Institute of Transportation, MOTC

According to Wikipedia, big data is a broad term for data sets that are so large that databases cannot store, compute, and/or process the data in a short amount of time and convert the data into interpretable information.

Valuable information may be hidden within big data. Such information includes key concealed or undiscovered information, such as unknown correlations, hidden patterns, and market trends. Therefore, not only have various fields placed greater focus on probing and analyzing big data in recent years, but in Europe and America they have applied their big data analysis results into governmental policies.

Computing technologies and structured data have also been the key thresholds of data analysis. However, following IBM’s introduction of the concept of big data in 2007, technology and information have developed to a point where big data can now be processed. Moreover, the progressive launch of innovative technologies, such as cloud computing and computing architecture, and the gradual prevalence of MEMS have resolved major obstructions in technology and information. The recent marketing, production, R&D, and design of technology giants, such as Google and Apple, are perfect examples of the application of big data.

The mass accumulation of transportation data most likely originated in the public transportation sector, such as Taiwan Railways Administration’s launch of telephone bookings on 12 March 1979 and Internet bookings on 15 April 1997; the activation of the MRT system in Taipei in June 2002; and the activation of the THSR on 15 January 2007. These three railway systems employ the automated procedures based on information and communication technologies (ICTs) for ticket sales and verification, which simultaneously began accumulating valuable data.

Moreover, electronic transportation tickets (formerly known as Taiwan Electronic tickets Payment; now known as electronic tickets; regulated by the FSC) have been able to be used to pay for MRT fares since 2002. The subsequent addition of Taipei City bus fares, highway passenger bus fares, parking fees, and micro-payments to the list of compatibilities have made the data collection and market prospects of electronic tickets limitless. Moreover, the interdisciplinary payment capability of electronic tickets has also enhanced the value of the data collected, which have made electronic tickets more consistent with the concept of big data, by which various data are meta-analyzed.

Small vehicle data mostly collected from vehicle detectors (VD) and Automatic Vehicle Identification Systems (AVI). However, the information quality could not be ensured due to the excessive maintenance costs and instability of these facilities. The activation of the eTag system on 10 February 2006 exponentially improved the data collection on highways, enabling the collection of more detailed data. The eTag database immediately became a popular target for the collection and analysis of data.

GPS chip modules were incorporated into mobile devices for the first time in 2003, increasing the possibility for users to pinpoint their locations. This became an instant popular topic of discussion. The launch of Google Maps and various traffic-reporting services, which use mobile phones to estimate traffic conditions, became conjecture that was taken as gospel by various sectors. Later, relevant specifications and details announced in developers’ conferences for Android and iOS, news concerning the transmission of data without the approval of the user, and various communications verified through laboratory testing suggested that collecting data using mobile devices to launch relevant services is not a prediction; rather, it is an operational model.

With the exception of mobile devices, most data collection and databases are typically based on frameworks that have specific functions when establishing and accumulating a foundation for data. For example, the THSR, MRT, and TRA systems are aimed at ticket booking, sales, and verification; electronic passes are aimed at convenient payments; and the collection of data by highway authorities is aimed at monitoring traffic and charging for the use of roads. However, this does not denote that system operators for mobile devices (regardless of device manufacturers or telecommunication vendors) are without goals. They clearly understand the importance and value of data. Instead of only collecting relevant data once goals are identified, they start to collect data during this process. The value of the data is highlighted with the decline in unit storage costs and computing costs.

The expansive and systematic collection of traffic data is considered to have begun in the late 90s. However, numerous technological and cost restrictions existed during that time, causing the data retained during that time to be incomplete. Even today, we often neglect data we consider to be irrelevant. Apace with the advancements in technology, we should immediately change our perspectives and past habits. The retention of complete data will be a key factor for future development.

Current big data research is based on institutes or units that possess large amounts of data. Due to technological developments, relevant technologies are applied for research. In the transportation sector, transportation departments use their collected data to analyze supervisory operations and facilitate decision-making. For example, data collected from the eTag system on highways are currently used to analyze the traffic conditions on highways, highlight problems, simulate measures, and facilitate decision-making. In the public transportation sector, data collected from swiping electronic passes are used to analyze network efficiency and passenger departure and destinations, thereby identifying appropriate bus route networks and formulating traffic control strategies.

Another value of big data is interdisciplinary analysis and research. Therefore, numerous private organizations, academic units, or even independent research teams are expectant on government units that possess big data. In response to the trend of data sharing, The Freedom of Government Information Law was passed in 2005, regulating the disclosure of information. However, to address the issue of personal information disclosure due to the growing requirements for in-depth and customized technological data, the Personal Information Protection Act was passed in 2010 to accelerate the establishment of a personal information protection system.

For the future application of big data in the transportation sector, we anticipate that “personal information protection” shall become a core foundation once relevant issues are clarified. We believe that big data will change extant operating models in transportation planning, management, safety, and R&D. We will no longer need to predict population conditions to create detailed behavioral models. We will be able to simulate strategies on a large scale and monitor performance and outcomes in real-time. Big data will be the primary driving force for change.

Recent popular topics in various sectors, such as Self-Driving Car and the Internet of Things, are all associated with data. The technology industry is currently endeavoring to research and provide countless solutions for the household, creating intelligent products by combining semiconductors with extant household appliances and devices. These products are able to automatically or semi-automatically provide feedback based on user habits. This is achieved through a back-end database. Past data are processed using an appropriate algorithm or model to provide users with feedback. Simultaneously, the “intelligence” of products can be elevated through machine learning, whereby they constantly revise and adjust their processing ability based on user responses. In the future, the business models of various industries may be altered, and data research would have to be incorporated into extant industrialized processes of production, marketing, and customer management, and eventually the overall industrial process. For example, the German government has invested in Industry 4.0 and the US citizen Brett King has proposed Bank 3.0. The integration of Self-Driving Car technology, roadside infrastructure, and ICT technologies in future will inevitably guide transportation into the next era.

A sound foundation must be established in order for new-era reform to be effortless. In response to technological trends and the reform of data application, the Ministry of Transportation and Communication (MOTC) commissioned the Institute of Transportation and the Census and Statistics Department (C&SD) to jointly arrange the “Application of Big Data Analysis in Transportation Management and Services Seminar” (hereafter referred to as “the Seminar”) in the Intelligent Technology Utilization and Transportation Management Optimization Seminar on 3 February 2015. The Seminar aimed to discuss the development trends of big data and share the successful implementation of big data analysis in transportation. Numerous affiliates, city and county governments, academic units, and consultation firms were invited to participate.

The Seminar was held on 24 July 2015. The heads of the transportation departments in the six municipalities of Taiwan were invited to attend. Viewing the concept implementation and application of big data from a practical perspective, the Seminar explored and discussed the importance, potential problems, cooperation opportunities, and policies concerning the future development of big data in Taiwan. Six core items were conceived in the Seminar, which will serve as a basis for the future development of big data in the MOTC, its affiliates, and the six municipalities. The Director of the MOTC further appointed the Institute, C&SD, and the Information Management Center, MOTC to participate in the future establishment of a big data consultation group, which will collaborate with the six cities to address and resolve major transportation issues through big data.

The big data of public transportation should be the primary focus. Practical traffic management requirements should be integrated and car/motorcycle information analysis results should be visualized to facilitate their application in resource allocation and monitoring activities for public transportation by the government, thereby improving the management models of passenger transportation operators. Travel monitoring data should be the secondary focus. The traffic trends in transportation corridors should be analyzed to enhance public transportation capacity and alternative route information services. In addition, data-based thinking should be adopted to develop a solution for bottlenecks in central and local road networks.

Big data visualization and geographical information analysis technologies should be used to diversely integrate transportation safety databases based on micro- and macro-transportation scientific perspectives, thereby effectively reducing accident mortality, reinforcing transportation safety information systems, and formulating preventive strategies.

Accumulating and sharing data is a rapid means to form big data for transportation. Complete and detailed big data are a basis for application and research development. We anticipate actuating the inter-governmental sharing of big data, enabling the MOTC to uniformly view problems, formulate information regulations and exchange standards, and establish an information exchange platform. Through this platform, local governments can propose their requirements for information from the central authority.

Big data volumes, structures, cross-applications, and other information techniques as well as various professional fields should be inter-exchanged to inspire more innovative application possibilities. Integrating transportation, disaster prevention, and internal affairs databases, and promoting a data exchange platform are essential for developing a smart city.

Establishing analysis environments for big data applications and cultivating analysts for information science are essential for Taiwan to keep up with the trend of big data analysis. Central and local authorities should establish a cooperation mechanism for information science teams, enabling the cooperation of big data holders, information scientists, and transportation experts, so that they may explore the application value of big data, accumulate technological practice and experience, and jointly create intelligent transportation for Taiwan.

Central and local governments should collaborate in establishing a transportation information science consultation services organization. This organization shall be responsible for the orientation of future transportation big data. The MOTC should formulate prospective plans based on the individual development features of the six municipalities. These plans should then be regularly discussed by teams established by the central and local governments, enabling business units to seamlessly operate, and big data analysis to effectively aid the leaders of the central and local authorities in their transportation management decisions.