Multi-mode fusion guide identification information grading dynamic collaborative layout method

Abstract

The invention discloses a multi-mode fusion guide identification information grading dynamic collaborative layout method, and relates to the technical field of guide identification planning. Multi-source data are collected and preprocessed; passenger behavior characteristics are analyzed based on the preprocessed multi-source data, wherein the passenger behavior characteristics comprise the moving characteristics of the passengers in the hub space, the staying characteristics of the passengers in the hub space and the decision behavior characteristics of the passengers in the hub space. Based on passenger behavior characteristics and real-time passenger flow prediction, space-time distribution of identifiers is dynamically adjusted, and the passenger way-finding time is shortened. Through information grading, information overload is avoided, and it is ensured that passengers rapidly obtain key guidance. And in combination with a passenger decision behavior model, the identification density is enhanced in an easily confused area. The average way-finding time of passengers can be greatly reduced, and the coverage rate of identification information is greatly improved. The congestion rate of the hub key nodes is reduced, and the passenger flow evacuation efficiency per unit time is improved.

Description

Technical Field

[0001] This invention relates to the field of wayfinding sign planning technology, and in particular to a hierarchical dynamic collaborative deployment method for wayfinding sign information that integrates multiple methods. Background Technology

[0002] The layout of signage and wayfinding information at passenger transport hubs is a bottleneck and key factor restricting the safe distribution function of these hubs, and it is also a research hotspot in the field of transportation engineering.

[0003] Hub buildings often feature multiple modes of transportation, exhibit diverse passenger behavior across different functional areas, and suffer from inconsistent standards and norms in signage for different modes of transport. Furthermore, the boundaries for multi-source video data collection and analysis are not uniform, leading to ambiguity in passenger choice behavior and regional passenger flow characteristics within the building space, making the construction of behavioral constraints difficult. Therefore, given the complex spatial arrangement of multiple modes of transportation and the trend of functional integration in hubs, it is essential to research scientific and quantitative methods for data collection and feature analysis. The wayfinding signage network, which takes into account the elements of passenger wayfinding behavior, exhibits dynamic hierarchical characteristics and requires information stability. However, under time-varying conditions, the nested, multi-flow, and continuous hierarchical structure of the information network, as well as the mechanism of the interaction between passenger behavior and the node network, are still unclear. This is one of the bottlenecks in the current intelligent deployment of wayfinding signage in hubs. The multi-mode guidance and dynamic identification information hierarchical deployment model faces a balance issue between commensurability and contradictions, affecting the efficiency of the solution algorithm. Currently, no stable, efficient, and convenient algorithm has been found. In summary, to meet the needs of safe and efficient passenger flow at hubs and high-quality travel at a higher standard, this study aims to provide intelligent and efficient guidance services for hubs. It focuses on the "network-behavior" elements of signage and guidance information under the integrated guidance services of multi-modal transportation functional areas within the hub. This reveals the dynamic characteristics of the interaction between passenger behavior and hierarchical guidance information. A dynamic hierarchical deployment method for guidance signage information, aimed at improving hub service efficiency, enhancing passenger travel experience, and ensuring the stability of the guidance system, is proposed. This method can support the safe and efficient passenger flow under the intelligent operation of integrated passenger transport hubs. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a hierarchical dynamic collaborative deployment method for wayfinding signage information that integrates multiple methods. The technical solution adopted is as follows: The hierarchical and dynamic collaborative deployment method for wayfinding signage information, which integrates multiple methods, includes the following steps: Step 1: Collect and preprocess multi-source data, which includes passenger flow data, traffic operation data, passenger behavior data, hub spatial data, and environmental data. Step 2: Analyze passenger behavior characteristics based on preprocessed multi-source data. Passenger behavior characteristics include passenger movement characteristics in the hub space, passenger dwell characteristics in the hub space, and passenger decision-making behavior characteristics in the hub space. Step 3, Constructing the Guided Information Network: The guidance information will be categorized based on passenger demand and hub functions; Based on the importance, urgency, and passenger needs, the wayfinding information is divided into different levels; The spatiotemporal distribution of wayfinding information is determined based on passenger behavior characteristics and the spatial layout of the hub; Step 4: Establish an interaction model between passenger behavior characteristics and guidance information; Step 5: Construct a multi-criteria objective function for the deployment of directional signs based on passenger behavior characteristics and interaction relationship model, and set constraints according to the actual situation of the hub and passenger needs; The optimization objectives of the multi-criteria objective function include: passenger experience optimization, hub efficiency improvement, signage reliability improvement, and signage stability improvement; Step 6: Use a genetic algorithm combined with an interaction relationship model to output the decision results of guidance and identification information for multiple optimization objectives.

[0005] Optionally, in step 1, an ARIMA model is established using historical passenger flow data to predict passenger flow density in future periods, and K-means clustering of passenger trajectories is used. By analyzing the proportion of different types of trajectories, critical path nodes are determined. Critical path nodes are the target points that are identified and deployed.

[0006] By adopting the above technical solutions, the spatiotemporal distribution of signs can be dynamically adjusted based on passenger behavior characteristics (movement path, dwell time) and real-time passenger flow prediction, thereby shortening the time for passengers to find their way.

[0007] By prioritizing information (such as emergency exit signage having the highest priority), information overload can be avoided, ensuring that passengers can quickly obtain key guidance.

[0008] By combining passenger decision-making behavior models (such as information search thresholds), the density of signage can be increased in easily confused areas (such as transfer nodes).

[0009] Simulation experiments show that the average time passengers spend finding their way can be significantly reduced, and the coverage of signage information can be greatly increased to over 90%.

[0010] By using ARIMA passenger flow forecasting and spatiotemporal network weight calculation, the signage deployment strategy can be adjusted in real time (such as adding temporary signs during peak hours).

[0011] Optimize the target area by improving signage and guidance routes in bottleneck areas (such as security checkpoints and turnstiles) to reduce pedestrian density. Integrate traffic operation data (such as train arrival and departure times) to link signage updates with traffic scheduling, thereby increasing the overall throughput of the hub.

[0012] Congestion rates at key hub nodes have decreased, and passenger flow dispersal efficiency per unit time has improved.

[0013] By dynamically feeding back multi-source data (environmental data, passenger behavior), abnormal situations (such as sudden passenger flow or equipment failure) can be identified, and backup deployment plans can be automatically switched.

[0014] During the optimization process of genetic algorithms, the cost of maintaining the identifier (such as budget constraints) should be considered to avoid resource waste caused by frequent identifier changes.

[0015] Introduce reliability indicators (such as shielding resistance and durability) into the objective function to ensure effective operation even under extreme conditions (such as power outages and high humidity).

[0016] By allocating weights to a multi-criteria objective function (passenger experience, efficiency, reliability, stability), short-term needs and long-term benefits can be balanced.

[0017] Precise deployment: Passenger trajectories are clustered using K-means to identify high-frequency path nodes and reduce the number of redundant markers (e.g., the density of markers in unnecessary areas is reduced by 50%).

[0018] Energy saving: Dynamically adjust the display frequency and brightness of electronic signs (based on ambient light data).

[0019] Optionally, in step 2, a Gaussian mixture model of movement speed distribution is established using historical passenger behavior data. This model is then used to extract the movement characteristics of passengers in the hub space. The core formula is as follows: ; in Here, v is the passenger flow density, K is the passenger speed, and K is the Gaussian component, where K=1 represents slow speed, K=2 represents medium speed, and K=3 represents fast speed. It is the weight of the i-th Gaussian component. It is the mean of the i-th Gaussian distribution. It is the standard deviation of the i-th Gaussian distribution; It is the probability density function of the i-th Gaussian distribution.

[0020] By adopting the above technical solution and establishing a Gaussian mixture model of movement speed distribution, passenger behavior at different speed levels can be described more accurately, thus providing a better understanding of passenger movement patterns in hub spaces. By setting different numbers of Gaussian components (K values) in the model, slow-moving, medium-moving, and fast-moving passengers can be effectively distinguished, providing a more detailed basis for hub space planning and passenger flow management.

[0021] By calculating the weight, mean, and standard deviation of each Gaussian component, the model can adaptively adjust to fit the actual data, improving the accuracy and reliability of predictions.

[0022] The model combines passenger flow density with passenger movement speed, revealing the intrinsic relationship between the two and providing a scientific basis for hub space design, passenger flow control, and safe evacuation.

[0023] Model-based passenger movement feature extraction can be used to optimize passenger flow allocation, reduce congestion, improve hub operation efficiency, and enhance the passenger travel experience.

[0024] Optionally, the Poisson distribution can be used to analyze the passenger dwell characteristics in the hub space. The core formula is as follows: ; in This represents the probability of a passenger staying in the hub space at time t, where t is the passenger's stay time in the hub space. It is the average number of times the device stays per unit time, and e is the natural constant.

[0025] By employing the aforementioned technical solutions and using Poisson distribution analysis to examine passenger dwell time characteristics within a hub, the probability distribution of passenger dwell time can be quantified, thus providing data support for hub space planning and management. The dwell time model based on Poisson distribution can predict the number of passenger dwell times within different time ranges, helping hub managers to rationally allocate service resources and facilities.

[0026] By analyzing passenger dwell characteristics, resource allocation within the hub space can be optimized, such as seating, rest areas, and catering services, thereby improving passenger satisfaction and hub operational efficiency.

[0027] Understanding the distribution of passenger dwell time helps in passenger flow management, enabling the development of corresponding passenger flow control strategies to reduce congestion and waiting time, and improve the passenger travel experience.

[0028] By predicting and regulating passenger dwell time, safety hazards within hub spaces can be reduced, such as overcrowding and disorder caused by prolonged stays.

[0029] Based on the analysis of dwell characteristics using the Poisson distribution, we provide data-driven decision support for hub managers, helping them make more scientific and rational planning and management decisions.

[0030] Optionally, Logistic regression can be used to model passenger decision-making behavior and extract passenger decision-making behavior features in the hub space. The core formula is as follows: ; It represents the probability that a passenger will make a route choice within time t in the hub space, where t is the information search time. This is the intercept term, representing the base decision probability without search time. It is the coefficient of time's influence on decision-making.

[0031] By employing the aforementioned technical solutions and modeling passenger decision-making behavior using Logistic regression, the probability of passengers making route choices within a hub space can be quantified, leading to a deeper understanding of the passenger decision-making process. The model can extract characteristics of passenger decision-making behavior within the hub space, particularly the impact of information search time on decision-making, providing crucial references for hub planning and management. Based on the Logistic regression model, the probability of passenger route choices under different information search times can be predicted, helping hub managers optimize route design and guidance measures. By analyzing the impact of time on decision-making, the way information is provided within the hub space can be optimized, such as the location of information displays and the frequency of content updates, thereby improving passenger decision-making efficiency.

[0032] Optionally, in step 3, a regression model is established to compare passenger behavior characteristics with the effectiveness of wayfinding information placement. A neural network or support vector machine is used to predict the impact of wayfinding information on passenger behavior. The model inputs include: passenger behavior characteristic vectors and wayfinding information placement scheme vectors. The model outputs include: passenger experience indicators, hub operation efficiency indicators, and signage usage effectiveness indicators.

[0033] Optionally, in step 4, the core formula of the interaction relationship model is as follows: ; in It represents the probability that a passenger chooses route k. It is the utility value of path k, determined by identifier visibility and path distance. It is the visibility score marked on path k. The k=1 flag indicates that the information is fully visible, while the k=0 flag indicates that the information is completely invisible. It is the influence coefficient of the visibility of the identifier. It is the influence coefficient of path distance.

[0034] By employing the aforementioned technical solutions and utilizing an interaction model, the probability of passengers choosing specific routes can be quantified, facilitating a deeper understanding of passenger route selection behavior within hub spaces. The model comprehensively considers two key factors: sign visibility and route distance, enabling a more holistic assessment of route attractiveness and providing a more accurate basis for hub planning. Analyzing the impact of sign visibility on route selection allows for optimization of the signage system design within hub spaces, improving sign visibility and guidance effectiveness, thereby enhancing passenger travel efficiency. The model's consideration of route distance's influence on passenger choice helps optimize route planning within hubs, designing more convenient and efficient travel routes.

[0035] Optionally, the core formula for the multi-criteria objective function is: ; in This is the average wayfinding time for passengers, calculated through simulation or historical data. It's the cost of congestion. , B is passenger flow density, and B is path width. This indicates coverage, which is the proportion of the area effectively represented within the passenger's field of vision. These are weighting coefficients. .

[0036] By adopting the above technical solutions, the multi-criteria objective function comprehensively considers multiple key factors such as average passenger wayfinding time, congestion costs, and signage coverage, thereby achieving comprehensive optimization of hub spatial planning.

[0037] Calculating the average wayfinding time for passengers using simulation or historical data and incorporating it into the objective function helps identify and optimize bottlenecks in the wayfinding process, thereby reducing passengers' wayfinding time.

[0038] Incorporating congestion costs into the objective function, taking into account the impact of passenger density and path width on congestion, helps to design more reasonable path layouts and reduce congestion and related costs.

[0039] By calculating the signage coverage rate, which is the proportion of effective signs within a passenger's field of vision, and incorporating it into the objective function, the design of the signage system can be optimized, improving the visibility and guidance efficiency of the signs.

[0040] Optional, the constraints for the multi-criteria objective function are as follows: The spacing constraints for the markers are: ;in Indicate the actual distance between i and j. It is the minimum allowable spacing; The budget constraint is: ; It is the cost of deploying marker k. It is a variable between 0 and 1. =1 indicates the deployment identifier k, and B is the total budget limit.

[0041] By adopting the above technical solution and constraining the spacing between signs, the minimum allowable spacing between signs is ensured, avoiding information overload and reduced readability caused by excessively dense signs, thereby improving the efficiency of passengers obtaining information.

[0042] Budget constraints ensured that the signage deployment plan was implemented within the total budget limit, achieving a reasonable allocation and effective use of resources and avoiding unnecessary waste.

[0043] Optionally, in step 6, a second-generation non-dominated sorting genetic algorithm is used to solve the multi-criteria optimization problem, and the fitness function is defined as follows: ; in It is the fitness function for identifying the deployment. It is a multi-objective function vector. As constraints, This means that when the constraint condition is met, It is the original objective function value , This indicates that when the constraint condition is not met, Through penalty items Adjustments will be made.

[0044] By adopting the above technical solution, the second-generation non-dominated sorting genetic algorithm (NSGA-II) is specifically designed to handle multi-objective optimization problems and can efficiently find the Pareto optimal solution set, that is, the optimal solution set that achieves a balance between different objectives.

[0045] By introducing constraints into the fitness function, we ensure that the generated solution not only meets the optimization objective but also conforms to practical constraints, such as marker spacing and budget limitations, thereby enhancing the practicality of the solution.

[0046] When constraints are not met, the objective function value is adjusted through a penalty term, which effectively guides the algorithm to search for a solution that better meets the constraints and avoids the generation of invalid solutions.

[0047] The algorithm can comprehensively consider multiple objectives (such as pathfinding time, congestion cost, and sign coverage) and find the optimal sign deployment scheme, thereby improving the overall efficiency of the sign system.

[0048] In summary, the present invention has at least one of the following beneficial technical effects: This invention provides a hierarchical and dynamic collaborative deployment method for wayfinding signage information, integrating multiple methods. Based on passenger behavior characteristics and real-time passenger flow prediction, it dynamically adjusts the spatiotemporal distribution of signs, shortening passenger wayfinding time. Information hierarchical classification avoids information overload, ensuring passengers quickly obtain key guidance. Combined with passenger decision-making behavior models, it enhances signage density in easily confused areas.

[0049] Simulation experiments show that the average wayfinding time for passengers can be significantly reduced, and the coverage of signage information can be greatly improved. The optimization objective is to optimize signage guidance routes in bottleneck areas and reduce pedestrian density. By integrating traffic operation data and linking signage updates with traffic scheduling, the overall throughput of the hub can be improved.

[0050] Congestion rates at key hub nodes have decreased, and passenger flow dispersal efficiency per unit time has improved. Dynamic feedback from multi-source data identifies anomalies and automatically switches to backup deployment plans.

[0051] The genetic algorithm optimization process takes into account the cost of identifier maintenance to avoid resource waste caused by frequent identifier changes. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating the hierarchical dynamic collaborative deployment method for directional signage information based on the multi-mode fusion of the present invention. Detailed Implementation

[0053] The present invention will be further described in detail below with reference to the accompanying drawings.

[0054] This invention discloses a hierarchical dynamic collaborative deployment method for wayfinding signage information that integrates multiple methods.

[0055] Reference Figure 1 Example 1, a hierarchical dynamic collaborative deployment method for wayfinding signage information integrating multiple methods, includes the following steps: Step 1: Collect and preprocess multi-source data, which includes passenger flow data, traffic operation data, passenger behavior data, hub spatial data, and environmental data. Step 2: Analyze passenger behavior characteristics based on preprocessed multi-source data. Passenger behavior characteristics include passenger movement characteristics in the hub space, passenger dwell characteristics in the hub space, and passenger decision-making behavior characteristics in the hub space. Step 3, Constructing the Guided Information Network: The guidance information will be categorized based on passenger demand and hub functions; Based on the importance, urgency, and passenger needs, the wayfinding information is divided into different levels; The spatiotemporal distribution of wayfinding information is determined based on passenger behavior characteristics and the spatial layout of the hub; Step 4: Establish an interaction model between passenger behavior characteristics and guidance information; Step 5: Construct a multi-criteria objective function for the deployment of directional signs based on passenger behavior characteristics and interaction relationship model, and set constraints according to the actual situation of the hub and passenger needs; The optimization objectives of the multi-criteria objective function include: passenger experience optimization, hub efficiency improvement, signage reliability improvement, and signage stability improvement; Step 6: Use a genetic algorithm combined with an interaction relationship model to output the decision results of guidance and identification information for multiple optimization objectives.

[0056] In Example 2, in step 1, an ARIMA model is established using historical passenger flow data to predict passenger flow density in future periods. K-means clustering of passenger trajectories is used, and the proportion of different types of trajectories is analyzed to determine critical path nodes. Critical path nodes are the target points that are marked and deployed.

[0057] Based on passenger behavior characteristics (movement path, dwell time) and real-time passenger flow prediction, the spatiotemporal distribution of signs is dynamically adjusted to shorten the time passengers spend finding their way.

[0058] By prioritizing information (such as emergency exit signage having the highest priority), information overload can be avoided, ensuring that passengers can quickly obtain key guidance.

[0059] By combining passenger decision-making behavior models (such as information search thresholds), the density of signage can be increased in easily confused areas (such as transfer nodes).

[0060] Simulation experiments show that the average time passengers spend finding their way can be significantly reduced, and the coverage of signage information can be greatly increased to over 90%.

[0061] By using ARIMA passenger flow forecasting and spatiotemporal network weight calculation, the signage deployment strategy can be adjusted in real time (such as adding temporary signs during peak hours).

[0062] Optimize the target area by improving signage and guidance routes in bottleneck areas (such as security checkpoints and turnstiles) to reduce pedestrian density. Integrate traffic operation data (such as train arrival and departure times) to link signage updates with traffic scheduling, thereby increasing the overall throughput of the hub.

[0063] Congestion rates at key hub nodes have decreased, and passenger flow dispersal efficiency per unit time has improved.

[0064] By dynamically feeding back multi-source data (environmental data, passenger behavior), abnormal situations (such as sudden passenger flow or equipment failure) can be identified, and backup deployment plans can be automatically switched.

[0065] During the optimization process of genetic algorithms, the cost of maintaining the identifier (such as budget constraints) should be considered to avoid resource waste caused by frequent identifier changes.

[0066] Introduce reliability indicators (such as shielding resistance and durability) into the objective function to ensure effective operation even under extreme conditions (such as power outages and high humidity).

[0067] By allocating weights to a multi-criteria objective function (passenger experience, efficiency, reliability, stability), short-term needs and long-term benefits can be balanced.

[0068] Precise deployment: Passenger trajectories are clustered using K-means to identify high-frequency path nodes and reduce the number of redundant markers (e.g., the density of markers in unnecessary areas is reduced by 50%).

[0069] Energy saving: Dynamically adjust the display frequency and brightness of electronic signs (based on ambient light data). In Example 3, step 2 involves establishing a Gaussian mixture model of passenger speed distribution using historical passenger behavior data. This model is then used to extract passenger movement characteristics within the hub space. The core formula is as follows: ; in Here, v is the passenger flow density, K is the passenger speed, and K is the Gaussian component, where K=1 represents slow speed, K=2 represents medium speed, and K=3 represents fast speed. It is the weight of the i-th Gaussian component. It is the mean of the i-th Gaussian distribution. It is the standard deviation of the i-th Gaussian distribution; It is the probability density function of the i-th Gaussian distribution.

[0070] By establishing a Gaussian mixture model of movement speed distribution, passenger behavior at different speed levels can be described more accurately, leading to a better understanding of passenger movement patterns in hub spaces. By setting different numbers of Gaussian components (K values) in the model, slow-moving, medium-moving, and fast-moving passengers can be effectively distinguished, providing a more detailed basis for hub space planning and passenger flow management.

[0071] By calculating the weight, mean, and standard deviation of each Gaussian component, the model can adaptively adjust to fit the actual data, improving the accuracy and reliability of predictions.

[0072] The model combines passenger flow density with passenger movement speed, revealing the intrinsic relationship between the two and providing a scientific basis for hub space design, passenger flow control, and safe evacuation.

[0073] Model-based passenger movement feature extraction can be used to optimize passenger flow allocation, reduce congestion, improve hub operation efficiency, and enhance the passenger travel experience.

[0074] Example 4 uses Poisson distribution to analyze passenger dwell characteristics in the hub space. The core formula is as follows: ; in This represents the probability of a passenger staying in the hub space at time t, where t is the passenger's stay time in the hub space. It is the average number of times the device stays per unit time, and e is the natural constant.

[0075] By employing the Poisson distribution to analyze passenger dwell characteristics in hub spaces, the probability distribution of passenger dwell time can be quantified, thus providing data support for hub space planning and management. A dwell time model based on the Poisson distribution can predict the number of passenger dwell times within different time ranges, helping hub managers to rationally allocate service resources and facilities.

[0076] By analyzing passenger dwell characteristics, resource allocation within the hub space can be optimized, such as seating, rest areas, and catering services, thereby improving passenger satisfaction and hub operational efficiency.

[0077] Understanding the distribution of passenger dwell time helps in passenger flow management, enabling the development of corresponding passenger flow control strategies to reduce congestion and waiting time, and improve the passenger travel experience.

[0078] By predicting and regulating passenger dwell time, safety hazards within hub spaces can be reduced, such as overcrowding and disorder caused by prolonged stays.

[0079] Based on the analysis of dwell characteristics using the Poisson distribution, we provide data-driven decision support for hub managers, helping them make more scientific and rational planning and management decisions.

[0080] Example 5 uses Logistic regression to model passenger decision-making behavior and extracts passenger decision-making behavior features in the hub space. The core formula is as follows: ; It represents the probability that a passenger will make a route choice within time t in the hub space, where t is the information search time. This is the intercept term, representing the base decision probability without search time. It is the coefficient of time's influence on decision-making.

[0081] By modeling passenger decision-making behavior using logistic regression, the probability of passengers making route choices within a hub can be quantified, leading to a deeper understanding of their decision-making process. The model can extract characteristics of passenger decision-making behavior within a hub, particularly the impact of information search time on decision-making, providing crucial insights for hub planning and management. Based on the logistic regression model, the probability of passenger route selection under different information search times can be predicted, helping hub managers optimize route design and guidance measures. By analyzing the impact of time on decision-making, the way information is provided within the hub, such as the location of information displays and the frequency of content updates, can be optimized, improving passenger decision-making efficiency.

[0082] In Example 6, step 3, a regression model is established to compare passenger behavior characteristics with the effectiveness of wayfinding information placement. A neural network or support vector machine is used to predict the impact of wayfinding information on passenger behavior. The model inputs include: passenger behavior characteristic vectors and wayfinding information placement scheme vectors. The model outputs include: passenger experience indicators, hub operation efficiency indicators, and signage usage effectiveness indicators.

[0083] In Example 7, step 4, the core formula of the interaction relationship model is as follows: ; in It represents the probability that a passenger chooses route k. It is the utility value of path k, determined by identifier visibility and path distance. It is the visibility score marked on path k. The k=1 flag indicates that the information is fully visible, while the k=0 flag indicates that the information is completely invisible. It is the influence coefficient of the visibility of the identifier. It is the influence coefficient of path distance.

[0084] Interaction models can quantify the probability of passengers choosing specific routes, facilitating a deeper understanding of passenger route selection behavior within hub spaces. The model comprehensively considers two key factors: sign visibility and route distance, enabling a more holistic assessment of route attractiveness and providing a more accurate basis for hub planning. Analyzing the impact of sign visibility on route selection allows for optimization of signage system design within hub spaces, improving sign visibility and guidance effectiveness, thereby enhancing passenger travel efficiency. The model's consideration of route distance's influence on passenger choice helps optimize route planning within hubs, designing more convenient and efficient travel routes.

[0085] Example 8: The core formula of the multi-criteria objective function is: ; in This is the average wayfinding time for passengers, calculated through simulation or historical data. It's the cost of congestion. , B is passenger flow density, and B is path width. This indicates coverage, which is the proportion of the area effectively represented within the passenger's field of vision. These are weighting coefficients. .

[0086] The multi-criteria objective function comprehensively considers several key factors such as average passenger wayfinding time, congestion costs, and signage coverage, achieving a comprehensive optimization of hub spatial planning.

[0087] Calculating the average wayfinding time for passengers using simulation or historical data and incorporating it into the objective function helps identify and optimize bottlenecks in the wayfinding process, thereby reducing passengers' wayfinding time.

[0088] Incorporating congestion costs into the objective function, taking into account the impact of passenger density and path width on congestion, helps to design more reasonable path layouts and reduce congestion and related costs.

[0089] By calculating the signage coverage rate, which is the proportion of effective signs within a passenger's field of vision, and incorporating it into the objective function, the design of the signage system can be optimized, improving the visibility and guidance efficiency of the signs.

[0090] Example 9: The constraints of the multi-criteria objective function are as follows: The spacing constraints for the markers are: ;in Indicate the actual distance between i and j. It is the minimum allowable spacing; The budget constraint is: ; It is the cost of deploying marker k. It is a variable between 0 and 1. =1 indicates the deployment identifier k, and B is the total budget limit.

[0091] By adopting the above technical solution and constraining the spacing between signs, the minimum allowable spacing between signs is ensured, avoiding information overload and reduced readability caused by excessively dense signs, thereby improving the efficiency of passengers obtaining information.

[0092] Budget constraints ensured that the signage deployment plan was implemented within the total budget limit, achieving a reasonable allocation and effective use of resources and avoiding unnecessary waste.

[0093] In Example 10, step 6, the second-generation non-dominated sorting genetic algorithm is used to solve the multi-criteria optimization problem. The fitness function is defined as follows: ; in It is the fitness function for identifying the deployment. It is a multi-objective function vector. As constraints, This means that when the constraint condition is met, It is the original objective function value , This indicates that when the constraint condition is not met, Through penalty items Adjustments will be made.

[0094] The second-generation non-dominated sorting genetic algorithm (NSGA-II) is specifically designed to handle multi-objective optimization problems and can efficiently find the Pareto optimal solution set, that is, the optimal solution set that achieves a balance between different objectives.

[0095] By introducing constraints into the fitness function, we ensure that the generated solution not only meets the optimization objective but also conforms to practical constraints, such as marker spacing and budget limitations, thereby enhancing the practicality of the solution.

[0096] When constraints are not met, the objective function value is adjusted through a penalty term, which effectively guides the algorithm to search for a solution that better meets the constraints and avoids the generation of invalid solutions.

[0097] The algorithm can comprehensively consider multiple objectives (such as pathfinding time, congestion cost, and sign coverage) and find the optimal sign deployment scheme, thereby improving the overall efficiency of the sign system.

[0098] The following specific embodiments illustrate the implementation principle of the present invention: A certain international airport integrated transportation hub includes multiple modes of transportation such as subway, airport express, and long-distance buses, with a daily passenger flow exceeding 500,000. Traditional static signage systems have the following problems: Severe congestion during peak hours: dense crowds at security checkpoints and transfer passages make it difficult for passengers to find their way; Signage overload: Signs for emergency exits, commercial facilities, etc. are mixed together, making it difficult to quickly identify key information; Poor dynamic adaptability: Unable to adjust guidance strategies in a timely manner when there is a sudden surge in passenger flow (such as flight delays).

[0099] Multi-source data acquisition and dynamic prediction: Real-time passenger flow sensor (to count the density of people in each area); Passenger mobile phone signaling data (extracting movement trajectory and dwell time); Flight / train arrival and departure timetables (forecasting future passenger flow fluctuations).

[0100] Use the ARIMA model to predict peak passenger flow in the next 2 hours (such as a surge in subway passenger flow 30 minutes after a flight arrives). K-means clustering was used to identify high-frequency passenger routes (such as “International Arrivals Hall → Subway Entrance → Security Checkpoint”).

[0101] Passenger behavior feature modeling: Movement speed tiers: The Gaussian Mixture Model (GMM) categorizes passengers into three groups: slow (shopping), medium (normal travel), and fast (catching a train), accounting for 30%, 50%, and 20% respectively. Fast passengers mainly congregate in the transfer corridor and need to be prioritized for guidance.

[0102] Dwell time analysis: Poisson distribution shows that the average dwell time in commercial areas is 15 minutes, indicating a need to increase the density of restaurant / toilet signage; If the waiting time at the security checkpoint exceeds 5 minutes, the electronic screen will dynamically display a message indicating the availability of an alternative lane.

[0103] Decision-making behavior optimization: Logistic models show that if passengers search for signs at intersections for more than 8 seconds, the probability of going the wrong way increases by 60%; adding ground projection signs at key intersections can shorten decision-making time.

[0104] Guiding Information Network Construction: Information Classification: Level 1 (Emergency): Fire exits and evacuation routes (red markings, highlighted 24 / 7); Level 2 (Core): Security checkpoints and boarding gates (dynamically updated queue times); Level 3 (Auxiliary): Commercial facilities, rest areas (displayed only during off-peak hours).

[0105] Spatiotemporal distribution strategy: During peak hours: Level 1 signs will be placed every 10 meters along the path from the subway entrance to the security checkpoint; In the event of a sudden flight delay, temporary "rebooking counter" signs will be added.

[0106] Multi-objective optimization and dynamic deployment: Objective function weight allocation: Peak hours: Passenger experience (50%), hub efficiency (40%), reliability (10%); Nighttime off-peak: Energy savings (30%), maintenance costs (40%), coverage (30%).

[0107] Genetic Algorithm Optimization: Inputs: Forecasted passenger flow, behavioral characteristics, and budget constraints (2 million yuan); Output: The Pareto optimal solution set selects the option of "reducing sign density by 20% and congestion rate by 18%"; Dynamic adjustment: The deployment plan is updated every 30 minutes based on real-time data.

[0108] Handling Abnormal Situations: Equipment malfunction: When the electronic signage in a certain area loses power, the sound and light guidance in the adjacent area will be automatically activated; Sudden surge in passenger flow: When large tour groups arrive, a backup channel will be temporarily opened and signage will be updated accordingly.

[0109] Implementation results: Passenger experience enhancement: The average wayfinding time decreased from 8.2 minutes to 5.5 minutes, a reduction of 33%. The coverage rate of signage increased from 75% to 92%, and the error rate at key nodes decreased by 45%.

[0110] Hub efficiency optimization: Security checkpoint congestion decreased by 22%, and peak-hour traffic speed increased by 25%. After the coordinated traffic scheduling was implemented, the delay rate between subway and flight connections decreased by 15%.

[0111] Operating cost control: The number of redundant identifiers is reduced by 40%, resulting in annual maintenance cost savings of 800,000 yuan. Electronic tags reduce energy consumption by 28%, saving 120,000 yuan in electricity costs annually.

[0112] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A hierarchical and dynamic collaborative deployment method for wayfinding signage information integrating multiple methods, characterized in that: Includes the following steps: Step 1: Collect and preprocess multi-source data, which includes passenger flow data, traffic operation data, passenger behavior data, hub spatial data, and environmental data. Step 2: Analyze passenger behavior characteristics based on preprocessed multi-source data. Passenger behavior characteristics include passenger movement characteristics in the hub space, passenger dwell characteristics in the hub space, and passenger decision-making behavior characteristics in the hub space. Step 3, Constructing the Guided Information Network: The guidance information will be categorized based on passenger demand and hub functions; Based on the importance, urgency, and passenger needs, the wayfinding information is divided into different levels; The spatiotemporal distribution of wayfinding information is determined based on passenger behavior characteristics and the spatial layout of the hub; Step 4: Establish an interaction model between passenger behavior characteristics and guidance information; Step 5: Construct a multi-criteria objective function for the deployment of directional signs based on passenger behavior characteristics and interaction relationship model, and set constraints according to the actual situation of the hub and passenger needs; The optimization objectives of the multi-criteria objective function include: passenger experience optimization, hub efficiency improvement, signage reliability improvement, and signage stability improvement; Step 6: Use a genetic algorithm combined with an interaction relationship model to output the decision results of guidance and identification information for multiple optimization objectives.

2. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 1, characterized in that: In step 1, an ARIMA model is built using historical passenger flow data to predict passenger flow density in future periods. K-means clustering of passenger trajectories is used, and the proportion of different types of trajectories is analyzed to determine critical path nodes. Critical path nodes are the target points that are marked for deployment.

3. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 1, characterized in that: In step 2, a Gaussian mixture model of movement speed distribution is established using historical passenger behavior data. This model is then used to extract the movement characteristics of passengers in the hub space. The core formula is as follows: ； in Here, v is the passenger flow density, K is the passenger speed, and K is the Gaussian component, where K=1 represents slow speed, K=2 represents medium speed, and K=3 represents fast speed. It is the weight of the i-th Gaussian component. It is the mean of the i-th Gaussian distribution. It is the standard deviation of the i-th Gaussian distribution; It is the probability density function of the i-th Gaussian distribution.

4. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 3, characterized in that: The Poisson distribution is used to analyze the dwell characteristics of passengers in the hub space. The core formula is as follows: ； in This represents the probability of a passenger staying in the hub space at time t, where t is the passenger's stay time in the hub space. It is the average number of times the device stays per unit time, and e is the natural constant.

5. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 4, characterized in that: Logistic regression is used to model passenger decision-making behavior and extract passenger decision-making behavior features in the hub space. The core formula is as follows: ; It represents the probability that a passenger will make a route choice within time t in the hub space, where t is the information search time. This is the intercept term, representing the base decision probability without search time. It is the coefficient of time's influence on decision-making.

6. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 5, characterized in that: In step 3, a regression model is established to compare passenger behavior characteristics with the effectiveness of wayfinding information placement. Neural networks or support vector machines are used to predict the impact of wayfinding information on passenger behavior. The model inputs include: passenger behavior characteristic vectors and wayfinding information placement scheme vectors. The model outputs include: passenger experience indicators, hub operation efficiency indicators, and signage usage effectiveness indicators.

7. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 6, characterized in that: In step 4, the core formula of the interaction relationship model is as follows: ； in It represents the probability that a passenger chooses route k. It is the utility value of path k, determined by identifier visibility and path distance. It is the visibility score marked on path k. The k=1 flag indicates that the information is fully visible, while the k=0 flag indicates that the information is completely invisible. It is the influence coefficient of the visibility of the identifier. It is the influence coefficient of path distance.

8. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 7, characterized in that, The core formula for the multi-criteria objective function is: ； in This is the average wayfinding time for passengers, calculated through simulation or historical data. It's the cost of congestion. , B is passenger flow density, and B is path width. This indicates coverage, which is the proportion of the area effectively represented within the passenger's field of vision. These are weighting coefficients. .

9. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 8, characterized in that, The constraints of the multi-criteria objective function are as follows: The spacing constraints for the markers are: ;in Indicate the actual distance between i and j. It is the minimum allowable spacing; The budget constraint is: ; It is the cost of deploying marker k. It is a variable between 0 and 1. =1 indicates the deployment identifier k, and B is the total budget limit.

10. The hierarchical dynamic collaborative deployment method for multi-mode integrated wayfinding signage information according to claim 9, characterized in that, In step 6, the second-generation non-dominated sorting genetic algorithm is used to solve the multi-criteria optimization problem. The fitness function is defined as follows: ； in It is the fitness function for identifying the deployment. It is a multi-objective function vector. As constraints, This means that when the constraint condition is met, It is the original objective function value , This indicates that when the constraint condition is not met, Through penalty items Adjustments will be made.

Images