Eiji Hato

Travel Mode Detection with Varying Smartphone Data Collection Frequencies

Smartphones are becoming increasingly popular day-by-day. Modern smartphones are more than just calling devices. They incorporate a number of high-end sensors that provide many new dimensions to smartphone experience. The use of smartphones, however, can be extended from the usual telecommunication field to applications in other specialized fields including transportation. Sensors embedded in the smartphones like GPS, accelerometer and gyroscope can collect data passively, which in turn can be processed to infer the travel mode of the smartphone user. This will solve most of the shortcomings associated with conventional travel survey methods including biased response, no response, erroneous time recording, etc. The current study uses the sensors’ data collected by smartphones to extract nine features for classification. Variables including data frequency, moving window size and proportion of data to be used for training, are dealt with to achieve better results. Random forest is used to classify the smartphone data among six modes. An overall accuracy of 99.96% is achieved, with no mode less than 99.8% for data collected at 10 Hz frequency. The accuracy is observed to decrease with decrease in data frequency, but at the same time the computation time also decreases.

Modeling the Cooperation Network Formation Process for Evacuation Systems Design in Disaster Areas with a Focus on Japanese Megadisasters

AbstractBy treating human relationships as networks, disaster response planners can capture the features of cooperative behaviors between residents, providing valuable insights for effective evacuation planning. In this paper, the writers outline residents’ evacuation behaviors during the 2011 Great East Japan earthquake and the resulting nuclear disaster and tsunami. An analysis of these evacuation behaviors shows the significance of cooperation between residents and of problems caused by two types of bias: normalcy and majority syncing. Next, the writers describe how residents’ cooperation behaviors accumulate to form cooperation networks and apply the fitness model to the formation of these networks. A case study of network formation in response to a mudslide disaster in the city of Niihama in 2004 is provided. The writers estimate the fitness parameters and analyze the network formation process and structure on the basis of the distribution of the fitness parameters and show the scale-free characteris...

Accounting for spatial correlation in tsunami evacuation destination choice: a case study of the Great East Japan Earthquake

This article analyzes the tsunami evacuation destination choice process, using as a case study the Great East Japan Earthquake of 2011. The contribution of this article is twofold. First, it sheds some light on the choice mechanism behind tsunami evacuation destination choice, an understudied aspect of the evacuation process. Second, and from a theoretical perspective, it addresses the issue of spatial correlation in discrete choice models. A spatially correlated logit model is estimated, where the allocation parameter is specified as a function of proximity and inter-zone altitude difference to capture more adequately unobserved similarities among alternatives in the specific context of tsunami evacuation.

A Suggestion of Path Enumeration Algorithm Based on Massive Behavioral Space Information

This paper on a path enumeration algorithm is from the proceedings of 14th international Conference on Urban Transport and the Environment in the 21st Century, which was held in Malta in 2008. The authors note that one of the difficulties of route choice modeling is to represent the process of generating a choice set explicitly in a network that includes infinite paths. Studies analyzing routes that a driver actually has used have not been conducted, so in most route choice models, the routes included in a choice set do not necessarily correspond to routes that a driver has actually used. However, the development of Probe Person Technology permits location data of the movable body with the use of GPS, so it is now possible to collect the data of routes a driver has actually used without burdening the driver with recordkeeping. In this paper, the authors discuss the choice set generation methods based on routes drivers actually used, and present case studies of route choice models that use these methods. They conclude with a brief discussion of other variables that must be considered, notably travel-time variances that influence the route choice behavior.