Comparison Of Three Accessibility Measurements Of Transit Equity With Gini Coefficient

In the context of increasing reliance on public financial support,the equalization of public transport service has become an important issue that the government cares about and urgently needs to solve.With the goal of improving service quality and balanced resource allocation,it has become a hot topic to study the scientific and reasonable evaluation theories,methods and applications of public transport service equity.In this paper,the buffer method,network ratio method and distance decay method are selected as three common and feasible methods to calculate the accessibility index of bus stop to express the difficulty of passengers to get to the stop,the radiation radius of the site is selected to be 400 m,and the Gini coefficient and Lorentz curve are combined to analyze the decomposition difference of the Gini coefficient of the three methods.In addition,the GWR model is used to analyze the reasons for the Gini coefficient decomposition differences of public transport by selecting social attributes(population,population density,traffic analysis zone area)and environmental variables(road network length,road network density,stop density).The results show that: compared with the buffer zone method,the Gini coefficient of the network ratio method is 25.33%,1.52% and 3.12% lower than that of the urban area,urban area and rural area respectively.The distance decay method was 33.33%,3.03% and 6.25% lower,respectively.Decomposition with accessibility of three kinds of calculation method of Gini coefficient is found that the differences between urban and rural overall unfairness contribution rate were 53.79%,33.93% and 20.00%,urban internal differences contributed respectively 28.24%,15.20% and 22.00%,rural internal difference contribution rate were 8.38%,35.37% and 34.00%,cross terms of contribution rates were 9.59%,15.50% and 24.00%.Then,GWR model was used to analyze the influencing factors of the results of the three accessibility methods,and it was found that the length of road network and the area of traffic plot significantly negatively affected the accessibility of bus stop,while the remaining variables significantly positively affected the accessibility of bus stop.Further "sensitivity analysis" of the independent variables in the model shows that the increase of road network density and site density first increases the calculation results of the three accessibility methods,and then leads to the reduction of Gini coefficient and the reduction of the decomposition results of each part of the Gini coefficient.Among them,the Gini coefficient of city area and rural area decreased the most,urban area the least.On the impact of the decomposed results,the buffer method calculated the difference between urban and rural public transport(14.29%,16.67%,23.81%,16.67%)was greater than the difference between urban and rural public transport(7.32% and 14.63%),and the difference between urban and rural public transport(15.79% and 21.05%)was greater than the difference between urban and rural public transport(11.11%,10.00%,5.36% and 10.00%).The distance decay method shows no regularity.The results show that: considering the difference of the influence of three accessibility methods on the Gini coefficient of public transport,it is beneficial to evaluate the fairness of public transport objectively.while the density of road network and the density of stations are the factors affecting the decomposition of Gini coefficient of public transport between the three methods.Therefore,it is necessary to carefully consider the selection of accessibility calculation method as the data source of Gini coefficient evaluation of public transport,or it will lead to the deviation of equity evaluation results.