Origin of the Project
The Brihanmumbai Municipal Corporation (BMC) manages waste collection for over 20 million residents through a network of routes serviced by dedicated vehicles and crews. The Chembur (M-East) ward alone handles roughly 30 sanitation workers spread across 10 active collection routes, each with a defined start point, end point, and a target staffing level set by the ward office. Through field visits to that ward office and direct interactions with Solid Waste Management (SWM) supervisors and sanitation workers, three operational pain points emerged repeatedly:
- ·Attendance fraud - workers signing the register at the ward office without reporting to their assigned routes.
- ·Late detection of route failures - supervisors had no way to detect understaffing until collections were already missed.
- ·Idle workers on completed shorter routes while adjacent longer routes remained understaffed.
SWRMS was scoped specifically against these three pain points rather than as a general-purpose municipal platform. Every feature in the system maps back to one of them: geo-fenced check-in and face-verified field photos address attendance fraud; live GPS tracking and automatic deviation alerts address late detection; and the workforce reallocation engine, surfaced on the supervisor dashboard, addresses the idle-vs-understaffed imbalance during a shift.
The Solution Architecture
Geo-fenced attendance verification ensures workers can only mark attendance when physically within 200 m of their assigned route start point, calculated using the haversine formula against coordinates stored on the route record. AI-verified field photos use 128-dimensional face embeddings to confirm worker identity at shift start, checkpoint, and shift end. Routes are road-snapped via OSRM so deviation detection runs against the actual street network rather than a straight-line approximation. Real-time route progress tracking and automated reallocation suggestions complete the loop.
On top of the core verification flow, the system layers a multilingual staff PWA (English, Hindi, and Marathi) for low-literacy field use, a tamper-evident audit log capturing every supervisory action with actor identity and IP address, an opt-in browser push channel that alerts supervisors within seconds of a route deviation, mock-GPS attempt, or missed shift, and a workforce reliability score per worker that aggregates 30 days of attendance signals into a single 0–100 figure for the ward office's monthly review.
UN Sustainable Development Goal 11
Sustainable Cities & Communities - Target 11.6
SWRMS contributes to Target 11.6 by improving accountability and route completion rates in municipal solid waste collection - directly addressing the per-capita environmental impact of waste accumulation in densely populated urban areas. The system makes the daily collection cycle measurable and auditable, which is the prerequisite for any subsequent operational reform.
Acknowledgments
This pilot was developed by the Department of Information Technology at Vivekanand Education Society's Institute of Technology (V.E.S.I.T) under the guidance of Prof. Archana Kshirsagar, with field access generously provided by BMC officials and Solid Waste Management staff at the Chembur Ward Office. The work draws on direct shadowing of supervisors during morning shift and on the Department's own operational records for the ward; no dataset was synthesised without ground-truth counterparts. The codebase, route definitions, anti-fraud thresholds, and audit schema are openly documented so the system remains transparent to BMC reviewers and available for adaptation by other wards.