| 4. In which ways is the initiative creative and innovative?
|
|
1. Discussing project tasks between Big Data Team and Transportation Policy Department (March 2013).
The first big data-based program was the late-night bus service as it was deemed to have immediacy, especially with regard to working women and the handicapped. Big Data Team and Transportation Policy Department held many meetings through which both agreed that the routes for late-night bus service the department has been considering needed to be revised based on the actual numbers of the citizens who would benefit most from the routes once selected.
2. Signing an MOU with KT (April 2013).
Finding the big data useful for selecting the most pertinent routes was not easy. Knowing that the data showing the movements of citizens in late hours would be ideal, the two decided on the idea of using data created by people using their phones at or near bus stations, which led to taxi GPS data containing such information as when and where passengers got on and off and phone data as to where, when and how many people used their phones. To get such phone data, the City signed an MOU with KT, a leading telecommunication company, and received over 3 billion phone logs for free.
3. Forming cooperative ties with the central government on big data.
The City and the Ministry of Science, ICT, and Future Planning of the central government agreed on selection of late-night bus service routes as a trial task to promote use of big data and pushed forward the task in unison. Under the basic administration principle of “Government 3.0”, the City and the central government cooperated with each other on technical and strategic aspects related to how to use big data in selecting the routes.
4. Laying a big data share base unique to Seoul City (Sept. 2013).
Aiming to adopt the practice of big data analysis throughout its municipal organization by 2014, the City has constructed a big data share/use platform and plans to carry out follow-on tasks such as fusing social big data with location information and developing and analyzing new policy tasks.
|
|
| 5. Who implemented the initiative and what is the size of the population affected by this initiative?
|
|
1. Cooperation between IT and administration units.
In solving city problems through use of big data, cooperation between the IT department capable of processing and analyzing data and other departments in search of solutions to city problems of their own has proved to be the most important factor in the Big Data Project. If these departments had not shared among themselves the need to use big data and the willingness to work together towards common goals, the Project would have been impossible. Because of these, the IT and Transportation Administration departments were able to agree on the feasibility of using call trend big data and, after meeting online and off-line over 50 times, find the nine most pertinent routes.
2. Private partners.
Formulating policies that are effective in addressing city problems takes not only public data but also a great deal of data that are scattered around various private sectors. For example, in the late-night bus service project in which big data was used for the first time, data from KT played a crucial role. In addressing issues related to public welfare, economy, culture and other areas, the City has found out that forming cooperative and strategic ties with private businesses early on is very important.
3. Working with groups of experts in various areas.
The City also has realized early on that involvement of experts was necessary in program development, particularly for securing public confidence in the reliability of programs to be implemented. By forming an independent council comprising experts from many different fields for each program, the City was able to ensure objectivity in the decision making process and thus enhance public confidence in the programs it sought to implement. In the case of selecting locations for senior welfare centers, for example, a researcher from a reputable and independent social welfare research center participated and offered his expertise on how to use social welfare-related big data.
|
| 6. How was the strategy implemented and what resources were mobilized?
|
|
1. Funding: Minimizing costs through MOUs with private businesses.
In developing the late-night bus service, the City spent about KRW 20 million all together, most of which went into outsourcing for construction of a computerized system for the service; as for data and data analysis, the City signed an MOU with a telecommunication company and received them for free. According to the budget estimate report from its IT department, the system construction would have cost the City over KRW 10 billion if it had to pay for the data and the data analysis service.
2. Data sources: Securing data through cooperation with public and private institutions.
As big data in diverse areas are the most important resource for the Big Data Project, the City has sought and succeeded in securing necessary data for free through formation of cooperative ties with the holders of the data: call data records from KT for the late-night bus service program; taxi operation data from Korea Smart Card for the “Taxi Matchmaking” program, and traffic accidents data from KoRoad for reducing traffic accidents.
3. Human and technical resources: through partnerships with non-profit organizations.
Through its partnerships with non-profit institutions the City also has secured human and technical resources necessary to its big data-based programs. Big Data Professional Association advises the City on the use of big data for solving city problems, and research institutions at Seoul National University and KAIST provide their research facilities and manpower.
|
|
| 7. Who were the stakeholders involved in the design of the initiative and in its implementation?
|
|
1. The scientific selection of late-night bus service routes.
Anyone who knows what is like to catch a taxi after regular public transportation services end would appreciate the idea of a bus service that runs all night. This seemingly simple idea, however, has one sticky problem to be solved: knowing which routes would be most pertinent to people who regularly return to their homes late at night. Any public service would be considered a waste of limited resources if it fails to serve its intended target group.
Thus began the “Late-night Bus Route Optimization” program which focused on identifying street sections with most foot traffic at night and grouping them into several bus routes. In the past, such routes would likely have been selected based on existing bus operation data and/or the discretion of city employees in charge. With the explosion of big data following the advances of information technology and a bit of open-mindedness towards unconventional policy approaches, however, the City has selected the routes most frequented by late-night commuters by using data gleaned from vast amounts of call data records and data containing use of smart cards by taxi passengers.
2. Selecting the best location for senior welfare service facilities.
In response to an increasing population of citizens aged 65 and over, the City operates a number of facilities throughout its municipality offering diverse welfare programs. But for a lack of established policy guidelines on the provision of welfare service, the supply-and-demand gap between the City and its senior citizens had been widening. To reduce the gap, the City also has turned to the use of big data: By analyzing the census records of its citizens aged 65 and over in conjunction with other types of data by administration section showing their income levels, the availability of existing welfare programs, and the proximity of public transportation services with respect to their places of residence, the City has developed an accurate distribution chart of senior people in each of its municipal districts and gained reliable information about welfare services needed by location and service type. And by applying such information, the City was able to fine-tune the availability of existing programs and better serve the needs of its senior citizens by building welfare facilities in locations deemed most accessible to them.
3. Posting PR materials in the most effective places.
The City produces and posts a great amount of PR materials to keep citizens informed of its business activity; while much of the materials are for the general public, some are for certain segments of citizens or citizens in certain age brackets. Prior to the use of big data, the City used to produce PR materials in numbers deemed appropriate and post them in any available places. But with the launch of the Big Data Project, the City now produces and posts PR materials in numbers and places derived based on big data analysis. For example, materials about youth job training are posted in areas most frequented by youths and young adults who are mostly likely to be interested in employment opportunities; materials about no-collateral, low-interest loans now are posted in low-income areas; and information materials concerning the safety of women returning home late at night are placed in areas with a high concentration of single working women. As such, through the use of big data, the City has been able to reach more of its target groups of citizens with a fewer number of PR materials.
4. An efficient big data application system established through construction of a “Big Data Task Development & Verification” process.
Based on the experience and achievements it has gained from such big data-based programs as the late night bus service and the selection of locations for welfare facilities, the City has established “Big Data Task Development & Verification”, a process through which to verify if a certain city problem can be solved through use of big data. The process centers on analysis by experts of city problems facing each municipal department in terms of the applicability of such data sets as census, traffic, income, social data and application to policy formulation of the data derived in cooperation with the IT department. With the adoption of the process, the City was able to establish a big data application system focusing on solving city problems from the citizen perspective.
|
|
| 8. What were the most successful outputs and why was the initiative effective?
|
|
1. Evaluating the late-night bus service.
In order to see if appropriate routes have been selected for the late-night bus service, the City has used “T-test”, a testing method for finding out if the average of differences between sub groups is statistically meaningful. Through the use of “T-test”, the City has found out that, on average, the number of users of regular bus services drops 17% on rainy days, increases 20% on weekends and holidays, and drops 19% a day after weekends and holidays. These numbers also are found to be identical to the figures derived from the case of the late-night bus service, which proves that bid data have contributed to the scientific and rational selection of the bus routes.
2. Setting up a reliable monitoring system with the help of experts.
The City constantly monitors the strategy development of all of its big data-based programs and evaluates their progress once implemented through use of the governing system it has formed with groups of industry experts. Through this system, for example, the City listens to experts’ opinions on the development of its strategies for improving existing taxi services and reducing traffic and supplements deficiencies if necessary.
3. Monitoring the use of the big data share base.
As for optimum system usage, the City has established a system for monitoring the process of collecting, storing, processing, analyzing data within its big data share base. By monitoring the process, the City has been able to keep data omissions or errors to minimum and thus enable accurate data analyses and forecasts.
|
|
| 9. What were the main obstacles encountered and how were they overcome?
|
|
1. Data securement and process: construction of a big data share base.
One of the major problems in the use of big data was a lack of the technology necessary for processing and analyzing big data. To solve this problem, the City has constructed a big data share base that allows its various departments to share their own data with one another and effectively analyze shared data using the built-in analysis tool. Through use of this system many of its departments have been able to analyze and process data of their interest.
2. Selection of tasks and difficulty of solving: fostering of big data experts.
Selecting tasks suitable for the Big Data Project among various city problems turned out to be a major challenge. The City thus launched a program of training people into “Big Data Curators” to analyze the suitability of target tasks and gather necessary data for selected tasks. By recruiting able bodies, mostly college graduates and corporate employees in search of better job opportunities, and training them in the areas of identifying and solving city problems through application of big data, the City has selected city problems and developed ways to address them in a satisfactory manner.
3. Tepid attitude among departments: Consensus build-up and cooperation system.
The nonchalant attitude of many of the City’s departments towards the Big Data Project also proved to be another huddle to overcome. In the early stage of the late-night bus service program, for example, relevant departments had trouble seeing the necessity of using big data eye-to-eye and thus were less than cooperative in their efforts to push forward the program. To deal with a lack of enthusiasm and cooperation among departments, the City informed its employees of the utility of big data and the importance of synergetic cooperation among departments through a series of education sessions and seminars, thereby developing a workable consensus on big data throughout its organization.
|