Residents photo-report overflowing bins from their phones. Staff get a dashboard that corrects for reporting bias and ranks the best places to add bins, shown as estimates and never as measured outcomes.
Orlando, FL · ports to NYC & Chicago by config
Illustrative: blue = expected overflow, pink dashed = under-reported
What-if: 20 bins, estimated change in demand within 150 m
How it works
01 · Report
Residents photograph an overflowing bin. The photo is classified, scored for severity and merged with reports of the same spot, so staff see one ticket and not ten.
02 · Rank
Hexagons score expected overflow from places, transit and nearby bins.
03 · Place
Greedy picks the site that serves the most demand next, with a reason for each.
Why it matters
Two areas can have the same overflow and very different reporting. Bin-It flags the quiet one, so the plan does not only serve the neighborhoods that already speak up.
Expected overflow
Reports
Expected overflow
Reports
Illustrative bars for the idea, not measured data.
cities, one config file each
hexagons scored
ranked bin sites per city
photo classes
Cities
Grids are illustrative, not real scores.
Estimates, not measurements
Orlando has no confirmed 311 sanitation dataset, so the model is trained on New York and Chicago and applied to Orlando. That transfer is only lightly validated. Every projected change on the dashboard is an estimate.