Optimization Model for Emergency Shelter Location Selection

Inspiration

• We want to do some real-world application of SVI index given the comprehensive data and infinite possibility this dataset entails.   ## What it does
• We want to figure out how to minimize the suffering of people after a disaster by pinpointing the best shelter location selection. So, we established a math model that to the greatest extent help people in need based on SVI index and ensure that the shelters are built in relatively safe locations based on FEMA dataset.   ## How we built it
• We used Tableau to construct the frequency maps of various natural disasters across all counties in US since 1954 based on the FEMA dataset. We constructed the SVI *Population map graph by multiplying the E_TOPTOP((Population estimate) and F_TOTAL (Sum of flags for the four SVI themes). We allow users to input the counties affected by a real-world disaster and assign each county with a weight according to its SVI * Population value (greater the population or/and SVI index, greater the weight). We employed RANSAC linearization model to calculate the optimized location of shelter that to the greatest extent help the people in need. This location is transferred to the disaster-frequency map to inform the users the safety of the location in the face of a particular disaster. We allow users themselves to determine the balance between the location that help most people in need and the location that is the relatively safe in a particular disaster.   ## Challenges we ran into
• We encountered great difficulties in creating an interactive map that accomplishes all the tasks as it requires as to connect python, c++, javascript, tableau, and flask. Due to the limited time available, we decided to concentrate on improving our math model instead. We want our system to achieve more accurate result.

Accomplishments that we're proud of

• We successfully built a website and do some manipulations on the map, such as map ordered and coordinates on the maps.   ## What we learned
• The ability of collaboration. The ability of searching information. The ability of learning new and unfamiliar knowledge in a short period of time.   ## What's next for Optimization Model for Emergency Shelter Location Selection
• In the future, we can do more better on user's interaction, and consider more variables like geological features in the real world.

Built With

• c++
• flask
• javascript
• python
• tableau