Bridge is an application built as a part of the Covid'19 Spaceapps Challenge. With this application, we're trying to mitigate the impact of Covid'19 on UN Sustainable Development Goals(SDGs).
To build a web application which can eradicate the food security problem in hard times like Covid'19 by using a ML Recommendation Engine.
According to USDA, 11.1% of households were food insecure at least some time during the year 2018 which is roughly 36.4 million Americans. On the contray, US has 897,400 (100 acres) of farmlands and 2,023,400 number of farms as of 2019, stated in a report by NASS. And according to another report, In 2018, Americans wasted around 30%-40% of the entire US food supply.
To understand the scenario, let's google something -
Connecting the dots based on the above statistics, one of the reason could be absence of a bridge to link the supplier and buyer in an optimized way leading to wastage of food at one region and food insecurity in another region.
And if this continues, by 2050 when the demand will be 60% higher than today, not to mention the scenario will get worse.
Barely-available Resources' Inter-region Distribution GuidE (BRIDGE) is a smart web application which leverages the power of machine learning to identify potential hotspots around the world where there is a rise in demand for resources(food/medical/human) but are not met. Simillarly hotspots are identified where excess production is made without a potential consumer. Our brief project motto is to bridge this divide and aid the government in empowering every single individual to achieve more.
BRIDGE follows the following steps to mitigate the issue-
- NASA Satellite Images are processed using Image Processing tools to determine county-wise cropland data and crops production.
- Two indexes are then calculated for every resource for a particular region, DemandIndexForResourceA and SupplyIndexForResourceA where an internal algorithm that takes in a ton of parameters like
Agricultural harvest of the region, GDP of the region, Population to Harvest ratio, Resource Cost at the region etc.... - After training the model, we prepare one recommendation engine which gives us the Top 5 potential supplier for a buyer region.
- These data and statistics are then showed in an Angular based Dashboard.
