Our NGO supports vaccinations that save millions of children’s lives each year – but many more still need immunisations.
A non-governmental organisation (NGO) that funds vaccinations for children in lower income regions. Each year, our NGO spends millions to get vaccinations to where they are needed the most. Since it launched, our NGO has supported immunisation of more than 580 million children and averted more than 8 million premature deaths.
Each year, funds have to be allocated across regions, countries, and types of immunisations.
The vaccination strategy has to focus both on short and long term goals - immediately saving lives and pursuing sustainability.
Immunisation funds are targeted at regions where there would otherwise be an undersupply and lives are at stake. Immunisations for diseases that pose an immediate risk to children are targeted. Specific factors play a role:
- some immunisations require multiple doses at separate times
- some diseases are on the brink of eradication
Generate data insights to effectively support a worldwide vaccination effort.
Use cutting-edge analytics to come up with a strategy that ensures funds for vaccinations have maximum impact.
Each year, our NGO has to allocate its budget across immunisation types and countries. This allocation of funds plays a crucial role in determining how effective the immunisation initiative is (i.e., how many lives are saved as a result of it). Our NGO has approached you, a team of independent analytics consultants, to help them plan their investments for the next 5 years. They are looking for a recommendation for how much to invest in what immunisations and what countries. Your model should depend on the following parameters:
- Objective: save as many children‘s (< age 10) lives as possible
- Budget: USD 1,000,000,000 (1 billion) per year
- Time horizon: 5 years
- Countries: all countries where our NGO has previously funded immunisation programs
- Immunisations: all types of immunisations that our NGO has previously funded
- Interactive dashboard (or similar) illustrating NGO’s disbursements since 2’000. Depending on your skill and time, this could be a pivot-table-like product and/or a fancy visualisation using a world map.
- A final budget allocation recommendation based on generated data insight. This will be presented to the judges and can be supported by a PowerPoint presentation or (ideally) scripts/visualisations/apps that your team built during the Hackathon.
You will be relying on data sets that are open to the public – and are you free to add your own data (if openly available)
- Birth rate, death rate, life expectancy, fertility rate, …
- Prevalence of obesity, malnourishment, vaccine waste
- International migrant numbers, net migration, refugee population by country or territory of origin
- GDP, poverty headcount
- Air pollution, improved sanitation facilities, water source
Ideas for supplementary data:
- Number of reported cases of Haemophilus influenzae Disease (Including HiB)
- Forecasts of epidemics
- Forecast of destabilizing elections
- GDP growth (meaning country can found own effort)
- Data on innovations for new ways of delivering vaccines in difficult circumstances
- etc.
Do not get overwhelmed by the data and remember to enjoy yourselves!
- Half the work is formalizing the problem. You will have to make assumptions. Make sure you explain why you made them.
- You don’t have much time so keep your analysis simple. You will need to discard many of the variables in the various data sets we give you.
- We are dealing with a serious topic but after all this is just a hackathon. Try to wow us and your competitors with cool visualisations and solid analysis. (Remember however: the recommendation is the final product of your work.)
- You can find inspiration from this talk on the importance of visualizing data here
Each team will present their work in front of a group of judges
- 5 min pitch
- 5 min QA
- Introduce team
- Problem Statement
- Presentation of solution
- Demo
Use any of the tools listed below or your own choice, provided the tool is available for free.
All code used must be open sourced. It can be existing code, new code written for the hackathon project, or a combination, but all code must be open sourced and licensed appropriately.
- QlikSense
- Tableau (trial available)
- GapMinder (link)
- PowerBI (link)
- R – Shiny, Plotly
- Python – Pandas, Bokeh
- Matlab
- SQL – Backend MS SQLServer
Github to download dataset (link)
Slack to bring communication together so that people can share their insights (link)
A number of clearly defined criteria will be used in the evaluation of your work. Each team’s work will be evaluated as a whole.
- Data indentification and integrity
- Analysis is relevant and results are easily reproducible
- Code is well-structured and commented
- Clearly state hypothesis
- Sophisticated analysis is used where appropriate
- Direct attention to the key message
- Appropriate chart type
- Accurate units of measurement
- Keep it clear
- Visuals are appealing
- Provides clear and well-founded recommendations
- Solves the initial business problem with actionable recommendations
- Is well-structured
- Tells a compelling story
- Is within the time limit
Good collaboration within the team will be key success factor.
Each participant will be part of a team of 2-4 people. People are free to team up as they wish (we recommend a good mix of skillsets in the same team: technical, business and presentation skills).
We suggest that you look for the following qualities in your team:
- Experience building small apps
- Experience with data visualisation
- Experience with statistics/machine learning
- Experience with business presentations/pitches
Friday | |
---|---|
9h – 10h15 | Introduction |
10h30 – open ended | Start of hackathon |
11h – 15h | Food (ongoing) |
19h – 22h | Bar opens & Apéro |
Saturday | |
---|---|
10h – 10h30 | Welcome breakfast |
10h30 – 12h | Presentation of results |
19h – 22h | Wrap up |