Centralized simulated annealing for alleviating vehicular congestion in smart cities

Vehicular traffic congestion is a serious problem arising in many cities around the world, due to the increasing number of vehicles utilizing roads of a limited capacity. Often the congestion has a considerable influence on the travel time, travel distance, fuel consumption and air pollution. This paper proposes a novel dynamic centralized simulated annealing based approach for finding optimal vehicle routes using a VIKOR type of cost function. Five attributes: the average travel speed of the traffic, vehicles density, roads width, road traffic signals and the roads' length are utilized by the proposed approach to find the optimal paths. The average travel speed and vehicles density values can be obtained from the sensors deployed in smart cities and communicated to vehicles and roadside communication units via vehicular ad hoc networks. The performance of the proposed algorithm is compared with four other algorithms, over two test scenarios: Birmingham and Turin city centres. These show the proposed method improves traffic efficiency in the presence of congestion by an overall average of 24.05%, 48.88% and 36.89% in terms of travel time, fuel consumption and CO2 emission, respectively, for a test scenario from Birmingham city in the UK. Additionally, similar performance patterns are achieved for the a test with data from Turin, Italy. •Proposed approach reduces the traffic congestion problem in smart cities.•The travel time, travel distance, the fuel consumption and the CO2 emissions have been reduced by using the CSA-VIKOR.•Proposed approach helps to obtain information about the wider road network can be embedded into the system.•Using multiple criteria in the cost function gives the best average travel speeds by finding the optimal paths.•The CPU time has increased slightly due to collects a large volume of information at each RSU.