Shanna Lucchesi – Project Coordinator, iRAP speaking at the Big Data Results workshop.

DA NANG, Vietnam – August 7, 2025 – Original article published by AIP Foundation

In recent years, the Government of Vietnam has demonstrated strong leadership in prioritizing road safety for children. Following the Prime Minister’s Directive No. 31 (December 21, 2023) and the Ministry of Transport’s Decision No. 64 (January 22, 2024), national efforts have intensified to ensure traffic order and safety around schools. These actions align with the long-term vision outlined in Decision No. 2060 (December 12, 2020), which sets the goal of ensuring that 100% of school gate areas located along national highways, provincial roads, and key urban routes are structured to guarantee traffic safety and reduce congestion.

La AI&Me: Leveraging AI tools for road safety program uses technology to boost change and promote a safe environment for children around schools. The project is funded by Google.org, granted in the call AI for the Global Goals. It is led by the International Road Assessment Programme (iRAP), with project partners including the Fundación AIPANDITI, y el Faculty of Electric Engineering and Computing, y el Faculty of Transport and Traffic Sciences, University of Zagreb.

To present the outcomes of the nationwide Big Data Screening, iRAP and AIP Foundation co-hosted a workshop on “Sharing the Big Data Screening Results” with the Government of Vietnam and national stakeholders. The results were intended to inform the local government in prioritizing improvements to transport infrastructure, thereby contributing to the reduction of injuries and fatalities caused by road traffic crashes involving students.

This event marked the first nationwide dissemination of the Big Data Screening methodology, following its successful pilot in three provinces from 2021 to 2024. It is the first time data science, AI, and satellite imagery have been systematically applied at scale in Vietnam to assess pedestrian safety risks for students.

“The integration of advanced technologies like AI and machine learning into road safety assessments demonstrates our commitment to innovative, data-driven solutions,” said Shanna Lucchesi from the International Road Assessment Program. “By working with Government partners and  international stakeholders to identify and address the most dangerous school zones across Vietnam, we are taking concrete steps to safeguard  children and support the goals of Vietnam’s National Strategy for Road Traffic Safety.”

A Systematic, Data-Driven Approach for Safer Schools 

The project partners, in collaboration with AIP Foundation, assessed road safety risks around schools throughout Vietnam utilizing AI and machine learning technologies in satellite imagery, and Street View images. This assessment identified high-risk schools, leading to recommendations for targeted infrastructure upgrades.

The Big Data Screening methodology systematically filters and ranks risk factors for pedestrians at the community level. It begins by screening nationwide data to identify high-risk provinces using socioeconomic indicators. These provinces are then narrowed down to the most at-risk cities or districts using cost-effective data. Next, the highest-risk schools are identified within these areas, and data on the surrounding built environment are collected. Finally, speed and traffic flow data generate a priority list of 500 schools.

These 400 schools will undergo further detailed assessment with the Star Rating for Schools methodology, utilizing AI to categorize road attributes, speed up the assessment process, and substantially expand the coverage area surrounding school zones. Approximately 40 schools have been prioritized for road infrastructure improvements as part of the program in Vinh Long and An Giang provinces, Vietnam.

Overall, the Big Data screening is an effective method for prioritisation due to its systematic and evidence-based approach. By leveraging large datasets, big data screening can identify patterns and trends that may not be apparent through traditional methods. This approach ensures that resources are allocated efficiently and effectively, leading to better decision-making and outcomes.

There are a wide variety of big data sources that come with varying coverage, frequency of collection, licensing conditions, and cost. The big data sources used in this project included: Government open data sources, Satellite images, Multinet MN-R data & telematics, and Street View images.  In addition, a dataset of 360-degree images from around 900 schools nationwide has been utilized for deeper analysis.

A Collaborative Path Toward Safer School Zones

The workshop brought together government officials, transport representatives, and international technology experts to explore how evidence-based, data-driven approaches can guide traffic infrastructure improvements around schools, aiming to reduce road injuries and fatalities among children.

“Harnessing the power of AI and digital mapping allows us to identify dangerous road conditions around schools faster and more accurately than ever before,” said Mirjam Sidik, Chief Executive Officer at AIP Foundation. “With this data, we can take targeted action to upgrade the most high-risk school zones and protect thousands of children across Vietnam. This innovative approach marks a major step forward in our mission to make every journey to school a safe one.”

The program is part of a broader effort to support the United Nations’ Sustainable Development Goals, particularly those related to quality education (SDG 4) and sustainable cities and communities (SDG 11). It also aligns with the United Nations Decade of Action for Road Safety 2021–2030, which calls on countries to reduce road traffic deaths and injuries by 50% by 2030. Every child deserves a safe journey to school, and through collective action, data-driven solutions, and unwavering commitment, that vision can become a reality in Vietnam.

Enlaces útiles:

  • To read the press release, please click aquí. 
  • To see photos, please click aquí.
es_COEspañol de Colombia