Incentives and Games in Large Scale Mobility
summary
Cellphone and GPS traces let origin destination inference replace slow surveys using convex programs that reconstruct the Hamilton‑Jacobi value function from sparse data, tested with 3s GPS at 2% sample density and validated against video ground truth. Public providers have up to 40% cell coverage and urban corridors like I‑10 operate with ~450 control assets, so routing apps add a fast feedback loop; simulations and a Nash‑Stackelberg model show ~20% app use can increase total delay and shift congestion onto neighborhoods.
AI-written summary · about
chapters (6)
- 0:00Start
- 4:28Optimization and control Algorithms for traffic flow control and optimization Playbooks among scenarios
- 4:38User interface Human in the loop for infrastructure control - Limited ability to actuate pre-planned scenarios (system-wide)
- 9:28Results Algorithm produces distribution of flows along the different paths - Approach is data driven, dous not make (Nash, UE, Social Optimum, etc.)
- 18:03The impact of traffic apps on system dynamics Fundamental promise of routing services
- 24:43Example for 3 miles in Pasadena Impact of increased app use for through traffic - Immediate massive reroute through Pasadena (2 reroutes) Travel time in Pasadena increases by 17%
related talks
01h 03mUnderstanding and Shaping Urban Mobility Using Dollars and Data II
01h 08mUnderstanding and Shaping Urban Mobility Using Dollars and Data I
34mData Drives Business
19mDigital Trace Data to Measure Human Mobility
21mExpert Panel | Going Places: Location Intelligence for Transport and Logistics
41mPredicting the Traffic Jam Congestion Aware Routing
29mDon't Get Distracted
24mBuilding Aggregated Solutions to Scale Your Geospatial Strategies
