Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
geographic information system, it is necessary to survey the contents of trips through ancillary questionnaire sur- veys and other means. The probe person survey system is one that, on the basis of the advantages and problems to be solved for such existing studies, aims to achieve the goals of (a) ensuring accurate travel records through spaceâtime position determination functions of high accuracy, (b) reducing recording omissions through timely travel behavior recording functions, and (c) ensuring improved efficiency of data coding by the investigator and improv- ing the sense of participation in surveys of subjects through a system that emphasizes real- timeness and interactiveness. A survey system of high timeliness is constructed that allows travel records to be input anywhere through the use of both mobile phones and the web as input media. This process allows subjects to record travel in a timely manner while they have a clear memory and to check and correct the travel records. Moreover, in a long- term survey in which mobile phones with position determina- tion functions are used to establish spaceâtime position, it is necessary to obtain data of high accuracy while reducing the burden on subjects. To this end, it is impor- tant to have a GUI that allows subjects to check visually and to edit both the data they recorded through the information input and the travel records obtained through mobile communication systems, which comple- ment each other. An example of such a survey is a suc- cessive diary survey of about 3 months. A survey has been proposed in which the contents of individualsâ behaviors are identified by using accelerom- eters, barometers, and sound sensors rather than imple- menting questionnaire surveys as a complement to a behavior survey that uses mobile phones and a GPS. In such a survey, transport mode is identified by the magni- tude of acceleration; the floor of a subject is identified by atmospheric pressure; and the behavior content is identi- fied by variability in acceleration and sound. It is aimed at automatically recognizing the behavior content by preparing sensor signal sequence data corresponding to behavior patterns, such as transport mode and activity content, as training data. Survey methods using multiple sensors can be said to allow a long- term behavior survey of high quality to be achieved by reducing the burden of input on subjects in the survey. APPLICATIONS OF SURVEY TECHNIQUES Wide Area: Probe Person Survey The most versatile survey technique of those described in the previous section is the probe person survey, in which subjects are asked to carry GPS mobile phones, which are in wide use in Japan. It is possible to record the travel path and departure and arrival times by using the phones as well as to correct input data and implement various ancillary surveys through a web diary in combination with them. Hato et al. (2006) implemented a successive day- to- day survey for several months on the same indi- viduals as a three- wave panel survey (Figure 2). The survey was composed of agent- type application software for GPS mobile phones, a web diary system, and a position data management server. Test subjects were asked to carry mobile GPS phones and to press the start or stop button at the time of departure or arrival, respectively, through the agent- type application soft- ware. Doing so determined originâdestination data and departure and arrival times for the trip. Position data were recorded every 20 s while the subject was traveling, and position data of about 5 to 10-m accuracy was transferred to the data management server if outdoors. By a series of such processes, path data were automati- cally determined. The results of trips were instantly reflected in a blog- type web diary system that could be viewed, with corrections being possible for any omis- sions related to the purpose of the trip and means of transportation. Furthermore, correcting the web diary was also possible from mobile phones by using their Internet function. Long- term recording became possible in a diary- like sense for test subjects because they could verify what was recorded about their trips via a map on the Internet. Through input via mobile phones, the agent application software transmitted transportation data. A mechanism to prevent omissions was made possible by using the transmitted traffic data as an incentive. Furthermore, this survey system proved to be stable with existing mobile phone systems and capable of supporting large- scale studies. 190 INNOVATIONS IN TRAVEL DEMAND MODELING, VOLUME 2 FIGURE 2 Outline of probe person systems.