Modern tourism resource investigation method and system based on big data

Abstract

The present application relates to big data and tourism resource management technical field, specifically for modern tourism resource investigation method and system based on big data, including: tourism resource perception module, data aggregation analysis module, constraint factor extraction module, dynamic deduction planning module and blueprint instruction execution module. The system collects multi-source data in real time through interface and sensor, constructs multi-dimensional space-time state tensor through semantic alignment to accurately describe resource state; based on coupling rule analysis tensor extracts core constraint factor, and uses dynamic framework containing positive and negative feedback path to carry out quantitative deduction and simulation intervention, generates multi-stage optimization configuration blueprint; finally, the blueprint is converted into instruction sequence to drive automatic execution. The present application realizes panoramic perception of tourism resource state, intelligent identification of core constraint factor, and dynamic planning and automatic execution of resource configuration scheme, improves the accuracy, adaptability and operation efficiency of resource management.

Description

Technical Field

[0001] This invention relates to the field of big data and tourism resource management technology, and in particular to modern tourism resource survey methods and systems based on big data. Background Technology

[0002] Traditional tourism resource surveys and management primarily rely on manual on-site inspections, historical data statistics, and report analysis. Existing technological systems often employ single sensor networks or tourist questionnaires for data collection, resulting in limited data sources, diverse data types, and delayed updates, hindering the formation of a real-time, continuous, and comprehensive understanding of resource status. At the data processing and analysis level, conventional methods lack the ability to deeply integrate multi-source, heterogeneous data. Data generated by different systems and devices differs in format, semantics, and spatiotemporal references, creating data silos and information islands. This prevents the construction of a unified and complete panoramic view of resource status, limiting the depth and accuracy of analytical decision-making.

[0003] In the planning and management decision-making stages, existing technologies primarily rely on static rule models or past experience for resource allocation, failing to effectively model the complex dynamic relationships and interactive feedback mechanisms among multiple elements within the resource system. This makes it difficult for the system to identify and quantify the core constraints affecting the sustainable use of resources and service quality from massive amounts of data, let alone dynamically simulate or proactively intervene in the future evolution of these factors. Consequently, existing solutions often only passively respond to existing problems, lacking adaptability and foresight in resource allocation strategies, and are ineffective in dealing with dynamic scenarios such as demand fluctuations and environmental changes. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a modern tourism resource survey method and system based on big data.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a modern tourism resource survey system based on big data, comprising: The tourism resource sensing module captures and aggregates heterogeneous tourism resource sensing data streams in real time through application programming interfaces deployed on mobile terminals and physical sensors deployed on tourism resource nodes. The data aggregation and analysis module performs multi-scale aggregation and semantic alignment operations on the tourism resource perception data stream to construct a multi-dimensional spatiotemporal state tensor describing the state of tourism resources. The constraint factor extraction module performs hierarchical decoupling and correlation analysis on the multidimensional spatiotemporal state tensor based on preset resource coupling rules, and extracts a set of core constraint factors affecting the effectiveness of tourism resources. The dynamic simulation planning module utilizes a dynamic adaptive adjustment framework that includes positive and negative feedback paths to quantitatively simulate and intervene in the set of core constraint factors, forming a blueprint for the optimal allocation of tourism resources that includes multi-stage planning nodes. The blueprint instruction execution module converts the tourism resource optimization configuration blueprint into an executable sequence of digital instructions, driving automated facilities to perform corresponding resource configuration and guidance actions.

[0006] Preferably, the step of performing multi-scale aggregation and semantic alignment operations on the tourism resource perception data stream to construct a multi-dimensional spatiotemporal state tensor describing the state of tourism resources specifically includes: The sequence of tourist locations from mobile terminals and the node environmental readings from physical sensors are aligned and stitched together according to a unified timestamp reference to generate spatiotemporally synchronized raw data frames. The original data frame is subjected to gridded downsampling in the spatial domain and event slicing in the temporal domain, respectively, to generate spatial slice sets and temporal segment sets with different granularities. Perform geographic feature-based regional clustering on the spatial slice set to identify interest region clusters with similar tourist distribution patterns; Simultaneously, the time segment set is divided into periods based on behavioral patterns to mark standard time periods with stable tourism activity characteristics; The boundary information of the interest region cluster is combined with the start and end information of the standard time period by performing a Cartesian product to generate a data cube for each combination of region and time period. Feature extraction and normalization are performed on all environmental readings and location sequences corresponding to the same data cube, and these are then filled into different dimensions of the data cube, ultimately stacked to form the multidimensional spatiotemporal state tensor.

[0007] Preferably, the hierarchical decoupling and correlation analysis of the multidimensional spatiotemporal state tensor based on preset resource coupling rules extracts a set of core constraint factors affecting the effectiveness of tourism resources, specifically including: The density dimension reflecting passenger flow pressure, the capacity dimension reflecting facility load, and the quality dimension reflecting environmental experience are extracted from the multidimensional spatiotemporal state tensor. Based on the nonlinear mapping relationship between density and capacity defined in the resource coupling rules, the critical areas where passenger flow density is close to the facility carrying capacity limit within the interest area cluster are calculated, and such areas are marked as first-level constraint factors. Based on the capacity and quality attenuation transmission relationship defined in the resource coupling rules, calculate the transmission chain of environmental quality perception decline caused by facility overload within the critical area, and mark the key attenuation nodes in the chain as secondary constraint factors. Based on the negative feedback relationship between quality and density defined in the resource coupling rule, the probability of tourists spontaneously leaving or avoiding a specific area due to a decline in perceived environmental quality is calculated, and high-probability areas are marked as third-level constraint factors. The critical regions, key decay nodes, and high-probability regions marked at all levels are combined to form the core constraint factor set.

[0008] Preferably, the use of a dynamic adaptive adjustment framework including positive and negative feedback paths to quantitatively extrapolate and simulate the core constraint factor set, forming a tourism resource optimization allocation blueprint containing multi-stage planning nodes, specifically includes: In the dynamic adaptive adjustment framework, a positive feedback path is set with the goal of improving the overall resource utilization rate. The positive feedback path is observed to increase the service capacity of the key attenuation nodes by simulation and to observe its chain effect on alleviating the pressure in the critical area. In the dynamic adaptive regulation framework, a negative feedback path with the goal of maintaining system stability is set. The negative feedback path is observed to reduce local density and improve the perception of environmental quality by simulating the implementation of tourist diversion in the high-probability area. The simulation process of alternating between the positive feedback path and the negative feedback path is carried out, and the change of the multidimensional spatiotemporal state tensor after each simulation intervention is recorded, and the deviation between the change and the expected target is evaluated. When the deviation of the simulation is lower than the preset convergence threshold in multiple consecutive simulations, the system is determined to have reached a quasi-steady state, and all triggered intervention measures and their execution parameters within the framework at this time are extracted. According to the time sequence of the simulated execution, the intervention measures and their execution parameters are arranged into a list of stage tasks with sequential dependencies, which are the multi-stage planning nodes in the blueprint for optimizing the allocation of tourism resources.

[0009] Preferably, the step of converting the tourism resource optimization blueprint into an executable sequence of digital instructions to drive automated facilities to perform corresponding resource configuration and guidance actions specifically includes: Each stage planning node in the tourism resource optimization allocation blueprint is analyzed and broken down into operation instruction elements for specific geographical coordinates or facility numbers; Each operation instruction element is matched with a preset facility control protocol template and filled with execution parameters defined by the stage planning node to generate standard equipment control instructions; According to the timing sequence defined by the stage planning nodes, all the device control instructions are sorted, and system status verification instructions are inserted between adjacent instructions to form the executable digital instruction sequence. The digital instruction sequence is sent in batches to the core controller of the corresponding smart traffic guidance screen, variable information sign or automatic service facility; The core controller executes the received instructions, specifically by adjusting the information display content, changing the physical guidance path, or reallocating service resources.

[0010] Preferably, inserting a system status verification instruction between adjacent instructions specifically includes: After a device control command is sent, a system status verification command is triggered. The system status verification command obtains the instantaneous status reading of the target area of ​​the command execution by calling the real-time data interface of the physical sensor. The instantaneous status reading is compared with the expected status threshold reached by the system status verification command to generate a status consistency verification result. If the state consistency verification result is as expected, then the digital instruction sequence is allowed to continue issuing the next device control instruction; If the state consistency verification result deviates from the expectation, the instruction adjustment sub-process is triggered. The instruction adjustment sub-process dynamically fine-tunes the execution parameters in the next device control instruction to be executed, or inserts a compensatory additional control instruction, based on the degree and direction of the deviation.

[0011] Preferably, the trigger instruction adjustment sub-process specifically includes: Analyze the specific dimensions in which the state consistency verification results deviate from expectations, whether it is the passenger flow density dimension, the facility load capacity dimension, or the environmental perception quality dimension; Based on the dimension of deviation, retrieve the most matching successful amendment case from the historical intervention case library, and extract the parameter adjustment strategy adopted in the successful amendment case; The extracted parameter adjustment strategy is fused with the current device control command to be executed to calculate a new set of execution parameters with corrective intent. The new execution parameters are used to overwrite the parameters in the original instruction, or a new compensation instruction containing the new parameters is generated and inserted into the corresponding position in the original instruction sequence. The adjusted instruction sequence fragment is resubmitted to the system state verification instruction for review to ensure that the adjusted expected state can converge to the target range.

[0012] Preferably, the method further includes a closed-loop optimization module: After a complete resource allocation cycle is completed, collect the actual effect data stream generated by the actions performed by the automated facilities; The actual effect data stream is compared with the tourism resource optimization and allocation blueprint driving this cycle in all dimensions, and the gap map between the blueprint prediction value and the actual value is calculated. The source analysis of the gap map identifies the key links with large prediction deviations. These key links exist in the construction process of the multidimensional spatiotemporal state tensor, the extraction process of core constraint factors, or the deduction process of dynamic adaptive adjustment. Based on the results of the source tracing analysis, the mapping relationship parameters in the resource coupling rules are corrected in reverse, or the weight coefficients of the positive and negative feedback paths in the dynamic adaptation adjustment framework are adjusted. The newly acquired tourism resource perception data stream is processed using the revised rules and framework to begin the next cycle of resource allocation optimization.

[0013] Preferably, the step of performing source analysis on the gap map to identify key links with large prediction deviations specifically includes: The gap map is decomposed according to the spatiotemporal dimension to locate the clusters of interest where the deviations are concentrated in space and the standard time periods where they are concentrated in time. For the deviation concentration area identified by spatiotemporal positioning, the process of marking the core constraint factors corresponding to the deviation concentration area is traced back when the tourism resource optimization allocation blueprint is formed; Verify whether the data in the multidimensional spatiotemporal state tensor on which the core constraint factor is based is complete, and whether the mapping relationship applied here in the resource coupling rule is accurate; Further tracing back to the process of constructing the multidimensional spatiotemporal state tensor, we examine whether there is information loss or distortion in the aggregation, alignment, and feature extraction operations from the original data stream to the data cube; By tracing back layer by layer, the final prediction deviation is attributed to at least one specific step in data capture, data aggregation, state representation, rule application, or inference simulation, and this step is marked as the key step to be optimized.

[0014] Preferably, the present invention also includes a modern tourism resource survey method based on big data, which is applied to the modern tourism resource survey system based on big data as described above.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: By introducing a dynamic adaptive adjustment framework incorporating both positive and negative feedback paths, the system can continuously perform quantitative simulations and intervention tests on the extracted core constraints. Positive feedback paths are used to identify and reinforce growth cycles in resource utilization, while negative feedback paths are used to monitor and suppress decay cycles that may lead to system imbalance. This mechanism enables the system to simulate the evolution trajectory of tourism resource states under different management strategies, automatically generating and iteratively optimizing resource allocation blueprints containing multi-stage action nodes. Unlike traditional static or linear programming, this method achieves dynamic capture and adaptive control of complex system behavior, making resource allocation schemes forward-looking and flexible, capable of proactively responding to potential future changes and disturbances.

[0016] Semantic alignment technology unifies and contextualizes heterogeneous data streams from mobile terminal interfaces and physical sensors, eliminating semantic barriers between data. Based on this, the system aggregates the aligned data to construct a multi-dimensional spatiotemporal state tensor, which simultaneously carries the characteristics and state changes of resources across multiple dimensions, including time, space, and attributes. This technology overcomes the shortcomings of conventional methods, such as flat data models and weak correlations, achieving a holographic and high-fidelity digital mapping of tourism resources. Based on this unified and rich tensor model, subsequent factor extraction and analysis obtain a high-quality, highly consistent data foundation, improving the depth of state perception and the accuracy of decision analysis. Attached Figure Description

[0017] Figure 1 This is a timeline diagram of the modern tourism resource survey system based on big data described in this invention. Figure 2 A flowchart for extracting the set of core constraint factors; Figure 3 A flowchart illustrating the operation of the Blueprint Instruction Execution Module; Figure 4 A bar chart comparing parameters in the sub-process stage of the instruction adjustment; Figure 5 A heat map showing the difference between predicted and actual visitor density in different areas of the seaside scenic area at different times. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] See Figure 1 The tourism resource sensing module captures and aggregates heterogeneous tourism resource sensing data streams in real time through application programming interfaces deployed on mobile terminals and physical sensors deployed at tourism resource nodes. The data aggregation and analysis module performs multi-scale aggregation and semantic alignment operations on the tourism resource sensing data streams to construct a multi-dimensional spatiotemporal state tensor describing the state of tourism resources. The constraint factor extraction module performs hierarchical decoupling and correlation analysis on the multi-dimensional spatiotemporal state tensor based on preset resource coupling rules, extracting a set of core constraint factors affecting the effectiveness of tourism resources. The dynamic deduction and planning module utilizes a dynamic adaptive adjustment framework including positive and negative feedback paths to quantitatively deduce and simulate interventions on the set of core constraint factors, forming a tourism resource optimization configuration blueprint containing multi-stage planning nodes. The blueprint instruction execution module converts the tourism resource optimization configuration blueprint into an executable sequence of digital instructions, driving automated facilities to perform corresponding resource configuration and guidance actions. All modules are sequentially connected to form an overall process from data acquisition to physical execution.

[0021] In one embodiment of the present invention, see [reference] Figure 2 Based on a seaside scenic area scenario, an application programming interface (API) deployed on tourists' mobile terminals continuously uploads location coordinates and dwell time. Physical sensors deployed in the scenic area's dining areas, rest areas, and viewing platforms periodically report temperature, humidity, and visitor count readings. The tourism resource sensing data stream includes the tourist location sequence reported by the mobile terminals and the node environmental readings reported by the physical sensors. The data aggregation and analysis module aligns and stitches the tourist location sequence from the mobile terminals and the node environmental readings from the physical sensors according to a unified Coordinated Universal Time (UTC) timestamp benchmark, generating a spatiotemporally synchronized raw data frame. A single alignment and stitching operation associates the coordinate data received at 10:05 AM from one hundred mobile terminals with the temperature readings from twenty physical sensors at the same time into the same raw data frame.

[0022] In some embodiments, a spatial domain gridded downsampling process is applied to the spatiotemporally synchronized raw data frame, dividing the entire scenic area map into a regular grid of 50 meters by 50 meters. Multiple visitor coordinate points within each grid are aggregated into a single value representing the visitor density of that grid, generating a spatial slice set with different granularities. Simultaneously, a temporal domain event slicing process is applied to the raw data frame, segmenting the continuous perceived data stream according to regular scenic area activity events. In a specific implementation, a geographic feature-based regional clustering operation is performed on the spatial slice set. The algorithm identifies the food court area, where visitor coordinate points are consistently clustered at high density, and the coastal walkway area, which exhibits a linear distribution. These two areas are identified as interest region clusters with similar visitor distribution patterns. A behavioral pattern-based periodic segmentation operation is performed on the time slice set, analyzing historical data to mark the "morning tour period" (9:00 AM to 11:00 AM) as a stable high-activity period and the "midday rest period" (12:00 PM to 2:00 PM) as a stable high-activity period. These periods are marked as standard time periods with stable tourism activity characteristics.

[0023] In practice, the boundary information of the region of interest cluster is combined with the start and end information of the standard time period using a Cartesian product to generate a data cube for each combination of region and time period. Features are extracted from all environmental readings and location sequences corresponding to the same data cube. For example, the average number of tourists per minute is extracted from the location sequences, and the average temperature and average humidity are extracted from the environmental readings. These features are then normalized. The normalization process can be calculated using the following formula:

[0024] in: Represents the normalized eigenvalues. Represents the original feature values. This indicates the minimum value of the feature in historical data. This represents the maximum value of the feature in historical data. The normalized feature values ​​are then filled into different dimensions of the data cube. Finally, all data cubes from different regions and time periods are stacked to form a multidimensional spatiotemporal state tensor describing the state of tourism resources. A slice of this multidimensional spatiotemporal state tensor can represent the state of the "dining plaza area during the midday rest period," including visitor density, average temperature, and average humidity.

[0025] In the implementation of the constraint factor extraction module, hierarchical decoupling and correlation analysis are performed on the multidimensional spatiotemporal state tensor based on preset resource coupling rules. The density dimension reflecting passenger flow pressure, the capacity dimension reflecting facility load, and the quality dimension reflecting environmental perception are extracted from the multidimensional spatiotemporal state tensor. Based on the nonlinear mapping relationship between density and capacity defined in the resource coupling rules, the critical region where passenger flow density approaches the facility's carrying capacity limit within the region of interest is calculated. In the example scenario, the calculation shows that the real-time passenger flow density in the food court area reaches 95% of the seating capacity during the lunch break, thus marking the food court area as a primary constraint factor. Based on the attenuation transmission relationship between capacity and quality defined in the resource coupling rules, the transmission chain of decreased environmental quality perception due to facility overload within the critical region is calculated. The calculation shows that full seating in the food court area leads to longer waiting times for tourists, which in turn causes a decrease in tourist satisfaction scores in the area. The node of extended waiting time is marked as a key attenuation node, i.e., a secondary constraint factor. Based on the negative feedback relationship between quality and density defined in the resource coupling rules, the probability of tourists spontaneously leaving or avoiding specific areas due to a decline in perceived environmental quality is calculated. The calculation shows that due to excessively long waiting times, there is a 60% probability that subsequent tourist flows will shift to other areas. Therefore, the food court area is also marked as a high-probability area, i.e., a level-three constraint factor. The critical areas, key decay nodes, and high-probability areas marked at all levels are combined to form the core constraint factor set.

[0026] In one embodiment of the present invention, based on a seaside scenic area scenario and an acquired set of core constraint factors, the set of core constraint factors includes a critical area of ​​the food plaza marked as a primary constraint factor, key attenuation nodes of prolonged visitor waiting time marked as secondary constraint factors, and a high-probability area of ​​the food plaza marked as a tertiary constraint factor. The dynamic simulation planning module utilizes a dynamic adaptive adjustment framework that includes positive and negative feedback paths to quantitatively extrapolate and simulate interventions on the set of core constraint factors. Within the dynamic adaptive adjustment framework, a positive feedback path is set with the goal of improving overall resource utilization. The positive feedback path observes the chain reaction by simulating the increase in service capacity of key attenuation nodes. For example, simulating the addition of twenty temporary rest seating units and two mobile food cart service points in the food plaza area, and observing the chain reaction of this intervention on alleviating the passenger flow pressure in the critical area of ​​the food plaza. In the dynamic adaptive regulation framework, a negative feedback path with the goal of maintaining system stability is set. The negative feedback path observes the adverse regulation effect by simulating the implementation of tourist diversion in high-probability areas. For example, on the main path leading to the food plaza area, an electronic guide map is used to display a notice that the nearby viewing platform area is vacant. The adverse regulation effect of this intervention on reducing the local density of the food plaza and thus improving the perceived environmental quality score of the food plaza area is observed.

[0027] In some embodiments, the simulation process of positive feedback paths and negative feedback paths is run alternately, and the changes in the multidimensional spatiotemporal state tensor after each simulation intervention are recorded. These changes involve the passenger flow density dimension readings for the food court area, the environmental perception quality dimension readings, and the passenger flow density dimension readings for the adjacent viewing platform area within the multidimensional spatiotemporal state tensor. The deviation between the changes and the expected target is evaluated, and the deviation is calculated using the following formula:

[0028] in: Indicates the overall deviation. This represents the actual change in passenger flow density in the food court area after simulation. This indicates a reduction in the target amount of expected customer flow density in the food court area. This represents the change in the perceived environmental quality score of the food court area after the simulation. This represents the target increase in the perceived environmental quality score for the food court area. It's important to understand that all changes in the formula are predicted values ​​derived from simulations, not actual measurements.

[0029] When the overall deviation of multiple consecutive simulations falls below a preset convergence threshold, the system is deemed to have reached a quasi-steady state, and all triggered interventions and their execution parameters within this framework are extracted. Optionally, triggering measures may include "deploying a mobile food cart at coordinates (X1, Y1) with a service radius of fifty meters" and "raising the display priority of the electronic guide map path P1 from level three to level one within time slices T1 to T2." In specific implementations, the interventions and their execution parameters are arranged into a list of stage tasks with sequential dependencies according to the chronological order of simulation execution. In some embodiments, the simulation of the dynamic adaptive adjustment framework is iteratively performed until the overall deviation meets the convergence condition. It can be understood that each simulation is a predictive calculation of the system state. The final tourism resource optimization allocation blueprint clarifies the timing and conditions for initiating different interventions.

[0030] In one embodiment of the present invention, see [reference] Figure 3Based on the seaside scenic area scenario and the generated tourism resource optimization blueprint, the tourism resource optimization blueprint contains multi-stage planning nodes. For example, the first planning node is "15 minutes before the start of the lunch break, activate and deploy temporary rest seating units" and the second planning node is "30 minutes after the start of the lunch break, if the density of the catering plaza area is still higher than the threshold K, trigger the electronic guide map diversion instruction." The blueprint instruction execution module parses each stage planning node in the tourism resource optimization blueprint and breaks it down into operation instruction elements for specific geographical coordinates or facility numbers. Parsing the first planning node yields operation instruction elements including "send an activation instruction to the temporary rest seating controller located at geographical coordinates (X1, Y1)" and "send an activation instruction to the temporary rest seating controller located at geographical coordinates (X2, Y2)".

[0031] In some embodiments, each operation instruction is matched with a preset facility control protocol template. For example, the operation instruction "send activation instruction to temporary rest seat controller" is matched with a Modbus RTU protocol control template, which defines placeholders for device address, function code, and register address. Execution parameters defined by the phase planning node are filled in. The execution parameters defined by the first planning node include device address "01", function code "05", and register value "FF00" indicating "activation", generating a standard device control instruction. The complete device control instruction is the hexadecimal string "01050001FF00". All device control instructions are sorted according to the timing defined by the phase planning node. For example, the activation instructions of two temporary rest seat controllers are arranged in coordinate order, and system status verification instructions are inserted between adjacent instructions to form an executable digital instruction sequence.

[0032] In practice, executable digital command sequences are sent in batches to the corresponding core controllers of automated facilities. For example, batches of commands containing activation instructions are sent to the core controllers of intelligent flow control screens and automatically deployed temporary rest seats. The core controllers execute the received commands. For intelligent flow control screens, this means adjusting the displayed information, changing the default welcome message to "Added seats available in the lounge area ahead." For automated service facilities, the core controllers reallocate service resources; for temporary rest seats, the core controllers drive motors to deploy seat units; and for variable message signs, the core controllers change the physical guidance path, directing the arrows on the signs towards the newly added seating area.

[0033] In some embodiments, the instruction sequencing logic needs to consider the response latency and communication latency of physical facilities, and a time interval parameter is introduced into the sequencing operation. Optionally, the time interval parameter is set based on the average response time of historical instruction executions, and the formula for calculating the time interval between adjacent device control instructions in the instruction sequence is as follows:

[0034] in: Indicates the time interval between adjacent instructions. This indicates the historical average response time of the target facility's core controller. This represents the historical average transmission delay of the communication network. The formula ensures that the previous device control command has sufficient time to be received and executed before the next device control command is sent. The digital command sequence is ultimately sent to the various distributed core controllers through the scenic area's IoT communication gateway.

[0035] In one embodiment of the present invention, the process of sending a digital instruction sequence by the blueprint instruction execution module in a seaside scenic area scenario includes an activation device control instruction for the temporary rest seat controller and a subsequent system status verification instruction. After sending a device control instruction, the system status verification instruction is triggered. The system status verification instruction obtains the instantaneous status reading of the instruction execution target area by calling the real-time data interface of the physical sensors deployed in the food plaza area. The instantaneous status reading includes the pedestrian count sensor value and the temperature sensor value of the target area within 30 seconds after the instruction is issued. The instantaneous status reading is compared with the status threshold expected to be reached by the system status verification instruction. For example, the actual pedestrian count value is compared with the target of expected reduction to below threshold K, and the actual temperature value is compared with the target of expected maintenance within the comfort range. A status consistency verification result is generated, and the status consistency verification result records the difference between the actual value and the expected value. See Table 1.

[0036] Table 1: State Consistency Verification Results Command ID Verification Dimensions Expected threshold Actual reading Verification results CMD_001 Customer flow density in the food plaza area ≤0.85 0.89 Deviation from expectations CMD_001 Average temperature in the food court area 22-26°C 24.5°C As expected If the state consistency check result is as expected, the digital instruction sequence is allowed to continue issuing the next equipment control instruction. If the state consistency check result deviates from the expectation, such as the actual reading of passenger flow density dimension being 0.89 higher than the expected threshold of 0.85 in the example, the instruction adjustment sub-process is triggered. The instruction adjustment sub-process dynamically fine-tunes the execution parameters in the next equipment control instruction to be executed, or inserts a compensatory additional control instruction, based on the degree and direction of the deviation.

[0037] In some embodiments, the trigger instruction adjustment sub-process specifically includes analyzing the specific dimensions that deviate from expectations in the state consistency verification results, such as passenger flow density, facility load capacity, or environmental experience quality. Based on the deviated dimension, the most relevant successful amendment case is retrieved from the historical intervention case library, for example, retrieving a successful amendment case from the historical case library stating that "passenger flow density in area A is higher than expected; density is reduced by enhancing guidance." The parameter adjustment strategies used in the successful amendment cases are extracted, for example, extracting the parameter adjustment strategy of "increasing the priority of the electronic guide map path display from +1 to +2." The extracted parameter adjustment strategies are then fused with the currently pending equipment control instructions to calculate a new set of execution parameters with corrective intent. The following formula can be used to fuse and calculate the new execution parameters:

[0038] in: This represents the calculated new execution parameters. This indicates the predefined execution parameters in the original equipment control command. This indicates the parameter adjustment amount extracted from historical successful amendment examples. This represents an adjustment coefficient that is positively correlated with the current degree of deviation. The adjustment coefficient is obtained by linearly mapping the absolute difference between the current actual reading and the expected threshold.

[0039] In practice, new execution parameters are used to override the parameters in the original instruction, or a new compensation instruction containing the new parameters is generated and inserted into the corresponding position in the original instruction sequence. For example, the calculated new parameter "set the priority of electronic map path P1 to +2" is used to override the parameter "increase value to +1" in the original instruction. The adjusted instruction sequence fragment is resubmitted to the system status verification instruction for review. The system status verification instruction predicts the state after execution based on the new parameters, ensuring that the adjusted expected state can converge to the target range. In some embodiments, the matching retrieval of the historical intervention case library is based on multi-dimensional feature similarity. Optionally, the retrieval conditions include the deviation dimension type, region type, time period type, and deviation magnitude range. When a highly matching historical successful amendment case cannot be retrieved, the instruction adjustment sub-process adopts a rule-based default parameter adjustment strategy, such as increasing the intensity of the diversion instruction by a fixed level by default.

[0040] See Figure 4This is a bar chart comparing parameters during the instruction adjustment sub-process stage, clearly showing the parameter changes of five core instructions in the seaside scenic area's food plaza before and after the instruction adjustments. All types of instructions showed higher parameter values ​​after the adjustments, indicating that the system generally adopted enhanced intervention strategies after detecting deviations from expectations. The number of activated rest seats increased from 5 before the adjustment to 8 afterward, representing the largest increase, indicating significant visitor pressure at the time, and the system prioritized increasing rest facilities to improve the visitor experience and alleviate pressure. The priority of electronic guides increased from 1 to 2, indicating that the system dispersed visitor flow by strengthening the electronic guide's path guidance. The intensity of flow diversion guidance increased from 2 to 3, indicating that the on-site manual or signage flow diversion efforts were strengthened. This data perfectly corresponds to the description of the "instruction adjustment sub-process" in the project background: when the system detects that indicators such as visitor density deviate from expectations, it matches strategies from the historical case library and dynamically adjusts the execution parameters of various instructions to alleviate on-site pressure and restore the expected state.

[0041] In one embodiment of the present invention, the state is based on the completion of a complete resource allocation cycle in a seaside scenic area scenario. The resource allocation cycle refers to the process from the start of driving automated facilities according to a tourism resource optimization blueprint to the end of the blueprint's planned timeframe. After a complete resource allocation cycle, the closed-loop optimization module collects the actual effect data stream generated by the actions performed by the automated facilities. This actual effect data stream includes new pedestrian density readings and temperature readings in the food court area during the midday rest period, obtained through physical sensors after adjustments to the intelligent flow guidance screen and variable information signs, as well as tourist satisfaction ratings sampled through a mobile terminal application interface. The actual effect data stream is compared across all dimensions with the tourism resource optimization blueprint driving this cycle, calculating a difference graph between the blueprint's predicted and actual values. For example, if the blueprint predicts that the pedestrian density in the food court area will drop to 0.85 after the diversion command is implemented, and the actual effect data stream shows a density of 0.78, then the difference at that point in time is recorded as +0.07.

[0042] In some embodiments, source analysis is performed on the gap map to identify key links with large prediction deviations. Spatiotemporal points in the gap map where the absolute value of the deviation exceeds a set threshold are defined as having large prediction deviations. Key links may exist in the construction process of the multidimensional spatiotemporal state tensor, the extraction process of core constraint factors, or the deduction process of dynamic adaptive adjustment. Based on the results of the source analysis, the mapping relationship parameters in the resource coupling rules are corrected in reverse, or the weight coefficients of the positive and negative feedback paths in the dynamic adaptive adjustment framework are adjusted. The corrected rules and framework are used to process the newly collected tourism resource perception data stream, and the resource allocation optimization for the next cycle begins.

[0043] It is understandable that the source analysis of the gap map specifically includes decomposing the gap map according to the spatiotemporal dimension, locating the clusters of interest where the deviations occur spatially, and the standard time periods where they occur temporally. In the example scenario, the decomposition revealed that positive deviations (i.e., actual values ​​are better than predicted values) are concentrated in the food court area and during the lunch break. For the deviation concentration areas located spatiotemporally, the process of marking the core constraint factors corresponding to the deviation concentration areas was traced back when the tourism resource optimization allocation blueprint was formed, and the predicted density data on which the food court area was marked as a first-level constraint factor was checked. The completeness of the data in the multidimensional spatiotemporal state tensor on which the core constraint factors were marked was verified, as well as the accuracy of the mapping relationship applied to this area in the resource coupling rules. For example, it was verified whether the parameters of the nonlinear mapping function used to calculate the facility carrying capacity limit needed to be adjusted.

[0044] In practical implementation, the process is further traced back to the construction of the multidimensional spatiotemporal state tensor. This involves examining whether there is information loss or distortion in the aggregation, alignment, and feature extraction operations from the original data stream to the data cube. For example, when constructing the data cube for the lunch break period in the food court area, it's examined whether the gridded downsampling of the tourist location sequence excessively smoothed the true density peaks due to improper parameter settings. Through layer-by-layer backtracking, the final prediction deviation is attributed to at least one specific step in data capture, data aggregation, state representation, rule application, or inductive simulation, and this step is marked as a key area for optimization. For instance, the conclusion might be that the inductive process of dynamic adaptive adjustment overestimates the efficiency of tourist diversion.

[0045] Optionally, a quantitative traceability index can be introduced into the source tracing analysis process to assist in location identification. The formula for calculating the traceability index is as follows:

[0046] in: Indicates the traceability index, Indicates the blueprint's predicted value. Indicates the actual effect value. Indicates the spatial dimension weight. This indicates the weight of the time dimension. The components with high values ​​will be prioritized as critical components for optimization and correction. Through the continuous comparison, tracing, and correction actions described above, the closed-loop optimization module enables the system's predictive and control capabilities to evolve iteratively.

[0047] See Figure 5This is a heatmap showing the difference between predicted and actual visitor density in different areas of a seaside scenic area at various times. It visually reflects the accuracy of the prediction model in the tourism resource optimization blueprint and is a core analytical tool for the "closed-loop optimization module" in the project background. During the lunchtime period, the actual visitor density in the food court was about 0.06 lower than the predicted value, representing the point with the largest deviation on the entire map. This indicates that resources prepared for the lunchtime peak were underutilized. During the lunchtime period, the actual visitor density in the rest area was about 0.04 higher than the predicted value, indicating that the prediction model significantly underestimated the visitor flow in the rest area during this period, potentially leading to facility overload and a decline in visitor experience. The negative deviations in the afternoon at the viewing platform and in the morning at the parking lot also reflect significant discrepancies between prediction and reality. This heatmap can directly guide the work of the "closed-loop optimization module." For areas with negative deviations like the food court during lunchtime, future resource allocation efforts can be reduced to avoid waste.

[0048] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A modern tourism resource survey system based on big data, characterized in that: include: The tourism resource sensing module captures and aggregates heterogeneous tourism resource sensing data streams in real time through application programming interfaces deployed on mobile terminals and physical sensors deployed on tourism resource nodes. The data aggregation and analysis module performs multi-scale aggregation and semantic alignment operations on the tourism resource perception data stream to construct a multi-dimensional spatiotemporal state tensor describing the state of tourism resources. The constraint factor extraction module performs hierarchical decoupling and correlation analysis on the multidimensional spatiotemporal state tensor based on preset resource coupling rules, and extracts a set of core constraint factors affecting the effectiveness of tourism resources. The dynamic simulation planning module utilizes a dynamic adaptive adjustment framework that includes positive and negative feedback paths to quantitatively simulate and intervene in the set of core constraint factors, forming a blueprint for the optimal allocation of tourism resources that includes multi-stage planning nodes. The blueprint instruction execution module converts the tourism resource optimization configuration blueprint into an executable sequence of digital instructions, driving automated facilities to perform corresponding resource configuration and guidance actions.

2. The modern tourism resource survey system based on big data according to claim 1, characterized in that, The process of performing multi-scale aggregation and semantic alignment on the tourism resource perception data stream to construct a multi-dimensional spatiotemporal state tensor describing the state of tourism resources specifically includes: The sequence of tourist locations from mobile terminals and the node environmental readings from physical sensors are aligned and stitched together according to a unified timestamp reference to generate spatiotemporally synchronized raw data frames. The original data frame is subjected to gridded downsampling in the spatial domain and event slicing in the temporal domain, respectively, to generate spatial slice sets and temporal segment sets with different granularities. Perform geographic feature-based regional clustering on the spatial slice set to identify interest region clusters with similar tourist distribution patterns; Simultaneously, the time segment set is divided into periods based on behavioral patterns to mark standard time periods with stable tourism activity characteristics; The boundary information of the interest region cluster is combined with the start and end information of the standard time period by performing a Cartesian product to generate a data cube for each combination of region and time period. Feature extraction and normalization are performed on all environmental readings and location sequences corresponding to the same data cube, and these are then filled into different dimensions of the data cube, ultimately stacked to form the multidimensional spatiotemporal state tensor.

3. The modern tourism resource survey system based on big data according to claim 2, characterized in that, The multidimensional spatiotemporal state tensor is subjected to hierarchical decoupling and correlation analysis based on preset resource coupling rules, and a set of core constraint factors affecting the effectiveness of tourism resources is extracted, specifically including: The density dimension reflecting passenger flow pressure, the capacity dimension reflecting facility load, and the quality dimension reflecting environmental experience are extracted from the multidimensional spatiotemporal state tensor. Based on the nonlinear mapping relationship between density and capacity defined in the resource coupling rules, the critical areas where passenger flow density is close to the facility carrying capacity limit within the interest area cluster are calculated, and such areas are marked as first-level constraint factors. Based on the capacity and quality attenuation transmission relationship defined in the resource coupling rules, calculate the transmission chain of environmental quality perception decline caused by facility overload within the critical area, and mark the key attenuation nodes in the chain as secondary constraint factors. Based on the negative feedback relationship between quality and density defined in the resource coupling rule, the probability of tourists spontaneously leaving or avoiding a specific area due to a decline in perceived environmental quality is calculated, and high-probability areas are marked as third-level constraint factors. The critical regions, key decay nodes, and high-probability regions marked at all levels are combined to form the core constraint factor set.

4. The modern tourism resource survey system based on big data according to claim 3, characterized in that, The dynamic adaptive adjustment framework, which includes positive and negative feedback paths, is used to quantitatively extrapolate and simulate the set of core constraint factors, forming a blueprint for the optimal allocation of tourism resources that includes multi-stage planning nodes. Specifically, this includes: In the dynamic adaptive adjustment framework, a positive feedback path is set with the goal of improving the overall resource utilization rate. The positive feedback path is observed to increase the service capacity of the key attenuation nodes by simulation and to observe its chain effect on alleviating the pressure in the critical area. In the dynamic adaptive regulation framework, a negative feedback path with the goal of maintaining system stability is set. The negative feedback path is observed to reduce local density and improve the perception of environmental quality by simulating the implementation of tourist diversion in the high-probability area. The simulation process of alternating between the positive feedback path and the negative feedback path is carried out, and the change of the multidimensional spatiotemporal state tensor after each simulation intervention is recorded, and the deviation between the change and the expected target is evaluated. When the deviation of the simulation is lower than the preset convergence threshold in multiple consecutive simulations, the system is determined to have reached a quasi-steady state, and all triggered intervention measures and their execution parameters within the framework at this time are extracted. According to the time sequence of the simulated execution, the intervention measures and their execution parameters are arranged into a list of stage tasks with sequential dependencies, which are the multi-stage planning nodes in the blueprint for optimizing the allocation of tourism resources.

5. The modern tourism resource survey system based on big data according to claim 4, characterized in that, The process of converting the tourism resource optimization blueprint into an executable sequence of digital instructions to drive automated facilities to perform corresponding resource allocation and guidance actions specifically includes: Each stage planning node in the tourism resource optimization allocation blueprint is analyzed and broken down into operation instruction elements for specific geographical coordinates or facility numbers; Each operation instruction element is matched with a preset facility control protocol template and filled with execution parameters defined by the stage planning node to generate standard equipment control instructions; According to the timing sequence defined by the stage planning nodes, all the device control instructions are sorted, and system status verification instructions are inserted between adjacent instructions to form the executable digital instruction sequence. The digital instruction sequence is sent in batches to the core controller of the corresponding smart traffic guidance screen, variable information sign or automatic service facility; The core controller executes the received instructions, specifically by adjusting the information display content, changing the physical guidance path, or reallocating service resources.

6. The modern tourism resource survey system based on big data according to claim 5, characterized in that, The insertion of system status verification instructions between adjacent instructions specifically includes: After a device control command is sent, a system status verification command is triggered. The system status verification command obtains the instantaneous status reading of the target area of ​​the command execution by calling the real-time data interface of the physical sensor. The instantaneous status reading is compared with the expected status threshold reached by the system status verification command to generate a status consistency verification result. If the state consistency verification result is as expected, then the digital instruction sequence is allowed to continue issuing the next device control instruction; If the state consistency verification result deviates from the expectation, the instruction adjustment sub-process is triggered. The instruction adjustment sub-process dynamically fine-tunes the execution parameters in the next device control instruction to be executed, or inserts a compensatory additional control instruction, based on the degree and direction of the deviation.

7. The modern tourism resource survey system based on big data according to claim 6, characterized in that, The trigger command adjustment sub-process specifically includes: Analyze the specific dimensions in which the state consistency verification results deviate from expectations, whether it is the passenger flow density dimension, the facility load capacity dimension, or the environmental perception quality dimension; Based on the dimension of deviation, retrieve the most matching successful amendment case from the historical intervention case library, and extract the parameter adjustment strategy adopted in the successful amendment case; The extracted parameter adjustment strategy is fused with the current device control command to be executed to calculate a new set of execution parameters with corrective intent. The new execution parameters are used to overwrite the parameters in the original instruction, or a new compensation instruction containing the new parameters is generated and inserted into the corresponding position in the original instruction sequence. The adjusted instruction sequence fragment is resubmitted to the system state verification instruction for review to ensure that the adjusted expected state can converge to the target range.

8. The modern tourism resource survey system based on big data according to claim 1, characterized in that, The method also includes a closed-loop optimization module: After a complete resource allocation cycle is completed, collect the actual effect data stream generated by the actions performed by the automated facilities; The actual effect data stream is compared with the tourism resource optimization and allocation blueprint driving this cycle in all dimensions, and the gap map between the blueprint prediction value and the actual value is calculated. The source analysis of the gap map identifies the key links with large prediction deviations. These key links exist in the construction process of the multidimensional spatiotemporal state tensor, the extraction process of core constraint factors, or the deduction process of dynamic adaptive adjustment. Based on the results of the source tracing analysis, the mapping relationship parameters in the resource coupling rules are corrected in reverse, or the weight coefficients of the positive and negative feedback paths in the dynamic adaptation adjustment framework are adjusted. The newly acquired tourism resource perception data stream is processed using the revised rules and framework to begin the next cycle of resource allocation optimization.

9. The modern tourism resource survey system based on big data according to claim 8, characterized in that, The source analysis of the gap map to identify key links with large prediction deviations specifically includes: The gap map is decomposed according to the spatiotemporal dimension to locate the clusters of interest where the deviations are concentrated in space and the standard time periods where they are concentrated in time. For the deviation concentration area identified by spatiotemporal positioning, the process of marking the core constraint factors corresponding to the deviation concentration area is traced back when the tourism resource optimization allocation blueprint is formed; Verify whether the data in the multidimensional spatiotemporal state tensor on which the core constraint factor is based is complete, and whether the mapping relationship applied here in the resource coupling rule is accurate; Further tracing back to the process of constructing the multidimensional spatiotemporal state tensor, we examine whether there is information loss or distortion in the aggregation, alignment, and feature extraction operations from the original data stream to the data cube; By tracing back layer by layer, the final prediction deviation is attributed to at least one specific step in data capture, data aggregation, state representation, rule application, or inference simulation, and this step is marked as the key step to be optimized.

10. A modern tourism resource survey method based on big data, characterized in that, Applied to the modern tourism resource survey system based on big data as described in any one of claims 1 to 9.

Images