Asakura Laboratory was established in 2011 when Professor Yasuo Asakura arrived at Tokyo Institute of Technology from Kobe University.

The main theme of the laboratory is traffic engineering, with the emphasis on observations, surveys, analysis and evaluation of human and vehicle movements on the transportation networks. The traffic behaviour that we usually and casually exhibit is indispensable in carrying out a wide range of activities such as commuting, shopping or leisure. To ease such traffic behaviour, it is important that transportation facilities, such as roads and railways are operated “safely”, “efficiently” and “attractively” in a “sustainable” way. Our laboratory conducts the research activities with the aim of achieving such traffic management. We investigate traffic dynamics not only under normal, but also under abnormal conditions, such as disasters, with the help of data obtained from the sensors, such as traffic counters installed on expressways and behaviour surveys which include probe person survey by mobile phone with Global Navigation Satellite System (GNSS) function. Recent topics are listed below. Please also refer to the PUBLICATION.

Ridesharing and autonomous vehicles

When we discuss the next generation transportation systems, we have to consider the developments related to information and communication technologies, as well as the changes in our life styles and social structures. There are two major forces driving the future transportation systems. One is the Autonomous Driving, and the other is the Sharing Economy including car-sharing and ride-sharing. By providing timely and convenient transportation, ride-sharing services are transforming urban mobility and show many positive impacts with respect to traffic congestion, pollution mitigation, energy consumption, etc. Social optimum strategies should be studied considering both operators and user’s behaviour for new transport services. In considering characteristics of autonomous vehicles, matching of multiple users, optimum vehicle operation and pricing options, we develop optimisation models and analyse the ride-share transport system from various viewpoints. In addition, field experiment on smart sharing will be conducted.

Traffic flow theory and network traffic flow analysis

Traffic flow analysis and related theoretical basis is an essential piece of knowledge for engineering a network. Without exaggeration, we can say that all the traffic of humans and things in modern society are related to these matters. We investigate various problems on different scales, such as estimation of the overall conditions based on partial information / data, discovering important events from a large amount of probe vehicle data, traffic optimisation in response to disasters, as well as analysing of network vulnerability based on aforementioned theories.

Selected publications:

• Takeshi NAGAE, Tomo FUJIHARA and Yasuo ASAKURA: Anti-seismic reinforcement strategy for an urban road network, Transportation Research Part A: Policy and Practice, Vol.46, No.5, pp.813-827, 2012
• Chong WEI, Yasuo ASAKURA and Takamasa IRYO: Formulating the within-day dynamic stochastic traffic assignment problem from a Bayesian perspective, Transportation Research Part B: Methodological, Vol.59, pp.45-57, 2014
• Toru SEO, Takahiko KUSAKABE and Yasuo ASAKURA: Estimation of flow and density using probe vehicles with spacing measurement equipment, Transportation Research Part C: Emerging Technologies, Vol.53, pp.134-150, 2015

Development of surveying method and statistical analysis framework

Understanding of people’s daily behaviour is indeed necessary in order to provide better planning and operations. As a means of acquiring behaviour data, unlike traditional questionnaire survey, we propose a method that can acquire a large amount of data semi-automatically at low cost. By using the GNSS function built in mobile phones, daily behaviour can be determined automatically with the help of statistical methods. Although the amount of data in the world is growing rapidly day-by-day, the knowledge which can be obtained by vaguely looking at this data is limited. In order to fully utilize the data, statistical methods, such as Bayesian estimation method is applied to mine the latent insights from the data.