To improve transparency, the Centre has created a catalogue of predictive models with information on ‘who is doing what, where and when’. We hope that this effort will make it easier for partners to get a quick overview of the models that are available and their current state of development. If your organization would like to be listed in the catalogue, please fill in this form. Learn more about the Centre’s Peer Review Framework for Predictive Models.

Model name Partners involved * Topic * Geographical scope * Description Update frequency Link to model code (if available) Main contacts Additional documentation
Project Jetson UNHCR Innovation Service, UN Global Pulse, Essex University Big Data & Technology Centre, Omdena Foundation Affected people: Internally displaced persons; Refugees and persons of concern Somalia, Ethiopia Jetson is experimenting with two methods: gravity model and multivariate TSA, machine learning-based (open source, Python and R applications). The target variable is forced displacement, and the nexus between displacement, climate/weather anomalies and changes and violent conflict. The model predicts at least one month of forced displacement movement (population flow, influx-only). This is the arrivals of IDPs and Refugees in the different regions of Somalia (admin level 1) and the southern border region of Ethiopia (Dollo Ado). Every 3 months since 2017 Code repository, including all the source code scripts necessary to run the model. Official Website
Unaccompanied and separated children projection model UNHCR, University of East Anglia Affected people: Refugees and persons of concern South Sudan, Democratic Republic of the Congo, Somalia Arrivals of unaccompanied and separated children in refugee settings. The outcome is a prediction of the number of arrivals of Unaccompanied and Separated Children (UASC), usually by month but sometimes by week. Under development - models created for several countries and updated each year.
The Joint Analysis of Disaster Exposure (JADE) OCHA (Asia-Pacific), WFP (Asia-Pacific), Pacific Disaster Centre Affected people: Humanitarian needs 20 countries supported in the Asia Pacific region Estimates the number of people likely to be in need after a sudden onset natural hazard. The JADE provides the number of people living in affected areas, the number of people living in the worst affected areas, and the number of people living in the worst affected area who were already considered "vulnerable" before the crisis. There are a number other values and figures provided. The spatial resolution is 1 x 1 km. The initiative was launched in 2018. The actual analysis is launched within 24 hours of a sudden onset emergency. It is not expected to be updated during the course of an emergency.
Flee Brunel University London Affected people: Internally displaced persons; Refugees and persons of concern The results cover case studies of three African countries at the moment but can be generalized to more contexts as per the authors. Flee aims to establish a tool that can (a) forecast the destinations of forcibly displaced persons, (b) estimate the effects on migration of major policy decisions such as border closures and camp relocations, (c) provide an estimation of the number of displaced persons in areas where the data collection is incomplete. The first major version of Flee was established in 2017. It is updated continuously and is funded until May 2023. An updated and more realistic rule set for agent movement is expected in Q1 or Q2 2020. Code repository, including all the source code scripts necessary to run the model. Derek Groen
INFORM Risk Index Joint Research Centre of European Commission, OCHA Coordination and context Global coverage (subnational in some countries) Risk for humanitarian crisis and disasters is evaluated along three dimensions coping capacity, vulnerability and hazard. The model has global coverage, and the output is national scale. It was developed in the 2014, and it is updated twice a year Detailed INFORM methdology
Foresight model Danish Refugee Council, IBM Affected people: Internally displaced persons; Refugees and persons of concern Afghanistan, Myanmar The project develops a forecasting model for total forced displacement from a country 1-3 years ahead. Total forced displacement is a combination of refugees and asylum seekers outside a given country and the internally displaced in the same country. The project was updated in 2019-2020 and it will be continuously updated with more countries. Code repository, including all the source code scripts necessary to run the model. Published research article
The Inundation Model - Google flood forecasting model Google AI for Social Good Coordination & Context: Affected areas Pilot in India (Bihar) with extended coverage in in 2019 in the Indian subcontinent. Simulation based real time flood forecasting model. Accurate predictions of severity and locations of flooding based on AI and physics simulation. Integrated with local partners to put an active warning system in place. Project started mid 2018 and latest 2019 iteration available online. Model details
African Risk Capacity African Union, Africa Risk Capacity Affected people: Humanitarian needs 33 African Union member states Large umbrella effort to have better responses towards extreme weather events and natural disasters. The frameworks uses predictive analytics and innovative financing (insurance to redistribute risk and risk pooling) to respond. Framework covers multiple disasters types: droughts, floods, diseases, and cyclones. The pure modelling efforts are ARV (Africa riskview). ARV model updated every year (sometimes for every harvest season) Proprietary software package - access can be requested. Model overview
Famine Early Warning Systems Network (FEWSNET) USAID, NASA, NOAA, USDA, USGS Food Security & Nutrition: Food security 28 countries with a focus on West, Southern and East Africa Early warning and analysis of food insecurity - livelihood based model. Efforts are made to understand impacts on livelihoods, markets/sources, nutrition. A key component of monitoring is satellite based remote sensing. Mixed methods analysis (quantitative and qualitative analysis) enables FEWSNET to build reliable scenarios and Integrated Food Classification (IPC). Updates may be issued every month. Consistent new datasets for all monitored countries are released every ~6 months (every harvest season)
Artemis- Famine Action Mechanism World Bank Food Security & Nutrition: Food security Pilot in 5 countries to be expanded to 21 additional countries The model combines qualitative, remote sensing, and data insights from multiple sources to generate a subnational signal of food insecurity at least 6 months to 1 year in advance. No planned updates of the model after publication in scientific journal. Scientific paper to be published in 2020
Hunger Map LIVE WFP and Alibaba Food Security & Nutrition: Food security Global coverage (90 countries) Advanced statistical/ML system with a focus on large scale nowcasting. The hunger map is a platform that uses streams of public data from multiple sources to create a complex dynamic subnational portrait of world hunger. In places where the data is missing ML algorithms are used to make extrapolations. Daily Official Website
Crisis Computing/ Artificial Intelligence for Digital Response Qatar Computing Research Institute Coordination and context: 3W-Who is doing What Where; Affected Areas Global coverage Natural Language processing model which focuses on extracting disaster and emergency related information from social media messages. The objective of the model is to gain actionable information from troves of social media to inform better humanitarian response In real time Code repository , partial code is published publicly.
Asset Impact Monitoring System (AIMS) WFP Geography and Infrastructure Pilots in Niger, Afghanistan, Sudan, South Sudan, Tajikistan The objective of this system is to monitor the rehabilitation of degraded landscapes using satellite imagery and remote sensing. The AIMS model mainly monitors the degraded land which is rehabilitated with with the Food Assistance for Assets (FFA) program. This approach gives evidence based quantifiable impacts of policy and landscape changes. AIMS Model in development/pilot phase. Last public update was given in Oct 2018 for the 5 pilot countries.
510 Typhoon Model Netherlands Red Cross - 510 Coordination and context The Philippines Predictive model to identify high priority areas in the wake of a natural disaster. The model uses a large numbers of factors to assess this including but not limited to socio-economic factors, housing types, impact data and administrative boundaries. One fully validated model for Philippines available. Pilots in development to test model scale ability in Mozambique. Code repository, including all the source code scripts to run the model.
  • Modeling Details included in the press release
  • Output datasets and additional details for other typhoons such as Haiyan, Rammasun, Hagupit and Glenda also available
Mongolia Dzud model National Agency for Meteorology and Environmental Monitoring (NAMEM) Coordination and context Mongolia Each year, National Agency for Meteorology and Environmental Monitoring (NAMEM) publishes Dzud risk map which uses summer condition, pasture carrying capacity, livestock number, anomalous precipitation and temperature, snow depth, biomass, drought index, temperature forecast etc. to predict which regions may experience severe winter condition. Yearly Overview of Methodology
REACH flood susceptibility model REACH Coordination & Context: Affected areas Yemen A model is build using high resolution data sets to account for topographical features, normalized vegetation index (NDVI), rainfall, drainage, land cover and soil type. These factors are combined in a weighted linear fashion to compute an area's flood susceptibility (not risk). Static model updated last in June 2019
Global Displacement Risk Model IDMC, UNDRR, ETH Zurich Affected people Global Coverage The model combines probabilistic analysis of events recorded in national loss databases (retrospective risk) with probabilistic risk estimates for low-frequency, high-impact events (prospective risk). Prospective disaster displacement risk is expressed as a function of hazard, exposure and vulnerability: Risk = Exposure x Hazard x Vulnerability The prospective risk profile is calculated using hazard scenarios provided by a wide range of partners. These events represent all of the possible disasters that could affect a country over different return periods. For each grid cell in the exposure map, which has a resolution of between 1km x 1km in coastal areas to 5km x 5km inland, the expected impact of a hazard is calculated using vulnerability curves to determine an expected level of damage to homes and other structures based on the hazard’s intensity. Whenever the simulated damage to a home is more than 55 per cent, it is considered uninhabitable and the household displaced. Model developed in 2018 using from the 2015 Global Assessment Report ( ) Model
Pastoralist Livelihood and Displacement Simulator IDMC, Climate Interactive Affected people Kenya, Ethiopia, Somalia System dynamics (a modelling technique often used to analyse population dynamics and the behaviour of complex systems) is used to understand displacement dynamics of pastoralists in the Horn of Africa. The model formalises the causal relationships and drivers of pastoralist displacement, providing a decision support tool to explore the impact of specific policies and interventions. The model user interface is designed to answer a decision-maker’s “what if?” questions. Developed in 2015 Published research paper
Global Cholera Risk Model (GCRM) University of Florida and University of Maryland Affected people Global coverage The algorithm uses precipitation, temperature, and population data from different data sources to project the risk of a cholera outbreak 4 weeks later. The risk values represent the conditions under which Vibrio’s are more likely to grow. Cholera risk increases if a region experiences a two month above average air temperature, followed by one month of above average precipitation. These two conditions must be accompanied by damage to WASH infrastructure, further facilitating interaction of cholera bacteria with humans and thereby leading to an outbreak. Daily, weekly and monthly updates available
PRIO Conflict Prediction Peace Research Institute Oslo (PRIO) Coordination & Context: Conflict Events Global Modelling effort that predicts internal armed conflict in a country. The scope of the study ranges from 1970 to 2050. The model uses a wide range of predictors from conflict history, development indicators of the country, neighborhood stability. The model subsequently goes through extensive testing and validation exercises. Last update in Jan 2013 Model Håvard Hegre
ViEWS: Violence Early-Warning System Uppsala University - Department of Peace and Conflict Research Coordination & Context: Conflict Events Africa An early warning system to track and predict three classes of political violence. The model can predict armed conflict involving states and rebel groups, armed conflict between non-state actors, and violence against civilians. More importantly the model can disaggregate results into three different levels of analysis. We can analyse violence at national level, subnational level and actor level. The model takes into account a large array of input variable including but not limited to history of conflict, political institute, economic development, temporal factors, demography, natural resources and proximity of conflict. Every month
Conflict Early Warning and Response Mechanism (CEWARN) Intergovernmental Authority on Development (IGAD) Coordination & Context: Conflict Events Djibouti, Eritrea, Ethiopia, Kenya, Somalia, South Sudan, Sudan, Uganda Information of conflicts, migrants and natural disasters is shared between participating countries in a transparent manner. Thus information about potentially violent situations are jointly analysed to develop case scenarios and optimal responses. The analysis, data and response options are also communicated to all IGAD member states
CrisisWatch The International Crisis Group Coordination & Context: Conflict Events Global coverage Crisis watch is a "nowcasting" tool that is updated every month to get an overview of security and political developments across 80 countries in almost real time. The tool provides a bird eye view of ongoing conflict trends, escalations and developmental impact for all decision makers. 1 month
The Early Warning Project Simon-Skjodt Center and Dartmouth College Coordination & Context: Conflict Events Global coverage The statistical model of this project computes risk of mass killing and genocide by studying previous episodes (1945- present) to identify predictors and risk factors. Various models are cross validated and the best performing ones are then used to make predictions for the next two years. The output of a model is an estimated risk (of an onset of a mass killing) and the country's vulnerability rank. New predictions made every 2 years Methodology for Generating Statistical Risk Assessment
OCHA-Bucky Johns Hopkins University Applied Physics Laboratory, UN OCHA Centre for Humanitarian Data Affected people Completed: Afghanistan, Democratic Republic of the Congo, Iraq, Somalia, South Sudan, Sudan. The OCHA-Bucky model consists of a series of adjustments to an existing COVID-19 model (JHUAPL-Bucky) that was developed as part of the Centers for Disease Control and Prevention (CDC) COVID-19 Mathematical Modeling Forecasting Ensemble. OCHA-Bucky is a collection of coupled and stratified SEIR models. Since COVID-19 exhibits heavily age-dependent properties, wherein a majority of severe cases are in older individuals, SEIR models are stratified via the age demographic structure of a geographic region to obtain accurate estimates of case severity and deaths. Additionally, to model the spatial dynamics of COVID spread, we consider a set of SEIR sub-models at the smallest geographic level for which we have appropriate data. For a full description of the methodology please refer to the OCHA-Bucky methodology paper. The Centre is producing bi-weekly reports with projections and observations for each country. Code repository, including all the source code scripts necessary to run the model.
Imperial College London Imperial College London, WHO Collaborating Centre for Infectious Disease Modelling, MRC Centre for Global Infectious Disease Analysis, Abdul Latif Jameel Institute for Disease and Emergency Analytics Affected people Pilot : Great Britain Global coverage Stochastic individual-based simulation model that uses population movements between households, schools, workplaces, and the community to estimate transmission. Healthcare demand and the effects of non-pharmaceutical interventions are estimated. Key Results: Total number of infections; expected deaths; hospitalizations; scenario building 2 weeks Code repository, including all the source code scripts necessary to run the model. Patrick Walker Charles Whittaker Oliver Watson
Robert Verity study Imperial College London Affected people China A study of age-stratified data from mainland China that estimates duration from onset-of-symptoms to death or hospital discharge, and case fatality rate. Static Model Code repository, including all the source code scripts necessary to run the model. Robert Verity Lucy C Okell Ilaria Dorigatti Published research, Estimates of the severity of COVID-19 disease
COVID Act Now Georgetown University Center for Global Health Science and Security, Stanford University Clinical Excellence Research Center, Grand Rounds Affected people United States Open source model based on SEIR model used to provide state-specific projections of infection rates and ICU bed usage. Subnational projections for USA are reported. Dashboards for data visualization and APIs to explore all datasets available. Key Results: Total number of infections; expected deaths; hospitalizations; scenario building Data and estimates updated daily Code repository , including all the source code scripts necessary to run the model.
COVID-19 Projections Institute for Health Metrics and Evaluation Affected people Argentina; Belgium; Bolivia; Brazil; Bulgaria: Canada; Chile; Colombia; Croatia; Cuba; Cyprus; Czechia; Denmark; Dominican Republic; Ecuador; Egypt; Estonia; Finland; Germany; Greece; Honduras; Hungary; Iceland; Ireland; Israel; Italy; Japan ; Latvia; Lithuania; Luxembourg; Malaysia; Mexico; Netherlands; Norway; Panama; Peru; Puerto Rico ; S. Korea; Moldova; Romania; Russia; Serbia; Slovenia; Spain; Sweden; Turkey; Ukraine; United Kingdom; USA Revised estimates released every 2-3 days Model details
Multi-Model Outbreak Decision Support (MMODS) MIDAS Coordination Center Penn State University Affected people Global This is a decision theoretic framework that takes in results from multiple individual models to produce a unified set of projections and recommendations. By taking into account multiple modelling efforts decision makers can avoid cognitive biases, and insights from many different modeling approaches can be included. Static Model Katriona Shea
Reich Lab COVID-19 Forecast Hub MIDAS Coordination Center University of Massachusetts at Amherst Affected people United States The ensemble forecast dashboard combines multiple models; unconditional on particular interventions being in place with those conditional on certain social distancing measures continuing. 21+ different models with 4 week-ahead forecasts ahead are included in the ensemble and their weekly predictions are visualized on the dashboard. Data and estimates updated daily Code repository, including all the source code scripts necessary to run the model. Nick Reich Forecast Hub
Nowcasting model for real time epidemic tracking MIDAS Coordination Center Harvard T.H. Chan School of Public Health Centers for Disease Control and Prevention Affected people Global Nowcasting model here attempts to estimate the complete case counts for a given reporting date, using a time series of case reporting that is known to be incomplete due to reporting delays. The model uses Bayesian Smoothing, accounts for serial autocorrelation and gives output estimates that are smooth and accurate in multiple disease settings. Depending on input time series Code repository, including all the source code scripts necessary to run the model. Published research Nowcasting for epidemic tracking
Stochastic model for the transmission of 2019-nCov in Wuhan Center for the Ecology of Infectious Diseases, University of Georgia Affected people Wuhan, China Stochastic SEIR based model that includes additional time-varying rates of case detection, patient isolation, and case notification parameters. By using forward simulation, the model enables the generation of predictions about the future trajectory of the epidemic under alternative scenarios for containment. The model parameters have been calibrated using multiple different input sources. Key Results: Total number of infections; scenario building Static Model Code repository, including all the source code scripts necessary to run the model. John M. Drake Official Website
Scenario analysis for the transmission of COVID‑19 Center for the Ecology of Infectious Diseases, University of Georgia Affected people Georgia,United States Stochastic SEIR based model which is employed for location- Georgia USA for scenario building under various interventions such as increased social distancing, increased hygiene, against returning to normal. Key Results: Total number of infections; expected deaths; hospitalizations; scenario building Data and estimates updated daily Code repository, including all the source code scripts necessary to run the model. John M. Drake Andreas Handel
Gravity model for spread of 2019-nCov in China Center for the Ecology of Infectious Diseases, University of Georgia Affected people China Gravity-based model the risk of spatial spread of the 2019-nCov at the admin2 (prefecture) level in China, and to determine the efficacy of quarantines and travel restrictions imposed in Wuhan and other prefectures. Static Model John M. Drake Official Website
Probability of Widespread Transmission Center for the Ecology of Infectious Diseases, University of Georgia Affected people NA The model attempts to project the potential size of the COVID-19 outbreak by repurposing a previously developed Ebola model. This model can provide global ballpark figures of worse, worst, average case scenarios. Ongoing work on this model includes parametrization to have more region specific insight. Key Results: Total number of infections; scenario building Static Model
Mass Testing and interventions impact Center for the Ecology of Infectious Diseases, University of Georgia Affected people China This model assesses the impact of a symptom-based mass screening and testing intervention during a novel infectious disease outbreak in China. Static Model Code repository, including all the source code scripts necessary to run the model. Published research Assessing the impact of a symptom-based mass screening
Nowcasting the size of COVID 19 Center for the Ecology of Infectious Diseases, University of Georgia Affected people United States This model aims to nowcast the number of latent and symptomatic infections from a point in the past forward to the present. The authors accomplish this by fitting a time varying autoregressive model, and then forecast the fitted model curve using an exponential smoothing (ETS) model. The model works to determine the ratio of detected cases to the true number of cases i.e the ascertainment rate. Key Results: Total number of infections Data and estimates updated daily Code repository, including all the source code scripts necessary to run the model. John M. Drake Official Website
SARS-CoV-2 mortality during the early stages of an epidemic Institute of Social and Preventive Medicine, University of Bern Columbia University Imperial College London Affected people Hubei, China; Austria; Baden-Württemberg (Germany); Lombardy (Italy); Spain; Switzerland This study attempts to simulate the transmission dynamics of SARS-CoV-2 using publicly available surveillance data. The model is then used to infer estimates of SARS-CoV-2 mortality, adjusted for biases. It also examines the case fatality ratio (CFR), the symptomatic case fatality ratio (sCFR) and the infection fatality ratio (IFR) in different geographic locations. Key Results: Total number of infections Static Model Code repository, including all the source code scripts necessary to run the model.
Estimating unobserved SARS-CoV-2 infections in the United States University of Notre Dame Affected people United States This model aims to estimate the extent of community transmission of SARS-CoV-2 in the US that occurred prior to its widespread recognition. The extent of community transmission of SARS-CoV-2 is estimated using a stochastic simulation model that combined importation and local transmission processes. Static Model - Published working paper
Real-time forecasting of infectious disease dynamics London School of Hygiene & Tropical Medicine Affected people NA A Bayesian semi-mechanistic SEIR model of infectious diseases, that was used in real time during the 2013–2016 West African Ebola epidemic is presented. This model can be used for real-time forecasting of infectious diseases. Specific SARS-CoV-2 projections are not a part of this model at this stage. Static Model Code repository, including all the source code scripts necessary to run the model. Sebastian Funks Published research Real-time forecasting of infectious disease dynamics
Effect of travel restrictions on the spread of the (COVID-19) outbreak Northeastern University, Bruno Kessler Foundation, ISI Foundation Affected people Global The authors use a global metapopulation disease transmission model (GLEAM) to project the impact of travel limitations on the national and international spread of the epidemic. The model is calibrated on the basis of internationally reported cases. The outputs are multiple dashboards which track the number of cases and risk of virus importation Data and estimates updated daily Code repository and datasets.
Machine Learning the Phenomenology of COVID-19 Rensselaer Polytechnic Institute Affected people United States Data-driven machine learning analysis of the COVID-19 pandemic from its early infection dynamics, specifically infection counts over time. The goal of this model is to project the infectious force i.e the rate of a mild infection becoming serious. Additionally the model estimates asymptomatic infections and makes overall predictions of new infections over time Static Model - Malik Magdon-Ismail Published researchMachine Learning the Phenomenology of COVID-19
CORVID: Modeling layered non-pharmaceutical interventions against SARS-CoV-2 Institute for Disease Modeling, Patient Knowhow Inc. Affected people United States Covid is an individual-based model that simulates the spread of SARS-CoV-2 in synthetic populations that represent communities in the United States.This model can be used to understand the potential effectiveness of non-pharmaceutical interventions (NPIs) for both suppression and mitigation efforts. Static Model Code repository , including all the source code scripts necessary to run the model. Published working paper
CHIME (COVID-19 Hospital Impact Model for Epidemics) Penn Medicine, University of Pennsylvania Affected people United States The CHIME model and web app is designed to assist hospitals and public health officials with understanding hospital capacity in response to the COVID pandemic. The model outputs provide estimates of total daily and running totals of inpatient hospitalizations, ICU admissions, and patients requiring ventilation. The model is built using a SIR (Susceptible, Infected, Recovered) epidemiological model. Key Results: Total number of infections; expected deaths; hospitalizations; scenario building Data and estimates updated daily Code repository , including all the source code scripts necessary to run the model.
  • Michael Becker
  • Corey Chivers
CoMo Consortium - COVID-19 Web App Covid-19 International Modelling Consortium (CoMo Consortium) Affected people Global The CoMo model is a simulation based tool designed to emulate multiple scenarios,impacts of non pharmacological interventions, to make reliable estimates and predictions about the spread, hospitalizations and deaths related to COVID-19. There are a range of input parameter values to customize simulations. Although explicit modeling parameters have not been made public it is being used extensively by decision makers where explicit country based models do not exist. Key Results: Total number of infections; expected deaths; hospitalizations; scenario building 2 weeks Code repository , including all the source code scripts necessary to run the model. Official Website
EPIDEMIC FORECASTING: COVID-19 Future of Humanity Institute, University of Oxford Affected people Global The epidemic forecasting model and online tool can be used to calculate the estimated effect of various combinations of COVID countermeasures. Multiple parameters and compliance rates can be adjusted by decision makers. The output of the model is effective reproduction number which is the expected number of infections directly generated by one additional infection. This model works in conjunction with the GLEAM model (also in the catalogue) Key Results: Total number of infections; scenario building Data and estimates updated daily Jan Kulveit Website
Safety Nets Alert Platform (SNAP) WFP and local governments Food security Algeria, Armenia, Azerbaijan, Egypt, Georgia, Iran, Iraq, Jordan, Kazakhstan, Kyrgyz Republic, Lebanon, Libya, Moldova, Morocco, Oman, Russian Federation, Saudi Arabia, State of Palestine, Sudan , Syria, Tajikistan, Tunisia, Turkey, Turkmenistan, Ukraine, Uzbekistan, Yemen The Food Prices Early Warning is done using an indicator that monitors the extent to which a local food experiences abnormally high food price levels. Food price crises are correlated with food security crises. Early detection of rising prices supports decision making and early action. Early warnings allow WFP, governments and other partners to be aware of the latest trends of food prices and take appropriate actions to absorb any expected shock. The Shock Impact Simulation is done through a macroeconomic modelling system called SISMod that brings new possibilities to allow timely quantitative assessments on the ex-ante and ex-post impact of various types of shocks (market, economic, sociopolitical, climatic, etc.) on household vulnerability. It identifies and profiles vulnerable groups, estimating to what extent they are in need. The model outputs are focussed on early warning and shock impacts. 2015, updated monthly Hazem Al Mahdy
InaSAFE - Flood impact model Kartoza and Red Cross Red Crescent Climate Centre Coordination & Context Indonesia Impact based Forecasting - The impact based forecast functionality in the InaSAFE platform will generate an estimate of districts, number of people and houses likely to be affected in the event of a flood with certain return period. It uses GLOFAS, JRC, OSM, and WorldPop as main sources of data. The model has the potential to cover all river basins in the country, for the moment focuses on the ones where there are measuring stations. The lead time provided is between 10 to 3 days, as per GLOFAS forecast. The model is to be used by Indonesia Red Cross as part of its Forecast-based Action system for floods. 2019, to be updated in 2020 when no data is available Code repository Official Website
CALM, the Cholera Artificial Learning Model The International Genetically Engineered Machine (iGEM) Foundation Affected people (Health) Yemen Cholera Artificial Learning Model (CALM) creates a system of four extreme-gradient-boosting (XGBoost) machine learning models that, in tandem, forecast the exact number (with an error margin of 4.787 cases per 10,000) of cholera cases any given Yemeni governorate will experience for multiple time intervals ranging from 2 weeks to 2 months. Details about features engineering, model tuning and results are publicly available. Static Model Code repository Rohil Badkundri
Cholera Prediction Modeling System University of Florida National Aeronautics and Space Administration Affected people (Health) Yemen This model uses satellite and ground-based data to forecast the risk of cholera in Yemen and other countries. The map linked alongside shows the forecasted risk of cholera in Yemen from August 10 to September 6, 2020. It was created with the Cholera Prediction Modeling System, which incorporates NASA precipitation data, air temperature data from NASA’s MERRA-2 reanalysis product, and population data. The number of cholera cases could increase in coming weeks, influenced by heavy rains that usually fall in August, though researchers predict the outbreaks should be limited to a few hotspots unless there is a large population displacement. The model got its first real-world test in 2017, and achieved 92 percent accuracy in predicting areas where cholera appeared that year. Static Model Antarpreet Jutla
Using Predictive Analytics to Identify Children at High Risk of Defaulting From a Routine Immunization Program Child Health and Vaccines program, Interactive Research and Development (IRD) Affected people (Health) Pakistan This predictive analytics algorithm is designed to identify the children who are likely to default on subsequent immunization visits. 47,554 longitudinal immunization records, which were classified into the training and validation cohorts. Four machine learning models (random forest; recursive partitioning; support vector machines, SVMs; and C-forest) were used to generate the algorithm that predicts the likelihood of each child defaulting from the follow-up immunization visit. Each of the models was assessed in terms of accuracy, precision (positive predictive value), sensitivity, specificity and negative predictive value, and area under the curve (AUC) and is seen to perform reasonably. This model demonstrates that predictive analytics can accurately identify children who are at a higher risk for defaulting on follow-up immunization visits and could be targeted for additional interventions. Static Model Published peer reviewed research on feasibility study for the project
Platform for Real-time Impact and Situation Monitoring (PRISM) World Food Program Japan Government USAID Food Security & Nutrition Global - Indonesia, Sri Lanka, and Cambodia, while deployments to Mongolia and Afghanistan are forthcoming PRISM enables WFP’s partners in government and WFP country offices to access the latest available climate hazard information alongside vulnerability data through an intuitive, map-based dashboard. PRISM combines information from satellites and other remote sensing sources with WFP data on vulnerability to create actionable climate information for decision makers, allowing them to prioritize assistance to those most in need.PRISM is capable of producing impact-based forecasts: converting climate information into programming which focuses on the socioeconomic impact of a hazard. Daily Updates
Managing Risk through Economic Development (M-RED) Mercy Corps Coordination and Context (Affected Areas) Nepal, Indonesia, and Timor-Leste M-RED will use drone technology to measure and monitor land to take more effective preemptive measures against land degradation and floods in Nepal (pilot). By observing targeted areas, the program should accurately detect and track land use changes. The data and predictive analytics are used to model floods and track shifting rivers and are designed to will benefit target communities. Model performance and details are documented but code is not public. Static Model
Predicting Demographic Trends for Global UNHCR Persons of Concern UNHCR Affected people Global The Demographic Projection Tool (DPTool) used uses data that are already available to UNHCR field operations. This model uses the demographic balancing equation to simulate different scenarios about the future size and composition of the refugee population.The tool at the moment only works well if the contextual conditions of a humanitarian emergency remain the same. The DPTool is better-suited for short-term projections in its current iteration. The tool also continues to rely heavily on UNHCR personnel who must define the parameters of the model based on their expert knowledge of the situation on the field. Static Model
Modelling Early Risk Indicators to Anticipate Malnutrition (MERIAM) Action Against Hunger, the Graduate Institute of Geneva, John Hopkins University, and the University of Maryland. Food Security & Nutrition Kenya, Niger, Nigeria, Uganda, Somalia MERIAM deploys a complementary combination of innovative methodologies – both econometric and computational modelling – and leverages a variety of existing and accessible data sets to rigorously capture causal factors of acute malnutrition. Taken together, these aspects allow the project to dynamically model the fluctuation of acute malnutrition in contexts where this information is most urgently required. By better understanding the leading risk factors for acute malnutrition, MERIAM can forecast who may be most at-risk of becoming wasted, when they are likely to become wasted, and where (in what geographical area they reside). Model in development Marcello Malavasi
Cooper/Smith population mobility model Digital Impact Alliance (DIAL), Cooper/Smith Affected people, Populated places Malawi In 2020 Cooper/Smith developed a population mobility model in Malawi with the support of the Digital Impact Alliance (DIAL) and the Malawi Ministry of Health (MoH). The mobility model provides insight on population localization and movement patterns, enabling the MoH and local stakeholders to have a near real-time view on where the population is, providing a much-needed complement to the census. daily N/A
  • Guillaume Foutry, Program Manager for West Africa
  • Cassie Morgan, Sustainability Coordinator, Kuunika Project, Malawi
  • Coopersmith
Final report of the peer review process conducted by the Centre according to the OCHA’s Peer Review Framework.