Data-driven PSO optimization of a smart-city metro line

The object of this paper is the data-driven optimization of public transport lines in smart-cities. This paper provides a solution to get the best line for urban public transport line planning based on public transport data, urban street data, and population data. Chapter 2 introduces the velocity score and sociality score to show the excellent degree of PTN, they can be understood as indicators of city transport accessibility and public transport operation capacity based on population distribution respectively. Also, a scoring model based on the principle of isochron is used to evaluate and calculate the sociality score of the city public transport network (PTN). Chapter 3 of this paper introduces the particle swarm optimization algorithm(PSO) and what kind of problems can be solved by PSO algorithm. In chapter 4, the greedy strategy is proposed to optimize the line by changing the position on the stops in line, and improved PSO is introduced. According to the sociality score of PTN, greedy strategy is adopted, and improved PSO algorithm is used to optimize the longitude and latitude position of the station of the new line to seek the maximum sociality score of city PTN after adding this line, so as to have the best operation ability based on population distribution. This paper optimizes the Metro D line under construction in Rome. By adjusting the location of the stops of Metro D in the planning, the maximum sociality score can be obtained as far as possible. By comparing the urban sociality score before and after optimization, we have achieved 10% progress. This paper confirms the role of improved PSO as an optimization algorithm in solving the line planning problem and gives a complete strategy that tends to do the line planning according to the population density distribution and PTN. The research significance of this paper is that the line planning scheme combined with scoring model and optimization algorithm can help city line planning to be data-driven to provide more schemes for engineers and planners reference, for the final line construction site selection to provide a data-driven scientific basis.