Efficient Traffic Routing in Smart Cities to Minimize Evacuation Time During Disasters

Abstract

AbstractEfficient traffic management during disaster evacuations is an essential component of intelligent transport systems in smart cities. In a natural disaster, a surge of vehicles from dense residential areas may simultaneously move towards the same nearest safe shelter following a shortest path for each individual vehicle, thereby often leading to congestion and resulting in increased evacuation time. In this paper, we consider time-optimal traffic distribution in such disastrous situations considering a Manhattan grid network of roads. Several research results on optimal-time traffic distribution in such a network exist in the literature, all of which consider a restricted scenario of a single safe destination at a corner point of the grid. In contrast, we describe a technique for minimizing average travel time of the vehicles assuming a general situation as experienced in real-life, where the destination node can be anywhere on a rectangular \(m \times n\) grid network with multiple sources of traffic injection. Simulation results using SUMO on a road network of Manhattan borough of New York city show that our proposed technique outperforms the existing techniques on dynamic traffic assignment in terms of average travel time.KeywordsTransportation networkRoute planningIntelligent transport system (ITS)Queuing delayEvacuation problemCongestion control