Spatial Optimization of Residential Areas in the Living Circle Based on Residents'Walking Simulations

CHE Guanqiong; LAN Xu; WANG Yitian

doi:10.3969/j.issn.1000-0232.2025.02.003

Abstract：Since the community living circle is delineated according to walking time and distance,analyzing the interaction mechanism between the built environment in the living circle and the walking activities of residents is crucial for creating a human-oriented living area.Simulation technology is not only conducive to visualizing the formation process and distribution trajectory of walking activities in the living area,but also identifies the spatial variables that have a significant impact on walking activities through the model parameters and their influencing modes.This method is effective for realizing human-oriented living areas. An agent-based model was applied in this study.According to the analysis framework of"model construction-data acquisition-simulation-result verification-mechanism analysis-optimization simulation",a daily walking activity simulation model of residents was built using the NetLogo platform.In this model,residents were represented as agents,while attributes and behavioral rules were set.Therefore,it can autonomously evaluate the utility of destinations and choose the shortest paths according to the built environment of the living area.Secondly,a case study based on the Huayuan living area in Tianjin was carried out.On the one hand,the built environment data was obtained through a"poi"data crawl and field survey,and a spatial environment index system covering three first-level indexes(street network structure,functional distribution,and street interface)and 80 second-level indexes were established by using the spatial measurement method,which act as independent variables affecting residents'destination choices.On the other hand,the origin,destination,and walking paths for three types of walking activities—commuting,shopping,and leisure—were obtained by using observation point counting and drone video recordings.Their spatial distributions were summarized to fit the simulation results.On this basis,the three types of walking activities were simulated.After the fitting test with the actual walking activities,the spatial indexes of the living area that have a significant impact on walking activities were identified,and their influencing modes were analyzed.Furthermore,the underutilized spaces and causes were analyzed;finally,the optimization strategy of the spatial environment was proposed and verified through simulations. The main conclusions are as follows:① the degree of fitting between simulation results and actual residents'walking activities reaches about 80％,which verifies the validity of the model;② the space factors that have positive influences on all types of walking activities include the residential density and the accessibility of streets within 1000m;③ there are some differences in the influencing factors of the three types of walking activities.In terms of the walking activities of commuting and shopping,roads with higher accessibility within 1000m are more attractive,and the functional mix is more attractive to commuting pedestrians.For shopping pedestrians,the attractiveness of commercial facilities is significant.In particular,the attractiveness of the density of commercial facilities along the street is far stronger than that of the commercial facility density on two sides of the street within 200m and 400m.The roads with high accessibility in the range of 500m and high width-height ratios are highly attractive to recreational pedestrians.④ The three walking types have different walking distances:within 1100m for commuting,within 800m for shopping,and within 300m for leisure.The results can be explained by the facilities'distributions in the study area:facilities with higher attractiveness to commuting and shopping residents are mostly located on urban secondary roads connecting various neighborhoods,while parks with higher attractiveness to leisure pedestrians are mostly distributed within the neighborhoods.⑤ For all three types of pedestrians,when pedestrians choose the path with the smallest angular distance,the simulation results show the highest fitting degree with the actual walking activity.This indicates that pedestrians in this area are more likely to choose paths with fewer turns and easier spatial cognition.⑥ Walking activity is the result of mutual adaptation and interaction of the network structure,functional facility distribution,and street interfaces.Therefore,it is crucial to conduct overall planning and coordinated layouts of the three types of walking activities to improve the attractiveness of community spaces to different walking activities. In future research,a model should be used to simulate the walking activities for all ages and in different types of living circles.This can help further analyze the spatial and population heterogeneity affecting the influencing mechanism of walking activities.Furthermore,the nonlinear influence and threshold range of the spatial environment of the living area on various types of walking activities can be analyzed more accurately by incorporating machine learning,which can accurately optimize the spatial environment in living circles.