The Smart Station project for future urban subway station has accelerated the digital transformation for public transportation industry, which has been slow to change, by applying fourth industrial technologies such as IoT, 3D digital twin technology, AI image analysis, and deep learning technology. Through this, the station has secured a control tower that can manage all lines for safe subway station operation, station workers can innovate their work through augmented reality UI systems.
Seoul Metro oversees 300km of track, 277 stations and 3,571 trains, but each line’s facility management systems are built in silos. Therefore the same system for Line 1 and Line 4 cannot communicate with one another while data generated from original sections are incompatible with data from newly built sections. As the scope of public transportation service continues to expand, the complexity of facility management and the burden of monitoring and control duties has also increased. For system patches or upgrades, engineers had to visit the site in person since remote access was impossible. Most facilities have been in operation for more than 20 years, and their outdated condition is an increasingly frequent source of disruption to subway service. The amount of time and money spent on maintenance has also continued to increase while in some cases maintenance is no longer possible.
Also, a control tower for governing emergencies was absent. Problems were only detected during regularly scheduled onsite inspections by the station staff, but while the staff was out on inspection, no one was in the station office to take control of emergencies. This resulted in an inability to detect problems in other station areas and it inhibited staff from taking the necessary immediate corrective. To address this issue, Seoul Metro had to increase the number of station staff members. However, cost efficiency could not be ignored, and due to labor costs Seoul Metro was not in a position to improve subway service quality by simply increasing personnel.
Beyond all these operational issues, Seoul Metro was under great pressure to satisfy heightened passenger expectations in the era of 5G. Since the video data quality of the station cameras was too low (0.4 MP), making accurate decisions based on that footage was to no avail. To make matters worse, the data collected from each system were saved in different storage facilities, preventing Seoul Metro from developing new data-based services for advanced passenger services.
By applying IoT, deep learning AI imaging analysis, and 3D digital twin systems, Seoul Metro has initiated the Smart Station Project to increase service quality for the public, reduce the workload of station staff and enhance their working environments.
- Transforming the station staff to digital workers with digital twin and automated operations
Previously, security issues were notated as dots on a 2D map, leaving station staff to determine the location based on the map’s scale. By adopting digital twin, station environments are virtually reproduced in the system to provide more accurate location data so station staff members can monitor sites without having to vacate the office while performing inspection duties. Station staff now have the ability to make decisions immediately and dispatch personnel as needed thanks to AI recognizing issues and sending proactive alerts. Automated operations lets staff handle a larger work scope and volume while also improving their operation rate by 65%. Now facilities, such as the shutter, are automatically managed by sensors and schedules rather than humans. When a pre-defined abnormal situation occurs, the facilities are operated automatically according to the predetermined rules.
- Improving safety with AI and deep learning technologies
With the combination of computer vision AI and deep learning, fare evasions and criminal activities are detected in real-time and the station officer can warn passengers or send staff to the site. Computer vision executes real-time monitoring for COVID-19 safety measures by identifying those who are not wearing face masks, and initiates immediate responses to solve the issues. The deep learning engine not only learns from video data, but also from numerical data collected from the system to automatically detect all abnormalities and deploy immediate responses by sending real-time alarms 24 hours a day. This reduced the response time to safety accidents by 75%. Subway safety is now guaranteed by AI technology that never rests and accurately detects what the human eye cannot.
- Control tower readiness, covering all subway lines with standardized integrated platform
Seoul Metro resolved the problem of non-standardized interfaces that once plagued the subway by adopting a standardized IoT platform that easily integrates heterogeneous devices. Previous system upgrades consumed lots of time and were expensive to undertake but now they can be remotely managed in a secure manner on a scalable, independent, and open system architecture. Furthermore, new data-based services can be planned or launched thanks to standardized data governance and simplified data extraction. Integrated new control systems will be the foundation for the upcoming control tower currently under construction. This will control all 277 stations remotely and manage any provisioning allowing staff at each station to focus on field-centered service going forward.
What Makes Your Project Innovative?
Seoul Metro’s digital convenience service for our users have grown in both variety and scope. Staff are now immediately notified when mobility impaired persons enter a station so that they can receive assistance without having to calling for help. Now, they are ready to transform to digital workers with digital twin and automated operations. Also, automated video analytics may help improvement in safety and increase protection again COVID 19 infection by monitoring and controlling the presence of face masks and responding accordingly.
IoT have been applied in the safety and health sectors. This project resolved the challenge of interpreting and pulling value from the various data stream, respectively, it in turn allowed Seoul Metro to provide the best quality service for public infrastructure users.
Moreover, newly implemented control tower covers all service types, e.g. metro schedule, and spaces from train platform, to public service areas, such as toilets, escalators, and so on.
What is the current status of your innovation?
Seoul Metro launched in 1974 and has grown to 8 subway lines that transport over 7 million passengers per day. However, its facility systems run on analog safety protocols which require a large workforce in order to maintain a high level of service. Furthermore, the metro system continues to expand to keep pace with Seoul’s growing population, resulting in a greater area to monitor and maintain.
Seoul Metro inaugurated the Smart Station Project to lower operational costs while improving service quality through the use of IoT, computer vision AI, deep learning, and digital twin. The scope of this project was for lines 1 and 4 out of a total of 8 lines, and we plan to expand the scope of implementation to all lines after that.
We have completed the smart station business implementation and are currently operating successfully in the field. We successfully completed the smart station system for lines 1 to 4, 8 and we are expanding it to lines 5 to 7. The service will be continuously.
Collaborations & Partnerships
In this project, Samsung SDS, Samsung Group's No. 1 SI company in Korea, formed a consortium with various partners to carry out the project. The Samsung SDS consortium has established a smart station system for lines 1 to 4 and completed integration and standardization. Samsung SDS applied it as a key engine using Brightics IoT solution, an IoT platform developed by the company, and implemented standardized protocol lake to integrate various system linkages and sensors.
Users, Stakeholders & Beneficiaries
- User : Seoul Metro (Government Official)
- Stakeholders : Seoul Metro (Government Official), National Police Agency, police station, fire station, Ministry of Public Administration, Office of the President, intelligence organization, etc
- Beneficiaries : Citizens of Republic of Korea, Seoul Metro Employees, Seoul Metro Decision-makers, Police and Fire officers, Seoul Metropolitan Government officials, persons with disabilities, etc
Results, Outcomes & Impacts
Seoul Metro decided to introduce new IT technologies to achieve maximum operational efficiency while reducing operating costs. This decision allowed for the provision of innovative public services where citizens can actually feel their benefits while achieving better cost and operational effectiveness.
Through the introduction of this new technology, we reduced annual operating costs by roughly $2.3 million (USD) by not having to add 72 additional personnel requested for operations and saving an additional 30% in equipment maintenance expenses.
Also, IoT sensors provide real-time subway information to users on their mobile apps to find out estimated arrival time and make decisions regarding delays and other abnormalities in subway operation.
Though real-time data from IoT sensors in this project is currently used for integrated control systems, it will be also used as key data for other Seoul Metro projects to expand data-based digital transformation.
Challenges and Failures
Seoul Metro had difficulty operating separately implemented systems for a number of different reasons, including additional device installation and system integration, application upgrades, and end of service issues. Silo systems prevented the creation of data-driven decision-making due to interrupted processes. In order to resolve this issue, Seoul Metro has decided to integrate all dissimilar systems across all lines (1-8) into one platform. Integrating 10 systems into one platform has proven to be a great challenge: a standard process had to be defined to govern how each system collects and processes data in real-time in the data pipeline. In addition, there are some cases where hardware devices end up being discontinued, so finding a way to integrate heterogeneous devices was essential to achieving maximum performance from connected devices.
Conditions for Success
The most successful result is the completion of the standard platform
- Implementing a standardized platform for rapid analysis of endpoint data
An integrated IoT platform able to simultaneously process more than 300,000 pieces of data has been introduced and the platform’s data pipeline decides whether to collect and analyze the data directly in the endpoint device in real-time or analyze it after transmission
- Applying a standardized methodology assuring compatibility between heterogeneous devices
After determining the used devices and the types of data that need to be saved, we defined the standard protocols needed to integrate those devices in advance. By providing integration protocol packages, including APIs, even non-experts can integrate a variety of devices with the IoT platform
- Establishing intuitive virtual stations for non-experts
We have completed a decision-making mechanism based on 3D and data that can quickly recognize, decide and respond to abnormal situations
Smart Station Project for future urban train has accelerated digital transformation for public transportation by adopting 3D digital twin technology, AI video analysis, deep learning, which can also be exploited in airports, ports, and railroads. Seoul Metro has enhanced station operation efficiency and built a control tower by digitalizing equipment management so that information is available anytime, anywhere as this video describes. This is a notable success case where innovations in public infrastructure have improved their operational efficiency along with citizen satisfaction, and this success can be replicated to other transportation agencies which need to adopt modern digital technology. In addition, we share data with national safety agencies such as the National Crisis Management Center and the National Police Agency to realize the vision of a true Smart Connected Metro that links the government, cities, and citizens.
The start of this project began with a simple logic of securing an easy operating environment that increases safety and efficiency. It was questionable whether the purpose of realizing an easy operating environment could be achieved. However, Seoul Metro found many areas for improvement through pilot projects, corrected them, and succeeded in implementing a high-quality system. It is the first system to be constructed in Korea, and now many domestic and overseas public institutions are visiting to benchmark the smart station system of the SeoulMetro, which has inspired many visiting customers to start their own 3dD-based integrated platform business. During system operation, Seoul Metro also made many consultations, trials and errors, and corrections with partners to solve the system's challenges. In the future, we will expand various services based on this standardized platform. Thanks to this project, We has reduced overlapping costs and standardized the facility management sector.
I advise organizations that want to pursue similar projects that the most difficult thing is system integration, and the most important thing is the IT policy of standardization. When integrating heterogeneous systems, a new platform should be the standard.
The method we chose was to focus on standardizing the new platform as a standard platform without relating to the existing system. In order to accomplish this, we equipped the integrated platform with an engine that can accommodate protocols that comply with global standards, updated and linked the protocols of existing systems and devices, and set policies to mandate all new equipment and devices purchased in the future to be adapted to the new platform. As a result, we predict that we will be able to achieve our goals of compatibility, scalability, and budget reduction when expanding services and systems in the future. We have established a standard integrated platform. This is our know-how.
- Implementation - making the innovation happen
- Diffusing Lessons - using what was learnt to inform other projects and understanding how the innovation can be applied in other ways
2 January 2023