The World Bank Group
GFDRR Labs – Unbreakable Resilience Indicator
As part of an ongoing, collaborative relationship, our partners at the World Bank wanted to redesign the Unbreakable Resilience Indicator interactive tool with which users can explore data from when the World Bank released Unbreakable: Building the Resilience of the Poor in the Face of Natural Disasters, a flagship report that examines how natural disasters affect people’s well-being and socio-economic resilience.
The result of Graphicacy’s partnership with the World Bank’s Unbreakable Global Facility for Disaster Reduction and Recovery (GFDRR) team is an engaging, intuitive data visualization platform that allows users to explore multiple data points by country or policy area to estimate the benefits of investing in resilience to natural disasters. For an audience of officials and practitioners who deal with natural disasters, the tool allows users to better understand how the term resilience is used, and which investments or policies build resilience based on data.
Graphicacy’s refined design provides customized tracks for general users who want to explore fundamental data points, and power users who want to manipulate the data. The design displays information and charts that are packed with data points without resorting to a long scroll.
The intention of the redesign was to reuse existing design elements and functionality to create a more user-friendly experience via an efficiently arranged layout. For example, instead of having the list of data points open up vertically, we compacted them with more elegant and user-friendly accordion tabs.
Our team also dedicated our efforts to re-thinking the chart types. We partnered with the GFDRR team to arrive at a chart selection that helps the readers more efficiently understand and contextualize the data that is presented and compared.
Our team was able to build an efficient, real-time encrypted pipeline for users to create their own scenarios around how certain country factors affect resilience in countries. As a user changes a Social Protection factor, such as “[R]eadiness to scale-up social protection,” the application hits a machine learning trained Python model to update how this shift will contribute to resilience in the selected country, as seen below. This process provides an almost infinite amount of exploration for a user who is trying to identify the optimal drivers in the evaluation of a policy scenario.
From an engineering perspective, the Unbreakable application builds on one of our core strengths: using data visualization to help explain the output of a data science model.