Resident travel feature and satisfaction analysis method and system based on different rural types

Abstract

The invention relates to the technical field of urban and rural traffic planning, resident travel behavior analysis and satisfaction evaluation, in particular to a resident travel characteristic and satisfaction comprehensive analysis method and system based on different rural types. According to the method and the system, the rural areas are divided into different types by comprehensively considering a plurality of factors such as the distance from the rural areas to a business center, the number of permanent residents, the regional area, the number of small and medium-sized enterprises, the number of large enterprises and the number of supermarkets and supermarkets, and the personal attributes, the travel attributes and the travel satisfaction of different types of rural residents are deeply analyzed. And finally, the method further focuses on the travel satisfaction degree of the rural residents, and modeling and verification are carried out on the relationship between the personal attributes, the travel attributes, the travel satisfaction degree and other factors of the rural residents and the travel satisfaction degree by adopting a second-order factor structure equation model (SFSEM). And a scientific basis is provided for optimizing rural traffic planning and improving the travel satisfaction degree of rural residents.

Description

Technical Field

[0001] The present invention relates to the technical fields of urban and rural traffic planning, resident travel behavior analysis and satisfaction evaluation, and in particular to a method and system for comprehensively analyzing travel characteristics and satisfaction of residents based on different rural types. Background Art

[0002] With the acceleration of the process of urban-rural integration, the travel needs of rural residents are increasing, and travel modes and habits are showing a trend of diversification. However, due to the significant differences in geographical, economic, and social conditions in rural areas, different types of rural areas show obvious differences in residents' travel characteristics. Traditional transportation planning methods often ignore this difference, resulting in transportation facilities and services being unable to meet the actual needs of rural residents, which in turn affects their travel satisfaction and quality of life.

[0003] At present, research on residents' travel characteristics and satisfaction has made some progress, but most of them are concentrated in cities or a specific type of rural areas, lacking systematic analysis and comparison of different types of rural areas. In addition, existing studies often use a single geographical or economic indicator when classifying rural types, ignoring the diversity and complexity of rural development.

[0004] In view of this, there is an urgent need to design a system and method that can comprehensively consider multiple factors, classify rural areas in a refined manner, and deeply analyze the travel characteristics and satisfaction of various types of rural residents. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a method and system for analyzing the travel characteristics and satisfaction of residents based on different types of rural areas. The method and system divide rural areas into different types by comprehensively considering multiple factors such as the distance from the rural area to the commercial center, the number of permanent residents, the regional area, the number of small and medium-sized enterprises, the number of large enterprises, and the number of supermarkets, and conducts in-depth analysis of the personal attributes, travel attributes, and travel satisfaction of different types of rural residents. Finally, the present invention also focuses on the travel satisfaction of rural residents, and models and verifies the relationship between factors such as personal attributes, travel attributes, and travel satisfaction of rural residents and travel satisfaction by adopting the Second-Order Factor Structural Equation Modeling (SFSEM). Provide a scientific basis for optimizing rural transportation planning and improving the travel satisfaction of rural residents.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] In a first aspect, the present application provides a method for classifying rural areas based on an agglomerative hierarchical clustering algorithm (AHC), comprising the following steps:

[0008] S1: Collect geographical, economic, and social data of the target rural areas, including but not limited to the distance from the rural area to the commercial center, the number of permanent residents, the area, the number of small and medium-sized enterprises, the number of large enterprises, and the number of supermarkets. Clean and organize the collected data to ensure the accuracy and completeness of the data;

[0009] S2: Agglomerative hierarchical clustering model (AHC) is used to classify rural areas. This model can gradually aggregate rural areas into different categories based on the similarity between data. In the process of model construction, it is necessary to determine parameters such as the number of clusters, clustering criteria, and clustering algorithm;

[0010] S3: Use cross-validation, cluster validity index and other methods to verify and optimize the classification results to ensure the accuracy and reliability of the classification results. At the same time, based on the actual situation, make necessary adjustments and optimizations to the classification results to better reflect the actual differences in rural areas.

[0011] S4: Based on the agglomerative hierarchical clustering algorithm, rural areas are classified into four types.

[0012] Furthermore, based on the seventh national census and the China County Yearbook (Township Volume), data such as the distance between each township and the commercial center (km), regional area (hectares), number of permanent residents (persons), number of small and medium-sized enterprises (units), number of large enterprises (units), and number of supermarkets (units) were obtained.

[0013] Furthermore, each data is regarded as a cluster, and different samples are merged according to certain clustering conditions to eventually form a cluster or reach the number of clusters.

[0014] Furthermore, the Ward method (mean sum of deviations), a standard function commonly used in AHC, is used to measure the degree of fusion of two clusters. The Ward method focuses on the sum of squared errors generated when two clusters are merged, and its calculation formula is as follows:

[0015]

[0016] Where: a k,i is the arithmetic square root of the distance from the i-th sample of the k-th cluster to the origin of the coordinate system; a k The sample mean of the kth cluster; D is the sum of squared errors between two clusters; g is the number of clusters before fusion.

[0017] Furthermore, the rural areas are clustered based on the agglomerative hierarchical clustering algorithm and divided into four rural types: rural areas of type I, rural areas of type II, rural areas of type III and rural areas of type IV.

[0018] On the other hand, the present application provides a method for investigating the travel characteristics and satisfaction of rural residents, the method comprising the following steps:

[0019] S1: According to the classification results of rural types, a targeted questionnaire was designed. The questionnaire content covers personal attributes of residents (such as gender, age, education level, annual income, number of cars owned, etc.), travel attributes (such as travel purpose, travel frequency, travel time, travel mode, travel cost, etc.) and travel satisfaction (such as comfort, safety, convenience, economy, punctuality, etc.);

[0020] S2: Use a combination of online and offline methods to survey various types of rural residents. Online surveys are conducted through social media, emails, etc., while offline surveys are conducted through on-site visits and distribution of paper questionnaires. The collected data needs to be cleaned and organized to ensure the authenticity and validity of the data;

[0021] S3: The internal consistency of each dimension of the questionnaire was analyzed using reliability test methods such as Cronbach α. At the same time, the validity of the questionnaire was tested using methods such as exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) to ensure that the measurement results of the questionnaire can accurately reflect the actual travel characteristics and satisfaction of residents.

[0022] Furthermore, the Cronbachα value range is between 0-1. The higher the test result coefficient value, the higher the reliability. The reliability coefficient greater than 0.7 is considered reliable. The EFA test results are evaluated using the KMO (Kaiser-Meyer-Olkin) statistic and the Bartlett sphericity test. It is generally believed that when the KMO test coefficient is ≥0.7, the data is suitable for factor analysis. AVE (Average Variance Extracted) and CR (Construct Reliability) are selected to measure the convergent validity and combined reliability of the scale. Under the premise that the CFA model has good fit, the convergent validity and combined reliability of each dimension of the scale will be further tested. AVE (Average Variance Extracted) and CR (Construct Reliability) are selected to measure the convergent validity and combined reliability of the scale. The AVE value reaches 0.5 and the CR value reaches 0.7, indicating good convergent validity and combined reliability. The calculation formulas for AVE and CR are as follows:

[0023] AVE=(∑λ 2) / n

[0024] CR=(∑λ 2 ) / ∑λ 2 +∑δ

[0025] Where: λ represents the standardized factor loading value; n represents the number of measurement indicators of the factor; δ represents the standardized residual.

[0026] Finally, the present application provides a rural residents travel characteristics and satisfaction analysis system, the system comprising the following steps:

[0027] S1: Conduct statistical analysis on the travel data of residents in different rural types. The analysis includes but is not limited to the characteristics of residents' travel purpose, travel frequency, travel time, travel mode and travel cost. Through comparative analysis, reveal the differences and patterns in travel characteristics of different types of rural residents.

[0028] S2: The second-order factor structural equation model (SFSEM) is used to analyze residents' travel satisfaction. This model can take into account the correlation between latent variables and more accurately evaluate residents' satisfaction with travel services. Through analysis, the differences and influencing factors in travel satisfaction among different types of rural residents are revealed.

[0029] S3: Use visualization tools such as charts and maps to intuitively display the analysis results. At the same time, interpret and discuss the analysis results in light of the actual situation, and put forward targeted improvement measures and suggestions.

[0030] Furthermore, the second-order factor structural equation model usually consists of two parts. The calculation formula of the factor model is as follows:

[0031] y=Λη+ε

[0032] Where y is the measurement indicator vector of the latent variable, that is, the preset measurement problem; Λ is the unknown parameter matrix to be estimated; ε is the measurement error term.

[0033] η=Γ i x i +ξ

[0034] Where η is the first-order latent variable vector, i.e., the travel satisfaction of rural residents, including comfort, safety, convenience, economy and punctuality; x i is the vector of exogenous observable variables that affect the latent variables, namely personal attributes and travel attributes; Γ i is the unknown parameter matrix to be estimated; ξ is the first-order factor measurement error.

[0035] The second-order factor model is represented by the first-order factors and the error term:

[0036] ρ=βη+ζ

[0037] Where: ρ is the vector of second-order latent variables, that is, the overall satisfaction of residents' travel; β is the unknown parameter matrix to be estimated; ζ is the second-order factor measurement error term.

[0038] The beneficial effects of the present invention include:

[0039] In order to more accurately understand the travel behavior of rural residents, the present invention first takes into account multiple factors from the perspective of residents' travel, such as the distance from the rural area to the commercial center, the number of permanent residents, the regional area, the number of small and medium-sized enterprises, the number of large enterprises, and the number of supermarkets, and divides the rural areas into four categories. On this basis, the present invention uses a questionnaire to deeply analyze the travel characteristics of rural residents from the aspects of personal attributes, travel attributes, and travel satisfaction of residents of different types of rural areas. Finally, the present invention also focuses on the travel satisfaction of rural residents. The relationship between personal attributes, travel attributes, and travel satisfaction of rural residents and travel satisfaction was modeled and verified by using a second-order factor structural equation model. The results of the present invention provide effective measures for targeted improvement of travel satisfaction of different types of rural residents. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A flow chart of a method for analyzing travel characteristics and travel satisfaction of residents based on different rural types provided by the present invention;

[0041] Figure 2 It is a graph of aggregation coefficients of the agglomerative hierarchical clustering algorithm in a specific implementation manner of the present invention;

[0042] Figure 3 It is a Ward pedigree diagram of rural classification in a specific embodiment of the present invention;

[0043] Figure 4 It is a CFA model diagram of the residents' travel satisfaction scale in a specific implementation manner of the present invention;

[0044] Figure 5 This is a diagram of the SFSEM model standardized path calibration result in a specific implementation manner of the present invention; DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings of the present invention specification. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other implementation methods obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0046] The embodiment of the present invention provides a method for analyzing the travel characteristics and travel satisfaction of residents based on different rural types, such as Figure 1 As shown, the following will refer to Figure 1 The specific implementation steps of the method of the present invention are described.

[0047] S101: Based on the data of the seventh national census and the China County Yearbook (Township Volume), the distance (km) between each township and the commercial center, the area (hectares), the number of permanent residents (persons), the number of small and medium-sized enterprises (units), the number of large enterprises (units), and the number of supermarkets (units) in Zaoqiang County;

[0048] In this specific implementation, Zaoqiang County of Hengshui City is selected as the research area. According to the data of the seventh national census and the China County Yearbook (township volume), the distance (km), regional area (hectares), number of permanent residents (people), number of small and medium-sized enterprises (units), number of large enterprises (units) and number of supermarkets (units) of each township in Zaoqiang County from the commercial center, as shown in Table 1. Since the commercial center of Zaoqiang County is located in Zaoqiang Town, the distance from Zaoqiang Town to the commercial center is 0. SPSS software is used to classify the townships in Zaoqiang County according to Figure 2 From the aggregation coefficient line chart in , we can see that when the number of categories is 4, the downward trend of the line slows down, so the number of categories can be set to 4. Figure 3 It can be seen that Zaoqiang Town is classified into a separate category, Daying Town is classified into a separate category, Matun Town, Zhangxiutun Town and Xintun Town are classified into a separate category, Encha Town, Jiahui Town, Xiaozhang Town, Tanglinzhen, Wangchang Township and Wangjun Township are classified into a separate category.

[0049] Table 1 Data of each township in Zaoqiang County

[0050]

[0051] S102: Obtain residents’ travel data through questionnaire surveys, including residents’ personal attributes, travel attributes, and travel satisfaction;

[0052] The travel data of rural residents in Zaoqiang County from September 28, 2014 to October 14, 2024 were obtained through questionnaire surveys, and a total of 903 residents were surveyed. After eliminating questionnaires with poor quality such as those with a longer filling time than the average level and continuous selection of extreme values ​​and the same value, 866 valid questionnaires were collected, with an efficiency of 95.90%. The survey included residents' personal attributes, travel attributes, and travel satisfaction, as shown in Tables 2 and 3.

[0053] Likert scale, also known as Likert scale, is the most commonly used type of scoring and summing scale. It consists of a group of statements, each of which has five answers, namely "very satisfied", "satisfied", "average", "unsatisfied", and "very unsatisfied", usually recorded as 5, 4, 3, 2, and 1 respectively. It can reflect the respondents' comprehensive views on specific things. In order to understand residents' satisfaction with travel services and thus improve the quality of travel services and residents' travel experience, it is particularly important to use scientific and effective measurement tools for evaluation. The Likert scale can meet this need.

[0054] Table 2 Personal attributes and travel attributes

[0055]

[0056]

[0057] Table 2 Travel satisfaction

[0058]

[0059] S103: Conduct reliability and validity analysis on the residents' travel satisfaction scale;

[0060] The present invention uses SPSS software to perform reliability analysis on the resident travel satisfaction scale. The results of the reliability analysis are shown in Table 3. From the results, it can be seen that the α coefficient value of each dimension of the questionnaire is greater than 0.7, indicating that the measurement indicators have high internal consistency. In addition, the resident travel satisfaction scale is subjected to EFA using SPSS software, as shown in Table 4. From the calculation results, the load coefficients of each indicator are all above 0.7, indicating that the explanatory power of the measurement variables is good and the structure between the latent variables and their measurement variables is reasonable. Then CFA is performed on the travel satisfaction scale, and the AMOS software is used to test the fitness of the CFA model, as shown in Table 4. Figure 4 The model fit test results are shown in Table 5. All indicators meet the requirements, indicating that the travel satisfaction CFA model has good fit. According to the analysis results in Table 6, it can be seen that the AVE values ​​of each dimension are all above 0.5, and the CR values ​​are all above 0.7. It can be seen that each dimension has good convergent validity and combined reliability.

[0061] Table 3 Reliability analysis results

[0062]

[0063] Table 4 Validity analysis results

[0064]

[0065] Table 5 Model fitness test

[0066]

[0067] Table 6 Convergent validity and combined reliability of each dimension

[0068]

[0069] S103: The second-order factor structural equation model (SFSEM) is used to analyze residents’ travel satisfaction.

[0070] This paper uses Mplus8.3 software to construct a rural residents' travel satisfaction model. The principle is to use the MLE algorithm to calculate the internal influence mechanism between the potential factors of residents' travel, such as Figure 5 In addition, the SFSEM is tested for fitness, as shown in Table 7. As can be seen from Table 7, the model meets the fitness evaluation criteria. Figure 5 It can be seen that personal attributes and travel attributes have a significant positive correlation with comfort, safety, convenience, economy and punctuality. The path coefficients of personal attributes and travel attributes for comfort are 0.689 and 0.622 respectively; the path coefficients for safety are 0.733 and 0.278 respectively; the path coefficients for convenience are 0.635 and 0.845 respectively; the path coefficients for economy are 0.690 and 0.437 respectively; and the path coefficients for punctuality are 0.622 and 0.284 respectively. In addition, comfort, safety, convenience, economy and punctuality have a significant positive correlation with overall satisfaction. The path coefficients of comfort, safety, convenience, economy and punctuality for overall satisfaction are 0.729, 0.699, 0.682, 0.626 and 0.624 respectively.

[0071] Table 7 SFSEM model fitness test

[0072]

[0073] An embodiment of the present invention is described in detail above, but the described content is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present invention, which should be covered within the protection scope of the present invention.

Claims

1. A method for analyzing the travel characteristics and travel satisfaction of residents based on different rural types, the specific implementation steps are as follows: S1: Rural areas are classified according to the data of the distance between each township and the commercial center, regional area, number of permanent residents, number of small and medium-sized enterprises, number of large enterprises, and number of supermarkets in the seventh national census and the township volume of the China County Yearbook; S2: Obtain residents’ travel data through questionnaire surveys, including residents’ personal attributes, travel attributes, and travel satisfaction; S3: Conduct reliability and validity analysis on the residents’ travel satisfaction scale; S4: A second-order factor structural equation model was used to analyze residents’ travel satisfaction.

2. According to the method of claim 1, obtaining data of each township comprises: Data on the distance between each township and the commercial center, regional area, permanent population, number of small and medium-sized enterprises, number of large enterprises, and number of supermarkets; The agglomerative hierarchical clustering model is used to classify rural areas. This model can gradually aggregate rural areas into different categories based on the similarities between data. Where: a k,i is the arithmetic square root of the distance from the i-th sample of the k-th cluster to the origin of the coordinate system; a k The sample mean of the kth cluster; D is the sum of squared errors between two clusters; g is the number of clusters before fusion.

3. The method according to claim 1, obtaining residents' travel data through questionnaire surveys, including residents' personal attributes, travel attributes and travel satisfaction, including: Residents’ personal attributes: gender, age, education level, annual income, number of cars owned; Travel attributes: travel purpose, travel frequency, travel time, travel mode, and travel expenses; Travel satisfaction: comfort, safety, convenience, economy, and punctuality.

4. According to the method of claim 1, reliability and validity analysis of the residents' travel satisfaction scale is performed, including: The Cronbach's coefficient reliability test method was used to analyze the internal consistency of each dimension of the questionnaire. Exploratory factor analysis and confirmatory factor analysis were used to test the validity of the questionnaire to ensure that the measurement results of the questionnaire could accurately reflect the actual travel characteristics and satisfaction of residents. The calculation formulas for AVE and CR are as follows: Where: λ represents the standardized factor loading value; n represents the number of measurement indicators of the factor; δ represents the standardized residual.

5. According to the method of claim 1, a second-order factor structural equation model is used to analyze residents' travel satisfaction, including: y=Λη+ε In the formula, y is the measurement indicator vector of the latent variable, that is, the preset measurement problem; Λ is the unknown parameter matrix to be estimated; ε is the measurement error term. h = C i x i +ξ Where η is the first-order latent variable vector, i.e., the travel satisfaction of rural residents, including comfort, safety, convenience, economy and punctuality; x i is the vector of exogenous observable variables that affect the latent variables, namely personal attributes and travel attributes; Γ i is the unknown parameter matrix to be estimated; ξ is the first-order factor measurement error. The second-order factor model is represented by the first-order factors and the error term: ρ=βη+ζ Where: ρ is the vector of second-order latent variables, that is, the overall satisfaction of residents' travel; β is the unknown parameter matrix to be estimated; ζ is the second-order factor measurement error term.

Images