AI system
- System
-
510 impact-based forecasting (IBF) model for typhoons in the Philippines, operated through the 510 dashboard with PAGASA forecasts
- Model
-
XGBoost.
- Release
-
Exact release not established.
- Application developer
-
510, an initiative of the Netherlands Red Cross, with the Philippine Red Cross and German Red Cross - Role
-
The model forecast damage to housing before a typhoon.
- Configuration
-
Not assessed
- Inputs and outputs
-
Not assessed
- Autonomy
-
Not assessed
- Workflow
-
510 IBF damage model fed by PAGASA typhoon forecasts. Pre-approved EAP with a 10% projected-damage trigger. IFRC anticipatory DREF funds released on trigger. Pre-positioned shelter kits, trained carpenters and pre-identified households.
- Human oversight
-
Philippine Red Cross staff checked local capacity and selected the municipalities for action.
- Relation to AI
-
The ML impact forecast was the trigger. Under the EAP, early actions fire on the forecast wind impact on housing 72 hours before landfall. The trigger: at least 10% of houses totally damaged in more than three municipalities. The notification records the Aurora municipalities as “Met (More than 10%)”. The 72-hour lead-time condition was not met. The final report says the 56-hour lead time was exceptionally granted. The notification says the projection came from the 510/IBF forecast incorporating PAGASA data. PRC staff confirmed chapter capacity and chose municipalities. The outcome claim is PRC's own post-storm survey. The NHESS paper and EGU abstract on the model have 510 staff as co-authors, not lead authors.
- AI contribution
-
The 510 model forecast that damage to housing would exceed the threshold for early action in Aurora. Its forecasts incorporated data from PAGASA.
Benefits
- Outcome identifiers
-
Not assessed
- Benefit
-
installed shelter kits for 78 Aurora households before Storm Nalgae
- Category
-
Reduced Risk
- Status
-
Not assessed
- Beneficiary groups
-
People. Organizations. Society.
- Beneficiaries
-
Not assessed
- Reach
-
Not assessed
- Distribution and access
-
Not assessed
- Duration
-
Not assessed
- Expected?
-
Yes
- Costs and harms
-
Not assessed
Measurement
- Studies
-
Not assessed
- Study design
-
Not assessed
- Measurement
-
Not assessed
- Setting and sample
-
Not assessed
- Measures
-
installed shelter kits for 78 Aurora households before Storm Nalgae
- Comparison
-
Not assessed
- Results
-
installed shelter kits for 78 Aurora households before Storm Nalgae
- Additional value from AI
-
Not assessed
- Limitations
-
The planned 72-hour lead time was not met; the final report records an exceptional approval with 56 hours remaining. The Philippine Red Cross’s own survey supplies the reported benefits after the storm.
- Denominators
-
Not assessed
- Materials for replication
-
Not assessed
Money
- Kind of value
-
Not assessed
- Amount
-
Not assessed
- Currency and prices
-
Not assessed
- Period and unit
-
Not assessed
- Realization
-
Not assessed
- Recipients and payers
-
Not assessed
- Gross or net
-
Not assessed
- Included costs
-
Not assessed
- Comparison
-
Not assessed
- Method
-
Not assessed
- Return and payback
-
Not assessed
People and places
- Participants
-
Developer:
510, an initiative of the Netherlands Red Cross, with the Philippine Red Cross and German Red Cross. Operator: Philippine Red Cross, with IFRC anticipatory DREF funding. - Jurisdictions
-
Benefit: Dipaculao, Aurora Province,
Philippines. Developer: 
Netherlands. Operator: Philippines. - Ownership and control
-
Not assessed
- Dependencies
-
Not assessed
- Cross-border connections
-
Not assessed
- Geopolitical implications
-
Not assessed
Follow-up
- Conditions for replication
-
Lead time near 72 hours. Forecast track and intensity accuracy. Weather permitting installation before landfall. Local government authorization in no-build zones.
- Replication attempts
-
1 follow-up entries. Outcome unresolved: 1.
- Later findings
-
30 June 2023. The IFRC final report of 30 June 2023 records all 78 households served and houses reinforced before impact, but a PRC lessons-learned workshop judged the trigger high and recommended reviewing the typhoon EAP.
- Availability
-
Not assessed
- Institutional decision
-
Not assessed
- Grounds
-
Not assessed
- Action and result
-
Not assessed
- Success criterion
-
Not assessed
Evidence
- Source numbers
-
AIBD-0098-source-0001–0004
- Citations
-
Numbered references below.
- Language
-
Not assessed
- Source relationships
-
Operator, Developer
- Original evidence
-
Not assessed
- Access and versions
-
Not assessed
- Supported findings
-
Not assessed
- Evidence links
-
Not assessed
- Evidence strength
-
Not assessed
- Missing evidence
-
Not assessed
Grades
- Assessed benefits
-
People. Organizations. Society.
- Depth
-
People: 2. Organizations: 1. Society: 1.
- Reasons
-
People: 78 households had roofs and walls reinforced before landfall. PRC reports they mitigated wind impact (implementer survey, no counterfactual). Organizations: First completed activation of the Philippine typhoon EAP. Lessons on lead time and weather constraints documented in the final report. Society: Small scale: 390 people reached in one municipality.
- Observed or intended
-
Not assessed
- Rubric
-
Not assessed
- Assessor and date
-
Not assessed
- Alternative assessments
-
Not assessed
- Fill
-
The deployer's account
Entry details
- Kind
-
Not assessed
- Related entries
-
Storm-impact model used for pre-landfall operational decisions underestimated Sandy's surge and NJ Transit lost equipment; 510's typhoon impact model triggers Red Cross action the same way.
- Field
-
Disaster risk reduction and forecast-based financing. Red Cross National Society.
- Purpose
-
Forecast typhoon housing damage per municipality to trigger pre-agreed early actions before landfall.
- Dates
-
Date basis unassessed: 2022 (precision unassessed). Updated: 21 October 2026.
- Stage
-
Not assessed
- Version
-
SHA-256: 4d2768523209…
References
- Other identifiers
-
AI Incident Database: 536.
- Category mappings
-
Not assessed
- Brief
-
Not assessed
- Revisions
-
7 revisions. Latest: 21 October 2026.
Sources
-
Peer-reviewed study Developer
Kooshki Forooshani, M., van den Homberg, M., Kalimeri, K., Kaltenbrunner, A., Mejova, Y., Milano, L., et al. (2024). Towards a global impact-based forecasting model for tropical cyclones. Natural Hazards and Earth System Sciences 24, 309–329.
In 2019, the 510 model was approved as the trigger model for the early action protocol for typhoons
-
Report Deployer
shelter damages are projected up to 10 per cent in different municipalities in the priority areas under the Early Action Protocol (EAP) for typhoon
-
Report Deployer
Despite of the challenges, 100 percent of 78 households who received the kits were successfully installed/reinforced.
-
Other Developer
Free, R., van den Homberg, M., Sperna Weiland, F., Teklesadik, A., Melillo, M., Bogaard, T. (2022). Extending a ML impact-based forecasting model for typhoons in the Philippines with a rainfall threshold for consecutive landslide events. EGU General Assembly 2022, EGU22-10474.
An impact-based forecasting model for typhoons currently operational in the Philippines predicts impact due to the high wind speeds associated with typhoons