Minimize disruption of education following natural hazards using Artificial Intelligence Natural Sciences – Disaster Risk Reduction

The physical vulnerability of school buildings to different natural hazards is found to be high as evidenced by global study on the subject. School infrastructure exposed to natural hazards has an estimated value of $13.6 trillion globally.

The global forecast of direct annual losses from earthquakes in school infrastructure is estimated in $3 billion while the direct annual loss from tropical cyclones is around $4 billion. Historical information indicate that the school infrastructure located in the East Asia & Pacific and Latin America & Caribbean regions may experience more than 20% of the total direct damage from large earthquakes and flooding.

UNESCO has been working on reducing the vulnerability of the school infrastructure to improve the safety level for multi hazards. UNESCO deploys a multi-hazard school safety assessment methodology known as VISUS (Visual Inspection for defining Safety Upgrading Strategies). The methodology has been successfully tested in seven countries, assessing the safety of more than 500,000 students and educational staff so far.

This highlights a compelling need for developing and implementing integrated, and ‘ground-real’ strategies for assessing and improving the safety and resilience of schools. This entails considering several aspects of the problem: characteristic and location of the school compounds and buildings, frequency and severity of several possibly concurring hazards, the resulting damage and interruption of the education experience for the students, the effect on this provided by social indicators and possible ways of mitigating detrimental effects. A systemic approach is needed to support decision making and investment that can be effective and fair at scale for the development and implementation of sustainable disaster risk reduction (DRR) and preventive solutions.

One way of addressing this issue is to use advanced computational and statistical approaches. Advance computational modelling, based on AI can help model the complex systemic interaction between school compounds and infrastructure at district or regional level and other critical infrastructure that serves it. Such models can help estimate the disruption to education due to multiple hazards to quantify its resilience, considering physical damage of buildings, accessibility loss, connectivity through road network, sheltering of displaced communities and social vulnerability aspects of the infrastructure.

The UNESCO is implementing a project in Dominican Republic aimed at developing strategies and action plans disaster risk reduction in schools to minimize the disruption of education for municipalities using the idea mentioned above.

UNESCO therefore seeks the individuals or organizations with experience on AI and DRR to develop the model to analyze the risk of the disruption of education integrating different aspects in case of natural hazards in Dominican Republic and provide strategy for investment to mitigate the disruption of education.

The individual or organization shall be expected to develop the following functions:
Identification of critical Data set and data collection
Identify and collect the essential sets of data to assess the risk of disruption of education such as:

• Natural hazards relevant to the site and their relevant data (frequency, intensity etc.)
• School infrastructure characterisation (structural vulnerability etc)
• Lifeline infrastructure characterisation (road network and its vulnerability etc)
•  Socio economic information on schools (contingency plan for shelter, number of students etc)

Modelling

Develop a probabilistic framework to determine causal effect and correlation between the variables and develop a model to represent the behaviour and response of the different interacting systems (educational, social, transport). Tasks might include:

• Develop algorithms to establish correlation among variable and their probabilistic representation based on data analysis
• Determine and model correlation between variables
• Model the interactive decision-making process between various stakeholders and the system.
• Consider possible use of advanced computational tools for simulating decision-making processes and optimisation strategies.
• Identify meaningful scenarios to describe performance and interaction at the system level.

Expected outcomes:

The framework shall be able to provide insight into:

• Hazard specific physical and functional vulnerability of the school infrastructure at district level.
• Hazard specific physical ad functional vulnerability of the road network serving the schools system and its community.
• Optimal strategies for preventive intervention or rapid response to increase resilience and facilitate recovery of the school system, its education function, and of other critical infrastructure.
• Determine the role of specific socio-economic indicators of the communities served on the causes and extent of disruption.

Travel fee

In case suggesting the physical meetings for collecting and demonstrating the outcome, please include the mission costs up to 10,000$.

Schedule

All the tasks above should be done by the end of March 2023.

How to submit the proposal

To submit the proposal, please send your proposal with the schedule and budget plan together with your organization/individual’s information and past similar experience to drr@unesco.org before 4 July 2022 (midnight).

Please note that only pre-selected proposals will be contacted.