Long-term impact of the Shinkansen on rail and air demand: analysis with data from Northeast Japan

This paper analyses how the high-speed rail construction in Northeast Japan (Tohoku) has affected total demand and interregional travel patterns. We use annual interregional passenger data from 1989 to 2012 and apply regression analysis with the demand between Tokyo and the Tohoku prefectures as the dependent variable. We distinguish particularly between the ‘Full-’ and the ‘Mini-’ Shinkansen, where the latter are branch services running with reduced speed. We find that the ‘Full-Shinkansen’ quickly increases rail and total public transport trips and generates additional rail demand year on year. The ‘Mini-Shinkansen’ impacts are less pronounced. Furthermore, our analysis shows that the Shinkansen has shifted some demand from air to rail once it started operation and increased rail share gradually. We therefore suggest that predictions of demand impacts should carefully distinguish immediate from gradual impacts. We also discuss differences in regional demand in that not all prefectures have gained equally from Shinkansen construction.     

Using Electronic Toll Collection Data to Understand Traffic Demand

In this study, we explored the potential of using electronic toll collection (ETC)-derived data that are a part of intelligent transport systems (ITS). Dynamic origin–destination (OD) traffic volumes were estimated using ETC data on the Hanshin Expressway. A dynamic OD estimation model that was suggested in a previous study was used, and abundant ETC data were input to improve the estimation accuracy. The results of OD estimation were analyzed to understand traffic demand and its variation. External factors were clarified that have an influence on variances in the OD flows, and statistical analysis methods for the variations were proposed depending on the factors. Moreover, the improvements in traffic simulation accuracy and performance as a result of using ETC data as input variables in the simulation models were discussed. According to the results of this study, ETC data have potential to assist in understaningd traffic demand and its variation, and the results can be applied to network management.