The maintenance gap

Chicago's 311 register tracks resident-filed requests across the city's seventy-seven community areas — the stable socio-demographic geography that pre-dates the city's redrawn political wards. Three of those request types are unambiguous physical-maintenance work: an alley light out, a street light out, and a pothole in the street. The Department of Streets and Sanitation and CDOT own the response. The work is the same on the north and south sides of the city; the timing isn't.

The map below shades each community area by the weighted-average number of days the city takes to close one of those three maintenance tickets, since January 2024. Areas with fewer than fifty closed cases are excluded — the average isn't stable enough. Darker fill is slower.

The next map drops the city's direct affordable-housing footprint on top of the same seventy-seven polygons — every municipally subsidized, deed-restricted rental unit the Department of Housing tracks, in cobalt. Multiply-blended with the lag layer in brick, the darkest areas are the ones high in both measures. This is the question a two-stacked-maps approach asks the reader to do in their head; the bivariate form does the overlay for you.

Many of the names that surface in both lists overlap. The community areas that carry the heaviest concentrations of subsidized units are concentrated on the South and West sides — Near West Side, Grand Boulevard, Douglas, Oakland, Washington Park. The streetlights stay out longer in some of those same areas. The potholes stay open longer.

The honest version of this story is structural: we have two open datasets keyed on the same geography, and putting them next to each other surfaces a pattern worth investigating with harder methods. Drill into the underlying data on the dataset page — the curated views or the SQL playground will let you slice by sr_type, ward, or zip code, and isolate which of the three maintenance categories is driving the lag in each area.