Transportation Deployment Casebook/2014/Japanese Bullet Train: Shinkansen

INTRODUCTION
Launching high speed passenger and freight trains have been at the hub of attention by transportation planners and policymakers, in recent decades. Before the development of airlines and motorized vehicles, transit railway was known as the first mode of rapid land transportation. One of the pivotal characteristics of the railways is speed which has triggered transportation planners constantly struggle to achieve higher speeds, and consequently diminish travel times. Hence, a considerable shift has happened in the speed of the railway systems around the world during the last few decades. For instance, Stephenson's Rocket which worked with steam power was the fastest train in 1829 running at speed of 46 km/h. In 1848, the Antelope train surpassed the Stephenson's Rocket with speed of 97 km/h. They were the fastest rail vehicles at the time in the world. In recent years, however, the world has witnessed incredible train speed records. The LGV Est and LGV Atlantique trains in France, by reaching 574.8 km/h broke the world speed record in April 2007.

The milestone of the railway systems in terms of speed might be the birth of the high-speed rail. Typically, a rail system which be able to surpass the speed of 200 km/h is defined as a high-speed rail system. Thanks to specialized rolling stock technology and dedicated tracks, high-speed rail operates considerably faster than conventional railways. In October 1964, the first high-speed rail system was launched in Japan to connect Tokyo to Osaka with trains running at speeds of 210 km/h. Although the success of the Japanese high-speed rail system, bullet train, inspired other countries to operate the same system, the spread of the system was relatively slowly around the world. For instance, the second high-speed train was inaugurated in 1981 to connect Paris to Lyons with speeds up to 270 km/h.

GEOGRAPHY OF JAPAN
Japan spans around 378 thousand square kilometers, has a population of more than 126 million, and is the 36th most densely populated country in the world. The country is located in the Pacific Ocean and is adjacent to China, South and North Korea, Russia, and Sea of Japan. The location characteristics of the Japan caused the country is known as the "Land of the Rising Sun." Japan is a long and slender archipelago containing more than 6,800 islands, four of which namely, Shikoku, Kyushu, Hokkaido, and Honshu, encompass around 97 percent of the Japan's land area. Tokyo, the de facto capital of Japan, is located in Honshu island in which with over 30 million residents, is the largest metropolitan area in the world. Japan is a developed country and not only is the third-largest economy in the word, but it is also the world's fourth-largest both exporter and importer. The population of Japan is concentrated in the major coastal cities namely, Tokyo, Nagoya, and Osaka. The Nagoya is located between Tokyo and Osaka which almost 523 kilometer are far from each other. A quarter of the Japan's population also live in the Tokyo. The population and characteristics of the largest cities in Japan is outlined in Table 1.

TABLE 1 Characteristics of the largest cities in Japan

The above-mentioned characteristics of Japan, persuaded the government to lunch the first high-speed train between Tokyo and Osaka.

MODE DESCRIPTION
The idea of the bullet train dates back to 1938 in which the first high-speed rail running at almost 201 km/h to connect Tokyo and Shimonoseki in Japan. Although the construction was started in 1938, the development of the network was blocked for the sake of the World War II. In 1957, the first committee of the Shinkansen was organized to develop the bullet train between Tokyo and Osaka. The Japanese National Railway began construction of the line for a total distance of 551 kilometers. The first link of the bullet train which is also known as Shinkansen came into service in October 1964 concurrent with Summer Olympics in Tokyo to disclose the first high-speed rail technology to the world. The newly inaugurated network dramatically diminished the travel time between two cities from 7 hours to less than 3 hours. Shinkansen also benefits from modern engineering features and facilities including, the monitoring system for drivers, passenger guidance information system, ATC speed control system, water-sprinkler snow-melting system, and train protection system from an earthquake. A Shinkansen train typically consists of 16 cars with a 396 meters total length with the capacity of around 1,300 passengers.

Birthing Phase
Tokaido rail line dates back to 1930, when an express train connected Tokyo and Osaka with an average of eight hours and 20 minutes travel time. Given the low capacity of the railway line, the government decided to replace the conventional lines with electric lines in 1950. Since after World War II, Japan was remarkably dependent on oil and other natural resources imports, the idea of electrifying the railway was warmly welcomed by the government. In 1958, Tokyo and Osaka was connected by a conventional rail line with a narrow gauge system. The conventional train, running at 91 km/h average speed, took around 7 hours between Tokyo to Osaka. Population of the Japan was also about 98 million, when the Shinkansen bullet train connected Tokyo to Osaka in 1964.

Growing Phase: Privatization
The political pressures triggered that debt surpassed $200 billion by 1987 leading to a financial crisis. It might be a spark for privatization. Japan National Railway began the process of privatization in 1987, just seven years after privatization of two other gigantic public entities namely, Monopoly Public Corporation and Nippon Telephone and Telegraph in Japan. The political pressures and demands for high-speed transit stations rooted in expected economic gains. Same as the inauguration of high-speed rail, the privatization of the Japan National Railway is known as the first comprehensive reorganization of a national railway around the world. The privatization of the Japan National Railway was accomplished step by step. Mizutani and Nakamura, in 1997, discussed about the main features of the privatization of the Japan National Railway. They divided the features into six distinguished categories that include: 1) regional subdivision, 2) passenger-freight distinction, 3) operation and infrastructure integration, 4) lump-sum subsidies for low-density Japan Railways, 5) the establishment of an intermediary institution, and 6) allowance of non-rail services.

Given some management and political influence issues, it was decided to split the Japan National Railway company into six regional passenger railway companies pursuant to the geographical demand. To gain the freight market share from trucking industry, also, the Japan Railway Freight was separated from the passenger Japan Railway. However, the Japan Railway Freight borrowed tracks from the passenger Japan Railway to decline the cost of infrastructure and avoid an excessive financial burden at the beginning. Studies show that the general performance of the Japan Railways, particularly efficiency and labor productivity, has been significantly improve after privatization. For instance, between 1987 and 1992, the average annual growth rate of the Japan Railways and private sectors was almost 11.4 and -0.7 percent, respectively. In 2000, Sumita also indicated that privatization triggered rising the rail fare.

Japan Railways benefited from the privatization and in less than five years the operating profits dramatically increased to $7.1 billion dollar. This trend continued over the years and enabled the Japan Railway Central, Japan Railway East, and Japan Railway West companies to purchase the rail lines from the government. Although this act begot some financial difficulties for the companies, the company revenues significantly increased in the long term. For instance, annual operating revenues for Japan Railway Central and Japan Railway East was reported, respectively, around 10.0 and 16.9 billions of dollars in 1998 which were far greater than the government prediction. Furthermore, privatization had a positive effect on the competition between rail and airplane to the extent that Japan Railways has continuously introduced new types of cars with new facilities including, silent rooms for passengers who want to take a rest, compartment rooms for meetings, and electrical outlets for personal computers in order to win the competition.

The Network
After the privatization, the Japan National Railway company had been totally broken up to six passenger an one freight railway companies. Further, to handle the debt and assets in long term, the government decided to formed the Japanese National Railway Settlement Corporation. The Japan Railway East and the Japan Railway West were at an advantage of the highest number of passengers in 1998. Annual operating revenue statistics also show that the companies profited from combining the infrastructure ownership with the operations.

The Shinkansen high-speed railway includes seven lines namely, Tokaido, Sanyo, Joetsu, Tohoku, Yamagata, Akita, and Hokuriku with the total 2,167 kilometers length of the lines. Tokaido Shinkansen line with 552 kilometers length was completed in 1964 as the first Shinkansen high-speed railway, and currently is operated by Japan Railway central company. In 1975, the second line of the Shinkansen i.e. Sabyo line was operated with 553 kilometer length as the largest line. This line is monitored by Japan Railway West company. The Japan Railway East company operates the four other companies with the almost 829 kilometers total length. The Tohoku and Yamagata lines are launched in 1991 and 1992, respectively; while, the both Akita and Hokuriku lines are inaugurated in 1997.

To attract more passengers, the Shinkansen decreased the average time schedule delays to only 24 seconds per train. Moreover, the system provides numerous options such as subway, monorail, express train, and automatic guideway transit to transfer passengers to and from major stations. In the Tokyo station, for instance, around 7.3 million passengers are carried by an extensive network of subways totaling 238 kilometers with only two minutes headway during the rush hours.

Construction Costs
Table 2 shows the Shinkansen construction costs. As per Table 2, the construction costs of the Tohoku and Joetsu lines are significantly higher than other lines. In 1992 Taniguchi conducted a research to investigate the construction costs of the Shinkansen railways. This study showed that Infrastructural developments such as cuttings, banks, bridges, and tunnels have the highest costs to construct the high-speed rails in Japan. For instance, 58 percent of the total cost for Sanyo line is dedicated to infrastructural developments, particularly, building bridges and tunnels. The proportion of each expense and infrastructural cost of the Sanyo line, between Shinosaka and Okayama, was estimated by previous studies. It is shown that more than 80 percent of the infrastructural costs is allocated for constructing the railroad viaduct and the tunnels. The Shinkansen system is more expensive than other high-speed rail systems such as Train à Grande Vitesse in France, Neubaustrecken in Germany, and Alta Velocidad Española in Spain. The reasons might be rooted in land price and unique topography of the Japan which burdens constructional costs of building tunnels and bridges for straight railways.

TABLE 2 Shinkansen Construction Costs

Advantage and Disadvantage
This part of the study attempts to provide a brief summary of the Shinkansen impacts. The most common impacts of the bullet train might be divided in four main categories: 1) Mobility, 2) Environmental, 3) Economic costs, and 4) Spatial and regional. Studies showed that operating a high-speed transit system not just attracts new passengers, but other service users significantly shift to this transportation mode. In first years of lunching the high-speed train in France, for instance, share of rail transportation dramatically increased from 40 to 72 percent; while, modal share of air and motorized vehicle traffic diminished around 24 and 8 percent, respectively. A wide range of studies have been also conducted to investigate the environmental impact of the high-speed railways. Overall, conducted studies indicates that high-speed train systems are more environmentally efficient than its competitor systems namely, aircraft, road, and conventional rail services. However, this technology is continuously criticized for environmental damages including, visual disruption, high consumption of electric energy, noise nuisance, and producing sulphur dioxide (SO 2 ) and nitrogen oxides (NOx) pollutions. Many studies, further, have been investigated the construction costs of the high-speed train systems around the word. The final results show that infrastructural costs of the high-speed railways have a positive correlation with number of operated bridges and tunnels, cost of land and labor, and passing through the city center.

Market Niche
Nagashima in Niche Marketing: 60 success stories, introduced four strategies for creating demand that leads to new products. He mentioned that the Shinkansen succeed to quickly get niche market and attract the attentions. Railway engineers came from all the countries around the world and admiringly called the Shinkansen "a magnificent innovation." While, the Japan Railway officials responded with “No, it’s not an innovation but the result of many small improvements added one by one.”

MARKET SHARE AND COMPETITION
The Shinkansen achieved successes in market just in first years and consequently effected socio-demographic characteristics of the cities. First, the number of passengers-kilometers quadrupled in just less than ten years. The Shinkansen triggered travel time savings around 400 million hours annually. Studies found a positive correlation between development of the high-speed transit railways and population growth. Hirota in 1985, for instance, indicated that population growth rates in cities with high-speed transit railways and those by passed by the service is almost 1.6 and 1 percent, respectively. This study further showed that in cities with a high-speed transit station, employment growth in wholesaling, retail, industrial, and construction is around 16-34 percent more than cities with no high-speed transit station. A study which was conducted in 1997, on the other hand, contended that high-speed transit lines had not a significant effect on population growth, and the population growth for selected Japanese cities in previous studies was expected independent of the high-speed transit lines. In the economic growth side, all the previous studies indicated that the tourism industry, service industry, and employment of Tokyo and Osaka significantly increased following the high-speed transit inauguration. The employment in Osaka, Kobe, and Kyoto, for instance, increased by 35 percent between 1955 and 1970.

Japan has 1.2 million kilometers paved road. However, given the insufficient capacity, toll highways, and high fuel prices, the private vehicle travel mode is not popular among Japanese for inter-regional transportation. Hence, the Shinkansen easily surpassed its competitor mode. On the other hand, around 175 airports including Haneda Airport, Narita International Airport, Kansai International Airport, and Chūbu Centrair International Airport are located in Japan. Haneda Airport is both the largest domestic and the second-busiest airport in Asia. The Nagoya Port is also the largest and busiest port in the country which organizes around 10 percent of trade values in Japan. Therefore, a broad spectrum of studies have attempted to investigate the competition between Shinkansen and Airline services. In terms of energy consumption, Shinkansen by consuming 30 Watt-hour electricity per passenger in a kilometer is six times as efficient as air services. Studies also show that high-speed rail is faster than air or conventional rail modes of travel in medium distances. However, there is not a significant advantage for high-speed rail in very short or very long journeys. Air and conventional rail mode of travel, respectively, enable journeys over long and short distances to be made quickly. A study attempted to find boundary distances where the high-speed rail performs better than aircraft and conventional railway. The results show that high-speed rail has a little advantage from conventional rail in less than 150 kilometers distances; while, for journeys of between 150 and 400 kilometers, the high-speed rail performs better than both air travel and conventional rail modes of travel. In distances more than 800 kilometers, finally, air travel remarkably faster than high-speed transit rails. A dedicated study to analyze the Shinkansen market share among other available transportation modes, in 2014, found a similar results in intra-Japan markets. The final results of the study show that for journeys of less than 500 kilometers the share of air and rail travel is around 5 and 53 percent, respectively. For longer distances, on the other hand, the inter-city passenger market share is completely different from short distances. For journeys of more than 1,000 kilometers, share of the rail transit is around 5 percent, while the passenger market share for air travel is almost 93 percent.

Overall, the results show that the most people use Shinkansen instead of airline services between Osaka and Tokyo. The reasons might be rooted in Cheaper transportation fare, better access to stations, Reliability of schedule, and Safety operation of the Tokaydo line.

INTRODUCTION
The main aim of the study is to investigate the life-cycle of the Shinkansen Bullet Train in Japan. In this vein, a single variable linear regression model along with a S-curve analysis were employed to identify the periods of birthing, growth, and maturity of the system. The rest of this part is dedicated to illustrate the method of the study and results of the final model in a concise fashion.

REGRESSION MODEL
The first application of regression models dates back to 1877, in which Galton investigated the height correlation between parents and their children. This method, then, was developed by Pearson, for statistical concepts. Although Galton applied regression analysis to emphasize the regress toward the average, the term of regression analysis focuses on modeling and examines the relationship among variables, in recent decades.

Regression models are vastly used in the fields of engineering, physics, economics, management, life science, biology, and social sciences for estimation and prediction. It can be said that regression analysis, is one of the most common method among statistical techniques. Equation 1 shows the correlation between endogenous and exogenous parameters in a multiple regression model. In this equation, Y i  is a dependent or response variable for observation i, X is explanatory or Predictor Variable, β is the vector of partial regression coefficients, and u i  defines as a stochastic disturbance term.

Y i =β 0 + β 1 X 1i + β 2 X 2i +…+ β k X ki + u i (1)

These models are named linear regression models because of being linear in parameters (i.e. β). In these models, β coefficients are unknown and estimated by different methods such as Ordinary Least Squares and Maximum Likelihood methods. In Ordinary Least Squares method, the best-fit model is a model in which the sum of squared of differences between observed Y and the estimated value from the model, is minimized. The method of maximum likelihood consists in estimating the unknown parameters in such a manner that the probability of observing the given Y is the high as possible.

DATA
This study attempts to find a simple correlation and the best fit between passenger-kilometers traveled and year variables. A single variable linear regression analysis, therefore, is adopted in this study, primarily because of its sound theoretical setting. Aggregate information (Table 3) was obtained from three main sources to explain share of passenger-kilometers traveled in a semi-log regression model. Given the Shinkansen transit has been operated by three main companies since the privatization time, the statistics were gathered in two steps. To obtain the passenger-kilometers traveled data, before privatization, the annual report of the Japan National Railway was used in the first stage. In the second stage, the required data was obtained from the fact sheets of the East Japan Railway, Central Japan Railway, and West Japan Railway companies which were formed after privatization.

TABLE 3 Explanatory Variables Used in the Model

MODEL AND RESULTS
The function of passenger-km traveled is given by Equation 2. In this equation, t is the observed year, t 0  is the inflection time, K is saturation status level, and b is a coefficient. To estimate the K and b parameters, a single variable linear regression was employed as per Equation 3. In this equation, X is the observed year and Y is given by Equation 4.


 * $$S(t) = \frac{K}{1 + e^{-b(t-to)}} $$ (2)

Y = bX + c (3)

Y = LN(Passenger-km/(K-Passenger-km)) (4)

To find the best fit of the model, a number of regression models, with various trials for K were applied. The statistical t-test was used to reject the null hypothesis that a specific variable does not affect share of passenger-kilometers traveled. Adjusted R-squared statistic was, also, used to assess the overall goodness-of-fit of each model. The adjusted R-squared value fluctuates between 0 and 1, and the greater the value, the more the model’s explanatory power is. The final model (Table 4) is estimated considering the year as an only independent variable, and pursuant to the saturation status level equals 88,400 million. The classical normal linear regression modeling approach that is implemented in this study is also based on some critical assumptions. The results are valid only if the model is correctly specified, residuals are normally distributed, residuals have a homoscedastic variance, and there is no autocorrelation in the residuals. The semi-log formulation was adopted to prevent a specification bias to some extents. Furthermore, the null hypothesis that the residuals are normally distributed was not rejected in the Kolmogorov-Smirnov test. As per the final results, life-cycle variables of the Shinkansen Bullet Train are reported in Table 5.

TABLE 4 The Final Regression Model

TABLE 5 The life-cycle of the Shinkansen Characteristics

Figure 1 shows the S-curve analysis for better understanding the life-cycle of the Shinkansen.



FIGURE 1 S-curve Analysis