Assessment Method And Optimization Of Spatial Equity Of Urban Health Service Facility

Reasonable allocation and distribution of urban health service facility are basic foundations of city's sound development and citizen's high quality of life.However,various factors,such as the complex relationship between supply and demand,difficult access to detailed data and low universality of assessment model bring great difficulties to the assessment of spatial equity,causing inadequateness of assessment system,low accuracy,poor reliability and applicability of assessment result.Accordingly,the absence of reasonable assessment method will inevitably affect the effectiveness of optimization method.Therefore,effective assessment of spatial equity and optimization method are of great significance in the planning of public service facility and development of spatial analysis method in geographic information system.Spatial equity assessment and optimization method of public facilities are still in fledging period at present,and the research method needs further exploration.This dissertation applies various dimensions,including the integrations of micro and macro,time and space,single transportation mode and multi-transportation mode,to analyze and estimate the spatial equity of health service facility in Nanjing central area.Furthermore,this dissertation builds spatial optimization model under the maximum equlity goal to solve technical difficulties of traditional spatial optimization methods which emphasize efficiency but neglects equity.The primary contents and conclusions are as follows:(1)Based on the attribute information of spatial event point,this dissertation proposes one method of marked value network kernel density estimation.The network cold and hot spot areas which are statistically significant can be explored under the application of integrating marked value network kernel density estimation method and Getis-Ord Gi*statistical method.This dissertation uses first-order and second-order method of network point pattern analysis theory to assess the network clustering characteristics of health service facility in Nanjing from different areas,categories,general clustering characteristics and spatial relationship.Results show that traditional network kernel density estimation tends to misjudge the intensity effect of event point,while the result of marked value network kernel density estimation is more in line with reality.Generally,the health service resources in the study area are overly concentrated in city-wall areas,where the majority of the roads is hot network segments with a 0.99 confidence level.From the perspective of the spatial equlity distribution of health service facilities,tertiary hospitals show the worst spatial equality,and small hospitals show the bad spatial equality,and secondary hospitals show the best spatial equality.In addition,retail pharmacies are obviously clustering with health service facilities in network space.(2)This dissertation analyzes the influence of time and transportation mode on the spatial equity of community health service centers.Using open geographic database based on Door-to-Door approach,this dissertation makes a systematic analysis on the nearest accessibility difference of community health service center under different moments and transportation modes within a day.What's more,Gini coefficient is introduced to estimate how the general spatial equity of community health service center changes along with time and transportation mode.It is concluded that the accessibility and general spatial equity of community health service centers fluctuates radically along with time under the dynamic changes of traffic flow and opening number of community health service centers,among them opening number is more significant.The spatial equlity result is very different when using different transportation mode.Riding have more advantages in the main urban area and nearby the community health center of sub-urban areas.Driving have more advantages in sub-urban areas.There are few areas where public transport takes advantage.The general spatial equity of community health service centers is assessed as "relatively reasonable"using public transport and driving,while the result is "poor" using riding.The results of general spatial equity of community health service centers tend to be overestimated if the Door-to-Door approach of driving is not adopted.(3)This dissertation proposes a Gaussian two-step floating catchment aproach(Ga2SFCA)which integrates multiple transportation modes with traditional accessibility model.The potential accessibility and accessibility Gini coefficient of large hospitals are calculated using the multi-mode Ga2SFCA method.And the differences between the results and other 3 single transportation modes' results are analysized.It can be concluded that the multi-mode Ga2SFCA method solves the problem of hard integration of traditional accessibility methods with citizen's multi-modes of travelling.This methoed could accurately assess spatial equlity level of health service facility.Compared with the multi-mode Ga2SFCA method,the accessibility result of single transportation mode can be severely overestimated or underestimated.The shortage areas of health resources of large hospitals are mainly located in sub-urban areas,where dispersed people lead to large difficulties to achieve the goal of spatial equlity.(4)This dissertation proposes a multi-targeted spatial optimization method,regarding maximum equalization as main target.This dissertation builds location optimization model and attribute optimization model based on Ga2SFCA method,and proposes the solution method to achieve location optimization and attribute optimization.The location of newly-build hospital is optimized in sub-urban areas,and the attribute optimization is applied in main urban area under two attribute optimizing scenarios.Results show that the proposed method overcomes technical difficulties of traditional spatial optimization methods which emphasize efficiency while neglects equity.After location optimization,the standard deviation and the Gini coefficient values of the large hospitals accessibility are significantly reduced in the three sub-urban areas,which prove that optimization effect is obvious.The attribute optimization of fixed total bed numbers have very obvious effect,while is difficult to achieve.Comparatively,the attribute optimization of fixed increasing bed numbers have inferior effect,while have more practical guidances.