Signed an MOU with Globiz to jointly develop a big data-based CBM platform for Hyundai Rototem.
Source: Hyundai Rotem Official Blog (https://blog.hyundai-rotem.co.kr/280)
The first industrial revolution in human history began in 18th century Britain. The spread of the industrial revolution worldwide was largely facilitated by the steam engine and railways. The railway vehicle industry, born out of the first industrial revolution and thriving through the third, is one of the few enduring sectors. And now, in the era of the fourth industrial revolution, the railway industry is again attempting a new leap forward.
The fourth industrial revolution brings technologies related to the railway industry such as big data, IoT, artificial intelligence, and AR. In particular, trains composed of tens of thousands of parts can collect vast amounts of data, which can be analyzed through a big data analysis platform using machine learning/artificial intelligence. This data analysis can provide information useful for train operation and maintenance. Using AR, this information can be made easily accessible to maintenance personnel on the vehicle, greatly assisting their work.
Today, on the Hyundai Rotem official blog, we will introduce you to the strides that Hyundai Rotem is making in response to the era of the fourth industrial revolution, starting with the development of smart maintenance technology utilizing big data.
Meeting of Railway Vehicle Maintenance and the Fourth Industrial Revolution
The traditional maintenance system for railway vehicles involved checking, replacing, and testing parts and devices at predetermined maintenance intervals (weekly, monthly, etc.) and replacing them as needed. This approach often led to issues such as inability to respond to sudden failures or defects, and unnecessary maintenance costs due to periodic part replacement. This is the very reason why Hyundai Rotem turned its attention to smart maintenance technology.
Smart maintenance of railway vehicles requires real-time analysis and diagnosis of the condition of the vehicle and its parts. Given the vast amount of data that needs to be collected and analyzed for condition diagnosis and fault cause analysis, the meeting of big data analysis and maintenance was inevitable.
▲Historically periodic maintenance of railcars
Hyundai Rotem embarked on the development of a big data analysis platform for developing a condition-based maintenance (CBM) system to manage railway vehicles efficiently. To understand CBM, let’s first define CM and CBM.
Condition Monitoring (CM) is the practice of systematically collecting data over a certain period, either automatically or manually, to understand the condition of a train’s parts/components. Condition Based Maintenance (CBM) is a maintenance management system that decides and performs maintenance work based on the condition of the parts/components. Therefore, to facilitate CBM, it is necessary to collect status data from trains and key components using sensor and IoT technology and analyze it on a big data analysis platform to perform condition diagnosis and fault analysis. By considering the degradation mode of each device and grading the condition of each part, failures can be detected and information that enables quick action by train drivers and maintenance personnel at the vehicle base can be provided. This is a new maintenance technology concept that reduces time-based maintenance (TBM) and increases CBM to optimize train maintenance environments and minimize operation costs.
▲Won-sang Lee, head of the R&D department at Hyundai Rotem Railway Technology Research Center
How important will the development of this big data analysis platform for CBM be in the railway industry? We asked Mr. Lee Won-sang, head of the research and development office at Hyundai Rotem Railway Technology Research Institute, which is in charge of developing smart systems and system engineering for Hyundai Rotem’s railway vehicles.
“Typically, the cost of buying and operating a railway vehicle consists of the vehicle purchase cost, operating cost, and maintenance cost, each accounting for about 30% of the total cost. However, looking at the Life Cycle Cost (LCC) of a railway vehicle that operates for over 30 years, the proportion of maintenance costs is gradually increasing. Maintenance costs are divided into preventive maintenance and corrective maintenance. Given the importance of safety in the railway industry, the proportion of preventive maintenance costs can be said to be very high.
The current maintenance of railway vehicles involved replacing parts or completely retrofitting on a daily, weekly, or annual basis, so it was more expensive than repairing only the faulty parts of a car. However, if a big data analysis platform is developed and applied, the vehicle’s condition can be constantly checked through sensors, and the lifespan and failure occurrence time of parts can be analyzed to prevent accidents before they happen. If you compare this to humans, it’s like using smart wearable devices to check your health status in real time and prevent diseases, a concept befitting the era of the fourth industrial revolution. Currently, as the vehicle price and maintenance costs are similar, we expect that the introduction of such smart maintenance technology will reduce maintenance costs by more than 30%.
Railway vehicle manufacturers can lower the LCC price by the amount of reduced maintenance costs, securing global competitiveness, and naturally, they will be able to invest more in R&D. We foresee this technology becoming a symbolic technology that allows Hyundai Rotem and the domestic railway industry to take one step further in the era of the fourth industrial revolution.”
Thus, CBM, which conducts maintenance based on the condition of the train and its parts through a big data analysis platform, has a tremendous impact on the LCC of railway vehicles. Not only does it improve maintainability through cost savings in maintenance, but it also directly links to safety by preventing faults. Being able to constantly understand the situation of railway vehicles and promptly respond to all situations also brings the advantage of improving availability.
The ins and outs of CBM (Condition Based Maintenance) technology
We will take a closer look at CBM technology, which many of you may be curious about. In railway vehicles, key component data is collected via the TCMS (Train Control and Management System) and communication networks. The main information of the components is measured through sensors such as pressure, temperature, current, and voltage, and the data collected in the TCMS is collected in the database of the big data analysis platform via a wireless device on the ground in real-time or after some preprocessing when the train enters the base.
Data collected through IoT (Internet of Things) technology that connects each component and vehicle to a network is reorganized into more useful information through diagnosis and analysis on the platform. All the digitalized states of the railway vehicles go through failure detection logic analysis and are made into a ‘failure prediction algorithm’. This provides relevant information in the form of a dashboard to perform preventive maintenance at the appropriate time.
Because the big data analysis platform is still under development, a demo project is included in the development process. This demo project plays a role in verifying the platform. It needs to verify whether the train data is being collected and stored in real-time, whether it is accurately diagnosing and analyzing the status of key parts. It needs to create a criterion to decide whether the algorithm that predicts failures is correctly applied and can predict actual major device failures, and at what point it is best to predict failures. Only when all these processes are verified, is it possible to actually use data for smart maintenance.
MOU for joint development of big data analysis platform
Hyundai Rotem has been setting up various response strategies to greet the fourth industrial revolution, but there were concerns about software technology as a comprehensive heavy industry company conducting railway, plant, and defense business. These concerns led to joint development with a company possessing technology to systematically collect, process, and analyze big data.
▲Signing ceremony for joint development of smart maintenance system big data analysis platform held on March 18th
Globiz, who conducted joint development of the big data analysis platform with Hyundai Rotem, has CM (Condition Monitoring) know-how in the railway field based on experience in status monitoring and data analysis of high-speed trains. Hyundai Rotem, who was looking for a company capable of developing a PHM (Prognostics and Health Management) algorithm with railway business experience, signed an MOU for a friendly partnership with Globiz and decided to jointly develop a unique big data analysis platform.
▲CEO of Globiz Im Jong-soon and Director of Hyundai Rototem Railway Technology Research Center Lee Won-sang at the MOU.
Hyundai Rotem and Globiz plan to develop the big data analysis platform from 2019 to 2020, proceed with detailed design and tests, and after thorough system verification, plan to construct a smart maintenance system using CBM. This system will be developed to be expandable and applicable to plant business such as constructing a smart factory as well as railway business, and it may take off as Hyundai Rotem’s asset management system.
When the big data analysis platform is developed, Hyundai Rotem will be able to propose smart maintenance systems at competitive prices in both domestic and foreign markets. Also, it is expected that the operation and maintenance business, a new business, can be secured not only for newly manufactured vehicles but also for vehicles that have already been supplied. You may look forward to the transformation of railway vehicles into a safer and more efficient advanced technology platform in the near future. Please pay a lot of attention to Hyundai Rotem, who is leading the development of technologies leading the global market in the fourth industrial revolution era.