Major government-funded research institutes that study disaster and safety technologies said the center of gravity in disaster response is shifting from "complete shutdown" to securing the golden time, and that they are advancing response technologies to block spread and enable early prediction.
They also said that while research autonomy and cross-institutional convergence are expected to expand after the abolition of the project-based system (PBS), safeguards such as performance management and accountability frameworks are needed to prevent autonomy from devolving into laxity.
On the 25th, Mun Byung-seop, vice president of the Korea Institute of Civil Engineering and Building Technology (KICT), Kwon Yong-jang, vice president of the Korea Railroad Research Institute (KRRI), and Lee Myung-jong, vice president of the Korea Institute of Geoscience and Mineral Resources (KIGAM), presented each institution's disaster response research status at the Korea Science and Technology Center in Gangnam-gu, Seoul.
Mun, the KICT vice president, cited "logistics facility fire response" as a representative research case. Mun said, "For new buildings, you can use good materials that minimize fire occurrence, but the problem is buildings already constructed," adding, "A 'non-demolition infill method' that reinforces an operating logistics center without tearing it down is an alternative."
The non-demolition infill method creates fire compartments by filling belt-shaped sections of wall with insulation that blocks fire spread. KICT said this approach delays spread while keeping logistics facilities in operation, buying time for evacuation and initial response. Mun said full-scale tests showed it is possible to secure 15 minutes of golden time.
Mun added, "We conducted a pilot application at a Lotte logistics center through a public call," and "after related details were reported by the media, LG also expressed interest."
KICT is also studying strategies to prepare for urban flooding. Mun said, "When sudden heavy rain hits, water rushes into city centers all at once due to impervious paving (asphalt and concrete that do not absorb rainwater), increasing damage. So strengthening advance warnings is a realistic solution," adding, "We are expanding flood prediction points with artificial intelligence (AI) and implementing a service that links with navigation to guide drivers through hazardous sections in advance." Mun also said, "There were 75 existing prediction points, but we can expand that to 223 using AI."
◇ Underground is fiber optics, tracks use satellites… disaster response becomes "precise" with AI
KIGAM proposed integrating the full cycle of detection, diagnosis, and response for underground disasters such as earthquakes, landslides, and sinkholes. Lee, the KIGAM vice president, introduced KIGAM as "a research institute that studies the underground," highlighting major results including building a seismic environment model for the Korean Peninsula and advancing early warning, a prediction model targeting secondary landslide risks driven by heavy rains following wildfires, and an urban sinkhole detection and stability assessment system.
Lee introduced landslides as a typical compound disaster, saying, "After a wildfire, when summer comes, heavy rains increase the probability of landslides in the affected area. We developed a technology that can predict this within two and a half hours, and we plan to incorporate AI to expand prediction accuracy and applicability."
In particular, KIGAM is envisioning a national-scale monitoring system that uses dark fiber (spare optical fiber) in the nationwide optical communication network as sensors to detect temperature, vibration, stress, and more. Lee said, "A lot of research has proceeded with detection, diagnosis, and response segmented from one another, so actual disaster response has lacked a consistent framework," adding, "we will aim for a national strategic integrated AI platform for geohazards."
Kwon, the KRRI vice president, said climate disasters are becoming a direct threat to rail safety, and announced plans to preemptively respond by using satellites to identify sections where rails expand during heat waves and risks rise. Kwon explained, "We are developing a system that automatically detects hazardous sections, issues warnings, and, if necessary, allows slow-speed operation only in those sections."
Unmanned and robotic field work for safety was also emphasized. Kwon added, "In areas like tunnel and track maintenance, where working conditions are poor and accident risks are high, robots and physical AI should replace people," and "we are even reviewing the feasibility of 24-hour urban rail operation by combining unmanned driving, unmanned maintenance, and robot-based infrastructure upkeep."
◇ Abolishing PBS… "From money-following research to self-directed and convergent"
The event also touched on changes in the research ecosystem at government-funded institutes after the abolition of PBS. PBS is a system under which government-funded institutes win national research and development (R&D) projects through competition. The government is pushing to phase out PBS and to reform the system so that, based on stable basic funding, government-funded institutes expand mission-centered research and cross-institutional convergence research.
Kwon said, "Under the PBS regime, repeated 'research you do when someone pays you' can lead to subordination to client demands, and at times progress checks can drive the work more than research depth," adding, "after the abolition of PBS, self-directed research will enable deeper studies and help restore researcher initiative."
However, Kwon added, "There is also a risk that autonomy could drift into laxity," and "internal performance management and accountability systems will become more important."
Meanwhile, regarding the strategic research groups (large-scale convergent projects) being promoted ahead of PBS abolition, Mun said, "The era of building self-reliance through competition has passed, and this is an era of creating better discourse through convergence," adding, "from early March, government-funded institutes have scheduled discussions on how to autonomously combine research to carry out strategic research. For example, combining KICT's ground subsidence prediction based on underground buried pipes (such as water and sewage) with KIGAM's groundwater and sinkhole analysis could create synergy."
Lee also added, "With strategic research as a catalyst, each institution's core programs will not remain fragmented but will evolve into convergent projects," and "the results could translate into outcomes the public can feel."