Agent-based Simulation Of Pedestrian Spatial Distribution And Improvement Of Spatial Vitality In Living Blocks

The pedestrian spatial distribution is the external manifestation of spatial vitality,and the spatial environment is the material basis for the formation of spatial vitality.Therefore,exploring the interaction mechanism between pedestrian and space is the key to create a dynamic space,and it is also an important way to promote the transformation of space planning and design from space-oriented to people-oriented,and from form and morphological composition to place and vitality creation.A review of the existing research found that although scholars have made notable progress on pedestrian spatial distribution and the improvement of spatial vitality,there are still certain limitations.For example,the spatial elements that affect pedestrian spatial distribution are obtained through regression analysis,which cannot fully reveal the interaction mechanism between space and pedestrians from the perspective of space constraints,pedestrians’ perception and choice of space.Besides,there is a lack of combined consideration of multi-scale and multi-dimensional spatial elements in analyzing the mechanism of pedestrian spatial distribution and spatial vitality improvement strategies.On the basis of reviewing the theory and practice of pedestrian spatial distribution,this paper uses a combination of mathematical analysis,simulation,and empirical research,and constructs a multi-agent-based model guided by the Complex Adaptive System theory(CAS)to explore the formation mechanism of pedestrian spatial distribution.The model was applied to four blocks with different spatial environmental characteristics in Tianjin.Through multiple adjustments of variables and parameters,as well as multiple simulation runs,good fits between the simulation results and the actual pedestrian spatial distribution were achieved,and the validity of the model was verified.On this basis,the key influencing factors of the pedestrian spatial distribution and their influence mechanisms are analyzed,and then strategies to enhance the space vitality are proposed.The main research content and conclusions include the following four aspects:Firstly,analyzes the pedestrian spatial distribution with the Complex Adaptive System theory.This paper proposes that the pedestrian spatial distribution is a complex adaptive system formed by the continuous interaction between pedestrian and space.Space provide pedestrians with spatial information by its form,scale,function,etc.,and serves as a physical carrier for walking activities,restricting pedestrians’ spatial distribution,scale,intensity,and activity types.While pedestrians can actively perceive the space and organize activities accordingly,so that the corresponding pedestrian spatial distribution pattern emerges.Secondly,constructs a multi-agent-based model to simulate pedestrian spatial distribution through programming on Netlogo platform.In this model,agents represent pedestrians and streets.Street agents have attributes,reflecting the characteristics of the street network structure,function distribution and street interface.While pedestrian agents evaluate and select the street as destinations and routes according to the attributes and behavior rules.The attributes of the street are transformed into the behavior rules of the pedestrian agents by mathematical methods.Behavior rules include destination selection rules and route selection rules.In the destination selection rules,the discrete selection model is used to make pedestrians choose based on the evaluation of the destinations’ utility,which includes 60 elements in aspects of street network structure,function layout,and street interface characteristics.In the route selection rules,this model allows pedestrian agents to choose the path with either the smallest angular distance or the shortest geographic distance to the destination.The model is used to simulate spatial distribution of pedestrians in three types of general,commuting,and shopping in four typical living blocks with different spatial environmental characteristics in Tianjin.Through the adjustment of elements and parameters in the behavior rules and multiple simulations,the simulation pedestrian spatial distribution and the actual pedestrian distribution achieved good fits,indicating that the model and the behavioral rules setting can truly and effectively reflect the actual pedestrian spatial distribution at block scale.Thirdly,analyzes the mechanism of pedestrian spatial distribution.It can be seen from the simulation results that the street network structure,function distribution,and street interface jointly affect the pedestrian spatial distribution.In general,in terms of street network structure,small-scale closeness and large-scale betweenness have a positive impact on pedestrians’ destination choices;in terms of function distribution,the construction intensity of commercial facilities,public service facilities,residential facilities,the degree of mixed functions,and public transportation stations have a positive impact on pedestrians,while the specific index have different influences on different areas and different kinds of pedestrians;in terms of street interface,street width to height ratio and green coverage have different influences in different areas.Further analysis of scatter plots found that the coordinated layout of mixed functions and small-scale high closeness is conducive to ecouraging nearby residents to carry out diversified walking activities;the coordinated layout of mixed functions and large-scale betweenness have a larger attraction radius and more diverse types of pedestrians;the mixed functions need to be complementary to attract pedestrians;streets with high closeness and betweenness have location advantages to attract pedestrians;the combination of too large or too small street width to height ratio and single residential function has limited appeal to pedestrians,and the combination of high green coverage and low functional density has limited appeal to pedestrians.Fourthly,proposes spatial vitality enhancement policies.Street network structure,function distribution,and street interface affect pedestrian spatial distribution joingtly.Therefore,the three aspects should be coordinated to provide benefits for the formation of diversified interactions between pedestrians and street space.The general ideas of improving the spatial vitality are: shaping an intensive and compact spatial texture,promoting the flexible combination of spaces and the establishment of diversified connections;optimizing the functions and street interfaces of streets with high betweenness at a large scale to make them the vitality axis;arrange function that are closely related to people’s daily life at streets with high closeness at a small-scale,making it a vitality center within a small-scale;add functions at street with high betweenness or closeness to make them a potential area for vitality improvement.By comparing the changes in the overall,commuting,and shopping simulated pedestrian flow before and after the space optimization,it is found that the number and types of pedestrians on some streets are effectively improved due to increased functional density and mix,improved street network connectivity,etc.Because they provide pedestrians with more diverse and convenient services within a walkable range or diverse path choices,which verifies the validity of the spatial optimization strategy.While when the neighborhoods of the street are already arranged with rich functions,even if the functional density and mixing degree of the street are improved,the attraction to pedestrian is relatively limited.Part of the reason is that existing space layout already provides pedestrians with convenient services within walking distance.This research focuses on the mechanism of pedestrian spatial distribution,guided by the theory of Complex Adaptive System,and proposes that the pedestrian spatial distribution is a complex adaptive system formed by the interaction of pedestrians and space.On this basis,a multi-agent-based model of pedestrian spatial distribution was constructed on Netlogo platform through programming,and its rationality and feasibility were verified in the application of actual cases,providing a bottom-up perspective and method to analyze the pedestrian spatial distribution pattern.Besides,street network structure,function distribution and street interface are included in the model,facilitating to analyze the formation mechanism of the pedestrian spatial distribution more comprehensively and to propose spatial vitality improvement strategies more precisely.