Smart Signals – Intelligent Traffic Management Prototype
30% reduction in wait times • 12,000+ hours saved annually
The Problem: Ahmedabad commuters wasted significant time at inefficient traffic signals—empty intersections had long green lights while busier ones were under-served, causing delays, fuel wastage, and frustration.
The Solution & Approach:
Validated problem: Interviewed 40+ commuters and 3 traffic officials to quantify pain points; surveyed traffic patterns at 2 key intersections.
Strategic trade-offs: Chose camera-based detection over manual retiming (real-time + scalable) and loop sensors (lower upfront cost, 3x maintenance burden).
Execution: Coordinated 6-member cross-functional team (2 developers, 1 hardware engineer, 3 testers) through prototype build and validation.
Stakeholder alignment: Secured pilot approval from Ahmedabad Traffic Police by addressing privacy concerns—processed all data locally without storing footage.
Iteration & Learning:
Initial loop sensor approach proved impractical; pivoted to cameras after cost-benefit analysis.
Early tests improved wait times by 15%; refined detection algorithm to reach 30% reduction.
Impact & Validation:
Achieved average wait time reduction: 30% (87s → 61s) over 200+ signal cycles across 2 weeks, extrapolating to ~12,000 hours saved annually per intersection.
Estimated fuel savings: ~5,000 liters/month across 2 pilot intersections.
₹2L SSIP grant (2018) awarded from 200+ competing projects, validating technical feasibility and policy relevance.
Presented at LJ Innovation Village 2018 and 2019 and LJ Startup Week 2018.
City-scale extrapolation: 100 intersections = 1.2M hours saved, ₹15 crores fuel annually.
Reflective Insight – If I Were Building This Today: Design a municipal revenue-sharing pilot (% of fuel savings) and build A/B testing infrastructure to iterate detection algorithms 10x faster—accelerating path to city-wide deployment.
₹2L SSIP grant (2018) awarded from 200+ competing projects (view grant certificate)
