議室
“Valuing Travel Time Reliability: Individual, System and Dynamic Perspectives”
Yu Xiao
(Abstract)
Traffic congestion is a widespread social problem and needs to be settled by sophisticated transport management and investment. The technological advances in monitoring traffic conditions allow the stochastic features of travel time to be better captured and managed, which leads to potentially large benefits to users of transport system. The guideline of cost-benefit analysis worldwide needs to be modified into one that accounts for the benefits of emerging reliability-improving schemes.
This dissertation is dedicated to the theoretical framework in including travel time variability (unreliability) into cost-benefit analysis, with a particular focus on the monetary value attached to unit improvement of travel time reliability. This thesis can be divided into two parts: (1) understanding traveler’s decision when facing variable travel time and (2) modeling transport system with variable supply.
The first part, including Chapter 3, is particularly concerned with estimation of the cost of travel time variability. It analyzes how systematic perception errors in travel time distribution might bias the estimates and undermine the theoretical equivalence between the structural model and reduced-form model. Empirical estimation on these bias is carried out using stated preference data.
The second part, including Chapter 4 and Chapter 5, is concerned with the system (social) cost when travelers are constantly searching for lower travel cost while the transport system are constantly facing random shocks. Taking travel time variability as given, Chapter 4 uses a stylized departure-time equilibrium approach to study how system cost of a traffic bottleneck varies with travel time variability when congestion profile depends on traveler’s collective behavior. It discusses how the conventional definition of value of travel time variability can be modified and fitted in the existing framework of cost-benefit analysis for transport investment to better capture the effects of endogenized congestion. On contrary, Chapter 5 challenges the assumption of stable equilibrium by arguing that the system might not have a stable equilibrium in some case and travel time variability is also a phenomena of traveler’s day-to-day behavior adjustment. It uses simulation to investigate how much travelers’ day-to-day departure time adjustment contributes to the travel time variability and the time-average travel cost in a long run.
In summary, the two parallel parts deal with the valuation of travel time variability from different angles, contributing new insights on using reliability as an indicator for transport user’s benefit. I hope it could shed some lights for both researches and practitioners on the route to providing more reliable transport services.
