A method for evaluating the accessibility of public transportation stations considering the travel environment

Abstract

This invention relates to a method for evaluating the accessibility of public transportation stations considering the travel environment, comprising: S1, obtaining a set of environmental indicators affecting walking impedance in the station area; S2, identifying key environmental factors; S3, determining the hierarchy of each key environmental factor; S4, obtaining travel route selection survey data based on a passenger travel intention survey; S5, establishing a binary logit model based on the travel route selection survey data to calculate the willingness to pay for each key environmental factor; S6, establishing a comprehensive walking cost model considering the walking environment; S7, using the number of points of interest reachable by walking under a certain travel cost as an indicator of walking accessibility to obtain a station area accessibility calculation model; S8, calculating the station area accessibility of each station based on the station area accessibility calculation model. Compared with existing technologies, this invention establishes a mapping relationship between the physical environment of travel and passengers' psychological perceived impedance, achieving an objective and reliable evaluation of station area accessibility.

Description

Technical Field

[0001] This invention relates to the field of urban public transportation technology, and in particular to a method for evaluating the accessibility of public transportation stations that takes into account the travel environment. Background Technology

[0002] As a high-capacity, centralized mode of transportation, public transportation cannot offer the same door-to-door service as private cars. Therefore, it needs to be effectively integrated with pedestrian and cycling networks to create a convenient and comfortable connection environment, thereby mitigating the negative impact of the connection process on travelers and increasing the attractiveness of public transportation.

[0003] Station accessibility is defined as the ease or difficulty of reaching a destination, and "station-area accessibility" is used to evaluate the degree of connection between urban public transportation stations and surrounding service points. Current methods for evaluating station-area accessibility lack quantitative analysis of factors influencing the travel environment, and a mapping relationship is not established between passenger psychological resistance and these influencing factors. Therefore, this paper proposes a public transportation station-area accessibility evaluation method that considers the travel environment, providing suggestions for clarifying the direction of public transportation system improvement and enhancing public transportation competitiveness. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a method for evaluating the accessibility of public transportation stations that takes into account the travel environment.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for evaluating the accessibility of public transportation stations considering the travel environment, the method comprising:

[0007] S1. Obtain a set of environmental indicators that affect the pedestrian impedance of the station area;

[0008] S2. Identify key environmental factors affecting the perceived impedance of pedestrian walkways in the station area based on the set of environmental indicators.

[0009] S3. Determine the levels of each key environmental factor;

[0010] S4. Based on the identified key environmental factors and hierarchical design, a passenger travel intention questionnaire is used to obtain travel route selection survey data.

[0011] S5. Based on travel route selection survey data, establish a binary logit model to calculate the willingness to pay for each key environmental factor;

[0012] S6. Establish a comprehensive walking cost model that takes into account the walking environment, namely, a passenger travel psychological perception resistance model.

[0013] S7. Under a certain travel cost, the number of points of interest that can be reached by walking is used as an indicator to measure walking accessibility, and a station area accessibility calculation model is obtained.

[0014] S8. Collect information on pedestrian roads, environment, and points of interest around public transportation stations. Calculate the walking perception impedance from each point of interest to the public transportation station based on the passenger travel psychological perception impedance model. Substitute the walking perception impedance as the travel cost into the station area accessibility calculation model to calculate the station area accessibility of each station.

[0015] Preferably, step S2 specifically involves: collecting passengers' cognitive data on various environmental factors in the environmental indicator set, using an expanded contribution model to analyze the importance of environmental factors, dividing them into importance levels, applying the interpretive structural model method to stratify and rank the environmental factors from largest to smallest importance, and selecting several environmental factors with higher importance as key environmental factors affecting the perceived impedance of walking in the station area.

[0016] Preferably, the key environmental factors affecting the pedestrian perception impedance of the station area include walking time, crossing method, separation of pedestrians and non-motorized vehicles, distribution of obstacles, and distribution of entrances and exits.

[0017] Preferably, step S4 specifically involves: treating each key environmental factor as a design factor, performing orthogonal design based on the stratification results of the key environmental factors, designing multiple travel scenarios, and designing two travel routes under each travel scenario, thereby forming a passenger travel intention survey form.

[0018] Preferably, step S5, which establishes a binary logit model based on the travel route selection survey data, is based on the framework of stochastic utility theory. It treats the survey participants as consumers and the available travel routes as choices. The established binary logit model is expressed as follows:

[0019] U in =V in +ε in

[0020]

[0021] Among them, U in It is the utility value of consumer n choosing product profile i, ε in V represents random utility; in It is system utility, related to the properties of the contour, x ink Let θ be the k-th attribute in consumer n's choice i. k X represents the fitting parameters corresponding to the k-th attribute. in Let θ' represent the attribute vector of option i, and let θ' represent θ. k The vector formed by the vector.

[0022] Preferably, step S5, calculating the willingness to pay for each key environmental factor, includes:

[0023] If the key environmental factor is a continuous variable, then the willingness to pay for that key environmental factor is:

[0024] WTP k =θ k / θ T

[0025] Among them, WTP k Willingness to pay for key environmental factor k, θ T θ represents the fitting parameter for the walk time attribute in the binary logit model. k These are the fitting parameters corresponding to the key environmental factor k in the binary logit model.

[0026] Preferably, if the key environmental factor is a discrete variable, then the willingness to pay for that key environmental factor is:

[0027] WTP ki =(θ ki -θ kp ) / θ T

[0028] Among them, WTP ki θ represents the willingness to pay for the i-th discrete quantity of key environmental factor k. T θ represents the fitting parameter for the walk time attribute in the binary logit model. ki Let θ be the fitting parameter corresponding to the i-th discrete quantity of the key environmental factor k in the binary logit model. kp is the fitting parameter corresponding to the p-th discrete quantity of the key environmental factor k in the binary logit model, and the p-th discrete quantity of the key environmental factor k is used as the benchmark quantity.

[0029] Preferably, the walking cost model in step S6 is expressed as:

[0030] C = T + W cro_sig ×β cro_sig +W cro_over ×β cro_over +W cro_under ×β cro_under +W div_par ×χ div_par +W div_no ×χ div_no +W Obs ×Obs+W Ent ×Ent

[0031] Where C is the total cost of walking, T is the walking time, and β is the total cost of walking. cro_sig βcro_over β cro_under χ represents the percentage of pedestrian crossings with traffic lights, the percentage of pedestrian overpasses, and the percentage of underpasses along the path; div_par , χ div_no These represent the proportion of road segments with separate pedestrian and non-motorized vehicle sections and the proportion of road segments without such separation, respectively; Obs represents the average number of obstacles per 100 meters in the path; Ent represents the average number of entrances / exits per 100 meters in the path; W cro_sig W cro_over W cro_under , W div_par W div_no , W Obs , W Ent This represents the willingness to pay corresponding to each key environmental factor.

[0032] Preferably, the station accessibility calculation model is expressed as follows:

[0033]

[0034] i and j represent public transportation stops and interests, respectively; C ij C represents the travel cost from public transport station i to point of interest j. ij The comprehensive walking cost model considering the walking environment is used to obtain the value; C0 is the travel perception time threshold from public transport station i to point of interest j; R i G represents the cumulative number of opportunities for public transport station i, i.e., walking accessibility; G represents the urban walking network.

[0035] Preferably, the method further includes calculating the influence coefficient:

[0036]

[0037] Where R i To assess the walkability of public transport station i under the psychological perception resistance of passengers; R i实 This represents the cumulative number of opportunities for public transport station i during actual walking time.

[0038] Compared with the prior art, the present invention has the following advantages:

[0039] (1) This invention fully considers the dynamism and scalability of the evaluation. The evaluation indicators used are obtained by screening the initial indicator set based on the survey data of passenger cognition during the evaluation process. They have strong objectivity and can be dynamically adjusted and changed with the changes in passenger cognition over time, making them more targeted. The evaluation method is still applicable and has strong scalability, which is in line with the current situation of rapid development of urban public transportation.

[0040] (2) The present invention evaluates the accessibility of the station area not only by considering the connectivity between the pedestrian road and the station, but also by quantifying the travel comfort affected by the pedestrian environment, which makes it easier to make more targeted suggestions and prioritize measures. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the steps of the station accessibility evaluation method of the present invention, which considers the travel environment.

[0042] Figure 2 This is a comparison of accessibility at each station within different perception time thresholds in Example 1;

[0043] Figure 3 This is a distribution diagram of the influence coefficients of each site in Example 1. Detailed Implementation

[0044] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Note that the following description of the embodiments is merely illustrative and is not intended to limit its applicability or use, nor is the present invention limited to the following embodiments.

[0045] Example

[0046] like Figure 1 As shown in the figure, this embodiment provides a method for evaluating the accessibility of public transportation stations considering the travel environment. The method includes:

[0047] S1. Based on Maslow's hierarchy of needs, a pedestrian walking needs hierarchy is established, and the dimensions of pedestrian environment evaluation indicators are divided. By summarizing the existing pedestrian connection environment evaluation indicator systems in the literature and further organizing and summarizing them, a set of environmental indicators affecting the pedestrian impedance of the station area is obtained;

[0048] S2. Design corresponding questionnaires and plans to collect passengers' awareness of each initial evaluation indicator. Based on this data, screen and determine the initial indicator set. Use the Expanded Contribution Model (ECR) to analyze the importance of factors, divide the importance hierarchy, and apply the Interpretive Structural Model (ISM) method to show the hierarchical ranking results of factor importance, and identify the key environmental factors affecting the perceived impedance of pedestrian access.

[0049] S3. Based on the final set of factors affecting passenger walking perceived impedance, and with reference to relevant standards, data and literature, determine the specific level of each walking environment factor.

[0050] S4. Based on the determined factor attributes and levels, an orthogonal design method is used to combine the attributes at each level to obtain a reasonable SP survey plan, which respondents can evaluate and judge by selection. A questionnaire is designed according to the survey plan to conduct a survey on passenger travel intentions. After completing the initial draft of the questionnaire, a usability test is used to conduct a preliminary survey within a certain scope. Based on the results of the preliminary survey, the questionnaire is revised, and the survey scope, methods, and sample size are determined. Finally, the survey is conducted, questionnaires are collected, and the data is processed.

[0051] S5. Based on travel route selection survey data, establish a binary logit model to calculate the willingness to pay for each key environmental factor;

[0052] S6. Establish a comprehensive walking cost model that considers the walking environment, i.e., a passenger travel psychological perception resistance model.

[0053] C ij =f(T) ij Z 1ij Z 2ij Z 3ij …)

[0054] C ij T represents the generalized travel cost between i and j, i.e., the perceived travel time; ij Z represents the actual walking time between i and j; 1ij Z 2ij Z 3ij … represents the walking environment factors between i and j.

[0055] The key environmental factors affecting pedestrian perception impedance in this invention include walking time, street crossing methods, pedestrian and non-motorized vehicle separation methods, obstacle distribution, and entrance / exit distribution. The comprehensive walking cost model is expressed as follows:

[0056] C = T + W cro_sig ×β cro_sig +W cro_over ×β cro_over +W cro_under ×β cro_under +W div_par ×χ div_par +W div_no ×χ div_no +W Obs ×Obs+W Ent ×Ent

[0057] Where C is the total cost of walking, T is the walking time, and β is the total cost of walking. cro_sig β cro_over β cro_under χ represents the percentage of pedestrian crossings with traffic lights, the percentage of pedestrian overpasses, and the percentage of underpasses along the path;div_par , χ div_no These represent the proportion of road segments with separate pedestrian and non-motorized vehicle sections and the proportion of road segments without such separation, respectively; Obs represents the average number of obstacles per 100 meters in the path; Ent represents the average number of entrances / exits per 100 meters in the path; W cro_sig 、W cro_over 、W cro_under W div_par 、W div_no W Obs W Ent The willingness to pay corresponds to each key environmental factor, where the form of crossing the street and the separation of pedestrians and non-motorized vehicles are discrete variables, while walking time, the distribution of obstacles, and the distribution of entrances and exits are continuous variables.

[0058] S7. Under a certain travel cost, the number of points of interest reachable by walking is used as an indicator to measure walkability, resulting in a station-area accessibility calculation model, which is expressed as:

[0059]

[0060] i and j represent public transportation stops and interests, respectively; C ij C represents the travel cost from public transport station i to point of interest j. ij The comprehensive walking cost model considering the walking environment is used to obtain the value; C0 is the travel perception time threshold from public transport station i to point of interest j; R i G represents the cumulative number of opportunities for public transport station i, i.e., walking accessibility; G represents the urban walking network.

[0061] S8. Collect information on pedestrian roads, environment, and points of interest around public transportation stations. Calculate the walking perception impedance from each point of interest to the public transportation station based on the passenger travel psychological perception impedance model. Substitute the walking perception impedance as the travel cost into the station area accessibility calculation model to calculate the station area accessibility of each station.

[0062] S9. Calculate the influence coefficient:

[0063]

[0064] Where R i To assess the walkability of public transport station i under the psychological perception resistance of passengers; R i实 This represents the cumulative number of opportunities for public transport station i during actual walking time.

[0065] The specific content of Maslow's hierarchy of needs theory is as follows:

[0066] Maslow's hierarchy of needs, proposed by American psychologist Abraham Harold Maslow in 1943, is an important theory in behavioral science. This theory posits that humans need motivation to fulfill five categories of needs, from lowest to highest: physiological needs, safety needs, social needs, esteem needs, and self-actualization needs.

[0067] (1) Physiological needs. Air, water, food, sleep, and physiological balance are the most basic requirements for human survival, and also the main driving force for human action. Maslow proposed that only after basic physiological needs are met and survival can be maintained will other needs emerge and become new driving forces.

[0068] (2) Safety Needs. The second level of needs mainly refers to safety, which includes not only personal safety and health, but also the protection of property and resources, moral protection, and family safety. The entire human body mechanism is driven by the pursuit of safety, and corresponding sensory organs, effector organs, and energy are all instruments used to obtain safety. Science and life philosophy can also be regarded as part of satisfying safety needs.

[0069] (3) Social needs. After satisfying the above two needs, people begin to seek emotional satisfaction, namely social needs, including emotional communication, interpersonal relationships, etc., which involve personal characteristics, experiences, education and religious beliefs.

[0070] (4) The need for respect. People all hope that their abilities and achievements will be recognized by others and society. The need for respect corresponds to this psychology. It is a higher level of emotional satisfaction and includes self-respect, confidence, achievement, respect for others, and being respected by others. Maslow believed that once this need is met, people can be full of confidence and enthusiasm for life.

[0071] (5) Self-actualization needs. This is the highest level of personal needs, involving morality, creativity, self-awareness, and fairness. It is an individual's own need and is unrelated to external factors. After the above four levels of needs are met, people begin to pursue the realization of their personal value, eager to exert their abilities and realize their personal ideals, thereby achieving self-recognition and accomplishing everything commensurate with their abilities.

[0072] The five needs mentioned above are considered the main driving forces behind human behavior and exhibit a clear hierarchical division. Once a lower-level need is satisfied, a higher-level need will take its place as the new motivating factor. However, human behavior is complex and is the result of the combined effect of multiple needs. Based on the above needs, more layers of needs can be derived.

[0073] Furthermore, the questionnaire design for screening the evaluation indicators aims to screen the influencing factors of perceived impedance while walking, and selects a relatively mature five-point Likert scale to test travelers' judgment of the importance of initial influencing factors.

[0074] The questionnaire is divided into two parts: an assessment of the importance of influencing factors and personal characteristic information. For the importance assessment section, the questionnaire design strictly adhered to the principles of simplicity, adaptability, purposefulness, and relevance. To improve readability, some factors were described in a more easily understood way. Each question in the questionnaire is a single-choice question, with options set as an importance scale: not important at all (1 point), not very important (2 points), moderately important (3 points), relatively important (4 points), and very important (5 points). The personal characteristic information section includes survey information on passengers' gender, age, and purpose of using public transportation.

[0075] Furthermore, the basic theory of the Expanded Contribution (ECR) model is as follows:

[0076]

[0077] The above formula represents the function of decision-maker l's preferences for factors; a i and a j These are two options representing the group's overall preference; g represents the group's overall preference level; R represents the group's preference factors; and m represents the number of decision-makers. Its meaning is the decision-maker's a i For a j A certain degree of preference; when the degree of preference is ≥0, it indicates that the decision-maker has a preference for a. i Preferences and a j The degree of preference is the same or greater.

[0078] a i The utility value can be used to u l (a i ) indicates that, let For a i and a j The difference in utility between them can be obtained as follows:

[0079]

[0080] The formula for calculating the g function is as follows:

[0081]

[0082] w l λ represents the weight of the decision-maker's opinion; λ (≥0) represents the weight of consensus within the group; θ (≥0) represents the difference excluding those with weak connections.

[0083] Furthermore, the application process of the Interpretive Structural Modeling (ISM) method is as follows:

[0084] The ISM model transforms preference values ​​between factors into an association matrix, clearly illustrating the relationships between factors and graphically representing the hierarchical ranking of factor importance, making it more concise and clear. The relationship between factors is represented by an adjacency matrix A, where elements a... ij The definition is shown in the following formula.

[0085]

[0086] Where, Group preference level.

[0087] Based on the adjacency matrix A, an identity matrix I is added, with the meaning of the matrix elements remaining unchanged, forming a one-step reachability matrix M. Further, based on the reachability matrix, the reachable sets of each factor are listed, i.e., factors with lower preference, and factors are extracted layer by layer to obtain the importance hierarchy classification results.

[0088] Furthermore, the scenario design in the aforementioned passenger travel intention survey questionnaire first determines one path option for each scenario based on the scenario design scheme, and then designs another path with similar impedance as a second option. The questionnaire uses descriptions that are easy for respondents to understand to avoid ambiguity.

[0089] Each questionnaire contains two parts: one part describes the respondents' demographic characteristics, including gender, age, purpose of public transportation use, public transportation connection time and mode of transport; the other part describes the respondents' choices for various travel scenarios. Additionally, the questionnaire includes trap questions—two questions with identical questions and options but in reverse order—to screen the questionnaire's validity.

[0090] Furthermore, the principle behind encoding all selection sets and results is that each respondent's choice for each path option should be presented as a separate data point. First, the original questionnaire data undergoes preliminary processing, removing answers to trap questions and personal information questions, and then encoding the options corresponding to those questions. Discrete variable encoding employs effect-coding, setting the reference level to -1 for all new variables, and the effect coefficient of the baseline variable is the negative of the sum of its coefficients.

[0091] Furthermore, the binary logit model is based on the framework of stochastic utility theory, and its principle is as follows:

[0092] When consumer n makes a choice within subset J, the observed utility value is divided into two parts: systematic utility and random utility.

[0093] U in =V in+ε in

[0094]

[0095] Where U in It is the utility value of consumer n choosing product profile i; ε in V represents random utility; in It is system utility, related to the properties of the contour; x ink Let X be the k-th attribute in consumer n's choice i. in The attribute vector representing i; θ k This represents the unknown parameter corresponding to the k-th attribute, and θ′ represents θ. k The vector formed by the vector.

[0096] When both option i and option j are in subset J, based on utility maximization theory, the condition under which consumer n chooses option i instead of option j is:

[0097] U in >U jn , i≠j, i,j∈J

[0098] The probability P that the consumer chooses option i in for:

[0099] P in =P[{ε jn -ε in}<{V jn -V in}],j≠i

[0100] The distribution of the extreme values ​​of independent, identical, type I random terms, therefore the probability P in The calculation can be performed using a multinomial logit model. When there are only two choices, the ML model is a binary logit (BL) model, which takes the following form:

[0101] or

[0102] For the present invention, step S5, which establishes a binary logit model based on travel route selection survey data, is based on the framework of stochastic utility theory, treating the passengers participating in the survey as consumers and the available travel routes as selection options, and thus establishing a binary logit model.

[0103] Furthermore, the willingness to pay (WTP) of a certain attribute k k The converted values ​​of each factor relative to time can be expressed as:

[0104] For continuous variables, the calculation formula is the ratio of the attribute coefficient to the time coefficient, indicating the amount of time that is willing to increase for each unit increase in attribute k. The calculation formula is as follows.

[0105] WTP k =θ k / θ T

[0106] For discrete variables, the calculation formula is the ratio of the difference in attribute level coefficients to the time coefficient, indicating the amount of time travelers are willing to increase for attribute k compared to attribute p, as shown in the following formula.

[0107] WTP k =(θ k -θ p ) / θ T

[0108] For the purposes of this invention, if the key environmental factor is a continuous variable, then the willingness to pay for that key environmental factor is:

[0109] WTP k =θ k / θ T

[0110] Among them, WTP k Willingness to pay for key environmental factor k, θ T θ represents the fitting parameter for the walk time attribute in the binary logit model. k These are the fitting parameters corresponding to the key environmental factor k in the binary logit model.

[0111] If the key environmental factor is a discrete variable, then the willingness to pay for that key environmental factor is:

[0112] WTP ki =(θ ki -θ kp ) / θ T

[0113] Among them, WTP ki θ represents the willingness to pay for the i-th discrete quantity of key environmental factor k. T θ represents the fitting parameter for the walk time attribute in the binary logit model. ki Let θ be the fitting parameter corresponding to the i-th discrete quantity of the key environmental factor k in the binary logit model. kp is the fitting parameter corresponding to the p-th discrete quantity of the key environmental factor k in the binary logit model, and the p-th discrete quantity of the key environmental factor k is used as the benchmark quantity.

[0114] Furthermore, the specified threshold T in the station accessibility calculation model is not the average time obtained by taking the Euclidean distance radius with the public transportation station as the center, but the walking perception time on the actual path from the POI to the public transportation station, which needs to take into account environmental factors on the walking path.

[0115] Furthermore, the influence coefficient β takes the value [0, 1], which means that, under the same time threshold, the number of POIs that can be reached considering perceived time impedance is equal to the number of POIs that can be reached considering actual time. When the walking environment is excellent and the traveler's perceived time is the same as the actual time, the ratio reaches 1, that is, the higher the value of β, the better the walking environment.

[0116] The following is a specific example to illustrate the above implementation plan:

[0117] Step 1: Construction of a set of environmental indicators affecting pedestrian impedance in the station area

[0118] Applying Maslow's hierarchy of needs to the process of pedestrian walking, we find that pedestrians' needs for the travel environment vary depending on factors such as road facilities and traffic conditions.

[0119] First, basic physiological needs correspond to pedestrians' accessibility needs. When sidewalks are too narrow or pedestrian traffic is too high, insufficient walking space makes passage difficult, but pedestrians don't yet have psychological needs. Once sidewalks are passable and pedestrian accessibility is met, the need for safety increases. Factors affecting safety, such as the separation between pedestrians and non-motorized / motorized vehicles, and the speed of non-motorized / motorized vehicles, receive more attention. When accessibility and safety are largely satisfied, pedestrians begin to focus on convenience, hoping sidewalks will provide easy access. The highest level of need is comfort, corresponding to Maslow's hierarchy of needs for esteem and self-actualization. This involves seeking comfort and dignity in travel, and relates to pedestrian-friendly facilities.

[0120] Based on the above analysis of pedestrian needs hierarchy, the pedestrian environment evaluation index is divided into four dimensions: accessibility, safety, convenience, and comfort. By summarizing the existing pedestrian connection environment evaluation index system in the literature and further organizing and summarizing it, the set of environmental indicators affecting the pedestrian impedance of the station area is shown in Table 1, which contains a total of 18 factors.

[0121] Table 1. Set of environmental factors affecting pedestrian impedance in the station area

[0122]

[0123]

[0124] Step 2: Identification of key environmental factors affecting the perceived impedance of pedestrian access

[0125] Regarding the survey on the evaluation of the importance of indicators, a total of 234 complete questionnaires were collected. The validity of the questionnaires was further screened based on the time spent in answering the questionnaires and the differences in the options, and 223 valid questionnaires were obtained, with an effective rate of 95.3%, which meets the statistical requirements.

[0126] The reliability and consistency of the questionnaire data were verified through descriptive feature statistics, questionnaire reliability analysis, and validity analysis. According to the ECR method, the weight of opposing opinions was set to 0.5, with λ = 0.5 and θ = 0. ECR preference scores were calculated and stratified to obtain the importance hierarchy results, as shown in Table 2.

[0127] Table 2. Ranking of Factor Importance

[0128] serial number factor Importance Preference value 11 Pedestrian crossing method (with or without traffic lights) 18 6.744 10 Frequency of entrance and exit settings 16 3.231 4 Methods of separating pedestrian walkways from non-motorized vehicles 14 3.935 8 Distribution of obstacles on the sidewalk 14 3.422 2 Number of lanes for motor vehicles and whether there is a central divider 13 1.166 17 Pedestrian lighting design 13 1.960 3 Pedestrian signs 12 1.809 18 Sidewalk pavement material 11 2.626 7 Sidewalk congestion 10 2.072 15 Shading and rain protection facilities on sidewalks 10 1.881 9 Location of pedestrian entrances and exits 9 1.251 12 Number of times crossing the street 8 0.798 6 pedestrian traffic on the sidewalk 7 1.020 1 width of sidewalk 5 0.594 16 Landscape design on the sidewalk 4 0.283 13 The nature and density of street-front shops 3 0.126 14 weather 3 0.224 5 Slope of sidewalk 1 0.000

[0129] The stratification of importance results shows a clear stratification in preference values ​​between the factors "distribution of obstacles on the sidewalk" and "number of motor vehicle lanes and presence of a central median". Therefore, the top four environmental factors and walking time were selected as the key factors influencing pedestrians' perceived walking impedance between their starting point / end point and public transportation stops.

[0130] Therefore, the functional relationship between perceived time and actual time can be expressed as follows:

[0131] C = f(Time, Cro, Div, Obs, Ent)

[0132] In the formula, C is the walking perception impedance; f(Time,Cro,Div,Obs,Ent) represents a function of variables such as walking time, crossing method, separation of pedestrians and non-motorized vehicles, distribution of obstacles, and distribution of entrances and exits, which is a correction function for the actual walking time.

[0133] Step 3: Determine the level of factors

[0134] Regarding walking time, based on literature review and data analysis, the threshold for walking connection time to urban rail transit stations in my country is approximately 16 minutes, with the longest time level set at 16 minutes. Considering that insufficient walking time leads to inadequate environmental awareness, the shortest time level is set at 5 minutes. Four levels are set, with 8 minutes and 12 minutes added between 5 and 16 minutes.

[0135] According to my country's urban road design specifications, pedestrian crossings include at-grade crossings and grade-separated crossings. At-grade crossings include unsignaled and signalized pedestrian crossings, while grade-separated crossings include pedestrian overpasses and pedestrian underpasses, totaling four levels.

[0136] The separation methods for pedestrian walkways and non-motorized vehicle lanes can be categorized into three types: complete separation, partial separation, and no separation. Complete separation refers to a system where the pedestrian walkway is raised by curb stones or separated by continuous green belts, guardrails, etc., allowing pedestrians and non-motorized vehicles to use their own lanes, significantly reducing conflicts. Partial separation involves the design of pedestrian and non-motorized vehicle lanes at different elevations, using trees, street furniture, and bollards for "flexible" separation, allowing pedestrians and non-motorized vehicles to share space and improving road utilization. No separation refers to a shared area for pedestrians and non-motorized vehicles, without physical separation facilities, using markings or pavement to distinguish their travel spaces. Based on this, this attribute is divided into three levels.

[0137] Regarding the distribution of obstacles on sidewalks, according to relevant research findings, on average, 0.5 obstacles within 100m indicate very good connectivity, 1 obstacle indicates moderate connectivity, and 2 or more obstacles indicate very poor connectivity.

[0138] The entry and exit of motor vehicles and non-motor vehicles will affect the pedestrian experience at the entrance and exit. Referring to the levels of "pedestrian obstacles", the number of entrances and exits is set to 4 levels.

[0139] The specific attributes and hierarchical settings of the pedestrian environment are shown in Table 3.

[0140] Table 3. Attributes and Hierarchical Settings of the Pedestrian Environment

[0141]

[0142]

[0143] Step 4: Questionnaire Data Collection and Coding

[0144] There are 5 pedestrian environment attributes involved, and a total of 16 scenarios are involved using orthogonal design, as shown in Table 4. For the design of options within each scenario, firstly, based on the scenario design scheme described above, one path option can be determined for each scenario. Then, another path with similar impedance is designed as the second option. For example, the scenario design is shown in Table 5.

[0145] Table 4 Orthogonal Design Table

[0146] plan Factor A Factor B Factor C Factor D Factor E 1 4 2 1 4 3 2 3 2 1 2 1 3 1 3 1 4 2 4 3 1 2 4 4 5 2 2 2 1 2 6 1 2 3 3 4 7 2 4 3 4 1 8 4 4 1 1 4 9 4 3 2 3 1 10 1 1 1 1 1 11 2 3 1 2 4 12 4 1 3 2 2 13 1 4 2 2 3 14 3 3 3 1 3 15 2 1 1 3 3 16 3 4 1 3 2

[0147] Table 5 Examples of Scenario Design

[0148]

[0149] An online survey was conducted, with two sets of questionnaires used to collect information. After verification, a total of 152 valid samples were obtained, with a total sample size of 1216, meeting statistical requirements. The data was recoded, with each row of the original questionnaire data split into 16 rows, resulting in a total of 2432 data entries.

[0150] Step 5: Calculation of Factor Preference Values

[0151] Based on the questionnaire data, a binary Logit model was established and tested. The parameter calibration results are shown in Table 6. In this embodiment, the crossing method and the separation method between pedestrians and non-motorized vehicles are used as discrete variables, while walking time, the distribution of obstacles, and the distribution of entrances and exits are used as continuous variables. Therefore, during calibration, the crossing method is discretized into 4 variables, and the separation method between pedestrians and non-motorized vehicles is discretized into 3 variables.

[0152] Table 6 Model parameter calibration results

[0153] variable coefficient Standard deviation T_value Time -0.158004 0.026815 -5.8923*** cro_sig -0.032943 0.012689 -2.5962** cro_nsig 0.662565 0.179072 3.7000*** cro_over -0.304085 0.133074 -2.2851** cro_under -0.325537 0.152369 -2.1365* div_all 0.483239 0.131589 3.6597*** div_par 0.481577 0.129281 3.7250*** div_no -0.964816 0.258906 -3.7265*** Obs -0.665786 0.203984 -3.2639** Ent -0.461692 0.157705 -2.9276**

[0154] As can be seen, at a 95% confidence level, the T-values ​​for all indicators are greater than 1.96, indicating that all variables are significant; the Macfadden R-squared model... 2 The value of 0.241 indicates that the model fits well.

[0155] Next, we calculated the willingness to pay for each indicator, that is, the converted value of each factor relative to travel time. The results are shown in Table 7.

[0156] Table 7. Willingness to Pay by Indicator

[0157]

[0158]

[0159] Step Six: Establishing a Passenger Psychological Perception Resistance Model

[0160] Further calculations of the proportion of intersections with pedestrian crossings and the proportion of path lengths for pedestrian-only routes yield the following comprehensive pedestrian cost model considering the pedestrian environment:

[0161] C = T + 4.40 × β cro_sig +6.12×β cro_over +6.25×β cro_under +0.01×χ div_par +9.16×χ div_no +4.21×Obs+2.92×Ent

[0162] In the formula: C is the walking sensing impedance; β cro_sig βcro_over β cro_under χ represents the percentage of pedestrian crossings with traffic lights, the percentage of pedestrian overpasses, and the percentage of underpasses along the path; div_par , χ div_no These represent the proportion of road segments with separate pedestrian and non-pedestrian sections and the proportion of road segments without such separation, respectively; Obs represents the average number of obstacles per 100 meters in the path; Ent represents the average number of entrances and exits per 100 meters in the path.

[0163] Step 7: Establishing the Site Accessibility Model

[0164] The following is a model for evaluating the accessibility of a website area:

[0165]

[0166] i and j represent public transportation stops and POI points, respectively; C ij Let C0 be the travel cost from public transport station i to point of interest j, represented by travel perceived impedance; C0 is the travel perceived time threshold from public transport station i to point of interest j; R i G represents the cumulative number of opportunities for public transport station i, i.e., walking accessibility; G represents the urban walking network.

[0167] Step 8: Site-wide reachability calculation

[0168] Six public transportation stations in Jiading District were selected for analysis of their accessibility: regular bus stations (Jiading Central Hospital, Liuliqiao, Chengliu Highway Shiqian Road, Boyuan Road Yutian South Road, and Xinhou Road Xinyu Road) and rail transit station (Jiading New City).

[0169] According to the "Urban Comprehensive Transportation System Planning Standard" (GB / T 51328-2018), the coverage radius of rail transit stations is taken as 800m, and the coverage radius of regular bus stations is taken as 500m. The POI properties within the coverage area of ​​the above 6 public transportation stations are statistically analyzed.

[0170] Previous studies have suggested that different distances provide pedestrians with varying travel experiences and corresponding evaluation levels for walking trips. Following this research, we defined walking perception time thresholds as 5 min, 10 min, 15 min, 20 min, 25 min, and 30 min, and counted the number of Points of Interest (POIs) reachable within these threshold ranges as accessibility results. (See below) Figure 2 .

[0171] (1) 5-minute accessibility

[0172] A 5-minute walk is comfortable for travelers, but the number of Points of Interest (POIs) reachable within 5 minutes is low at each station. Given the generally low absolute number of POIs, the relative differences between stations are not significant.

[0173] (2) 10-minute accessibility

[0174] When the walking perception time threshold is increased to 10 minutes, the accessibility of all stations increases, and the advantages of stations with better accessibility begin to stand out, showing a rapid upward trend. The data in the figure shows that when the walking perception time is less than 10 minutes, the accessibility of Jiading New Town and Xinhou Road-Xinyu Road stations is the best, while the accessibility of Liuliqiao and Chengliu Highway-Shiqian Road stations remains poor, and the change with the increase of range is slow.

[0175] (3) 15min

[0176] When the perceived walking time is 15 minutes, the stations are considered generally accessible. Within this range, compared to the 10-minute range, station accessibility increases significantly, especially at the Xinhou Road and Xinyu Road stations. The Liuliqiao and Chengliu Highway / Shiqian Road stations, which have poorer accessibility, have fallen significantly behind other stations.

[0177] (4) 20min

[0178] When the perceived walking time reaches 20 minutes, although it is still possible to reach the destination on foot, the comfort level will be significantly reduced. It can be seen that the perceived walking time for POIs within the coverage area of ​​each station is almost entirely within this threshold, which is consistent with reality. Beyond this coverage area, travelers are highly unlikely to choose to walk.

[0179] Step Nine: Calculation of Influence Coefficient

[0180] Calculate the pedestrian environment impact coefficient β for each station under different thresholds, see Figure 3 .

[0181] The distribution of β shows that the β values ​​are consistently low when the walking time threshold is 5 minutes, indicating that the walking environment has a significant impact on accessibility within this range. As the walking time threshold increases, the impact of the walking environment on accessibility gradually decreases. Analyzing each station individually, the walking environment around Jiading New Town and Boyuan Road / Yutian South Road stations is excellent, while the walking environment around Liuliqiao and Chengliu Highway / Shiqian Road stations is poor, which aligns with the aforementioned situation. Based on the field survey, it is recommended to design the walking environment around the Liuliqiao and Chengliu Highway / Shiqian Road stations, primarily by clarifying lane functions and increasing pedestrian-non-motorized vehicle separation facilities.

[0182] The above embodiments are merely illustrative and do not constitute a limitation on the scope of the present invention. These embodiments can also be implemented in various other ways, and various omissions, substitutions, and modifications can be made without departing from the technical spirit of the present invention.

Claims

1. A method for evaluating the accessibility of public transportation stations considering the travel environment, characterized in that, The method includes: S1. Obtain a set of environmental indicators that affect the pedestrian impedance of the station area; S2. Identify key environmental factors affecting the perceived impedance of pedestrian walkways in the station area based on the set of environmental indicators. S3. Determine the levels of each key environmental factor; S4. Based on the identified key environmental factors and hierarchical design, a passenger travel intention survey form is used to obtain travel route selection survey data. S5. Based on travel route selection survey data, establish a binary logit model to calculate the willingness to pay for various key environmental factors; S6. Establish a comprehensive walking cost model that takes into account the walking environment, namely, a passenger travel psychological perception resistance model. S7. Under a certain travel cost, the number of points of interest that can be reached by walking is used as an indicator to measure walking accessibility, and a station area accessibility calculation model is obtained. S8. Collect information on pedestrian roads, environment, and points of interest around public transportation stations. Calculate the walking perception impedance from each point of interest to the public transportation station based on the passenger travel psychological perception impedance model. Substitute the walking perception impedance as the travel cost into the station area accessibility calculation model to calculate the station area accessibility of each station. Step S2 specifically involves: collecting passenger cognitive data on various environmental factors in the environmental indicator set, using an expanded contribution model to analyze the importance of environmental factors, dividing them into importance levels, applying the interpretive structural model method to stratify and rank the environmental factors from largest to smallest importance, and selecting several environmental factors with higher importance as key environmental factors affecting the station area walking perception impedance.

2. The method for evaluating the accessibility of public transportation stations considering the travel environment according to claim 1, characterized in that, The key environmental factors affecting the perceived impedance of pedestrians in the station area include walking time, crossing methods, separation of pedestrians and non-motorized vehicles, distribution of obstacles, and distribution of entrances and exits.

3. The method for evaluating the accessibility of public transportation stations considering the travel environment according to claim 1, characterized in that, Step S4 specifically involves treating each key environmental factor as a design factor, performing orthogonal design based on the stratification results of the key environmental factors, designing multiple travel scenarios, and designing two travel routes under each travel scenario, thereby forming a passenger travel intention survey form.

4. The method for evaluating the accessibility of public transportation stations considering the travel environment according to claim 1, characterized in that, Step S5 establishes a binary logit model based on travel route selection survey data. This model is based on the framework of stochastic utility theory, treating survey participants as consumers and the available travel routes as choices. The resulting binary logit model is expressed as follows: U in =V in +e in Among them, U in It is the utility value of consumer n choosing product profile i, ε in V represents random utility; in It is system utility, related to the properties of the contour, x ink Let θ be the k-th attribute in consumer n's choice i. k X represents the fitting parameters corresponding to the k-th attribute. in Let θ' represent the attribute vector of option i, and let θ' represent θ. k The vector formed by these vectors.

5. The method for evaluating the accessibility of public transportation stations considering the travel environment according to claim 4, characterized in that, Step S5 calculates the willingness to pay for each key environmental factor, including: If the key environmental factor is a continuous variable, then the willingness to pay for that key environmental factor is: WTP k =θ k / i T Among them, WTP k For the willingness to pay for key environmental factor k, θ T θ represents the fitting parameter for the walk time attribute in the binary logit model. k These are the fitting parameters corresponding to the key environmental factor k in the binary logit model.

6. The method for evaluating the accessibility of public transportation stations considering the travel environment according to claim 5, characterized in that, If the key environmental factor is a discrete variable, then the willingness to pay for that key environmental factor is: WTP ki =(θ ki -θ kp ) / θ T Among them, WTP ki θ represents the willingness to pay for the i-th discrete quantity of key environmental factor k. T θ represents the fitting parameter for the walk time attribute in the binary logit model. ki Let θ be the fitting parameter corresponding to the i-th discrete quantity of the key environmental factor k in the binary logit model. kp is the fitting parameter corresponding to the p-th discrete quantity of the key environmental factor k in the binary logit model, and the p-th discrete quantity of the key environmental factor k is used as the benchmark quantity.

7. The method for evaluating the accessibility of public transportation stations considering the travel environment according to claim 2, characterized in that, The comprehensive walking cost model in step S6 is expressed as follows: C=T+W cro_sig ×β cro_sig +W cro_over ×β cro_over +W cro_under ×β cro_under +W div_par ×χ div_par +W div_no ×χ div_no +W Obs ×Obs+W Ent ×Ent Where C is the total cost of walking, T is the walking time, and β is the total cost of walking. cro_sig β cro_over β cro_under χ represents the proportion of pedestrian crossings with traffic lights, the proportion of pedestrian overpasses, and the proportion of pedestrian underpasses along the path, respectively; div_par , χ div_no These represent the proportion of road segments with separate pedestrian and non-motorized vehicle sections and the proportion of road segments without such separation, respectively; Obs represents the average number of obstacles per 100 meters in the path; Ent represents the average number of entrances / exits per 100 meters in the path; W cro_sig 、W cro_over 、W cro_under W div_par 、W div_no W Obs W Ent This represents the willingness to pay corresponding to each key environmental factor.

8. The method for evaluating the accessibility of public transportation stations considering the travel environment according to claim 1, characterized in that, The site accessibility calculation model is represented as follows: i and j represent public transportation stops and interests, respectively; C ij C represents the travel cost from public transport station i to point of interest j. ij The comprehensive walking cost model considering the walking environment is used to obtain the value; C0 is the travel perception time threshold from public transport station i to point of interest j; R i G represents the cumulative number of opportunities for public transport station i, i.e., walking accessibility; G represents the urban walking network.

9. The method for evaluating the accessibility of public transportation stations considering the travel environment according to claim 1, characterized in that, This method also includes calculating the influence coefficient: In the formula, R i The walkability of public transport station i under the psychological perception resistance of passengers; R i实 This represents the cumulative number of opportunities for public transport station i during actual walking time.

Images