A security check resource allocation optimization method and system

Abstract

The application discloses a security check resource allocation optimization method and system, and relates to the field of resource allocation.The method comprises the following steps: predicting the initial passenger flow of a security check area in a future period; inverting a passenger flow abnormal fluctuation event of the security check area, correcting the initial passenger flow according to the passenger flow abnormal fluctuation event, and obtaining the corrected passenger flow of the security check area in the future period; inputting the corrected passenger flow into a security check resource allocation model to output an initial resource allocation scheme of the security check area; constructing a security check simulation scene, deploying the initial resource allocation scheme to the security check simulation scene, simulating the passing behavior of passengers in the security check area in the future period, and optimizing the initial resource allocation scheme based on the simulation result of the passing behavior.The application accurately predicts and corrects the security check passenger flow in the future period by combining the time series prediction of historical data and the inversion of the passenger flow abnormal fluctuation event, thereby providing accurate basis for the resource allocation of the security check area.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of resource allocation, in particular to a security check resource allocation optimization method and system. BACKGROUND

[0002] Security check resources refer to various resources in the airport security check area for ensuring efficient operation of the passenger security check process, including security check channels, security check staff, security check auxiliary facilities, etc. The allocation of security check resources is because the passenger flow of the security check area will dynamically change with time, date, flight plan and sudden abnormal events. By reasonably allocating resources, the resources can be accurately matched with the passenger flow demand, avoiding long waiting time of passengers, security check congestion due to insufficient resources, or waste of manpower and material resources due to idle resources, while ensuring the orderly and efficient security check process and improving the passenger travel experience.

[0003] If the security check resources cannot be allocated based on the passenger flow, on the one hand, the number of security check channels cannot be accurately predicted during the peak passenger flow period, resulting in insufficient number of security check channels and insufficient staff, which further prolongs the passenger waiting time. On the other hand, due to the difference in passenger flow distribution of different channels, resources cannot be allocated, which easily causes congestion in some channels and passenger queue congestion, while some channels are in an idle state due to the lack of passenger flow, which further affects the overall operation order of the security check area.

[0004] At present, there is no effective solution to the problems in the related art. SUMMARY

[0005] In view of the problems in the related art, the present application provides a security check resource allocation optimization method and system to overcome the above technical problems existing in the prior art.

[0006] To this end, the specific technical solutions adopted by the present application are as follows:

[0007] According to a first aspect of the present application, a security check resource allocation optimization method is provided, which comprises:

[0008] S1, constructing a time series prediction model based on pre-collected historical airport operation data, and predicting the initial passenger flow of the security check area in the future period through the time series prediction model;

[0009] S2, inverting the passenger flow abnormal fluctuation event of the security check area, correcting the initial passenger flow according to the resource control factor set according to the passenger flow abnormal fluctuation event, and obtaining the corrected passenger flow of the security check area in the future period;

[0010] S3, targeting the minimum waiting time of passengers in the security check area, and taking the security check area carrying capacity as a constraint condition, a security check resource allocation model is constructed; the corrected passenger flow is input into the security check resource allocation model to output an initial resource allocation scheme of the security check area;

[0011] S4, a security check simulation scene is constructed, the initial resource allocation scheme is deployed into the security check simulation scene, the passenger traffic behavior in the security check area in the future period is simulated, and the initial resource allocation scheme is optimized based on the simulation result of the traffic behavior.

[0012] Preferably, the time series prediction model is constructed based on the pre-acquired historical airport operation data, and the initial passenger flow of the security check area in the future period is predicted through the time series prediction model, and the method further comprises the following steps:

[0013] Acquire airport point cloud data, sequentially filter and denoise the airport point cloud data to obtain optimized point cloud data; and establish an airport visualization model according to the optimized point cloud data;

[0014] Identify geometric feature points associated with the security check area in the airport visualization model, and randomly select seed points for region growing based on the geometric feature points;

[0015] Taking the seed points as the starting point, gradually expanding and aggregating the geometric feature points according to the preset geometric similarity condition to divide the security check area in the airport visualization model.

[0016] Preferably, the passenger flow abnormal fluctuation event of the security check area is inverted, the initial passenger flow is corrected according to the resource control factor set according to the passenger flow abnormal fluctuation event, and the corrected passenger flow of the security check area in the future period is obtained, which comprises the following steps:

[0017] S21, screening the passenger flow fluctuation parameters associated with the security check area from the historical airport operation data, and inversely obtaining the passenger flow abnormal fluctuation event of the security check area based on the passenger flow fluctuation parameters;

[0018] S22, setting a control factor function according to the correction rule of the passenger flow abnormal fluctuation event, the control factor including a time offset factor, a passenger flow distribution factor and a passenger flow amplification factor;

[0019] S23, dividing the priority correction order based on the pre-defined rule engine mechanism, and correcting the initial passenger flow through the control factor function according to the priority correction order to obtain the corrected passenger flow of the security check area.

[0020] Preferably, the passenger flow fluctuation parameters associated with the security check area are screened from the historical airport operation data, and the passenger flow abnormal fluctuation event of the security check area is inversely obtained based on the passenger flow fluctuation parameters, which comprises the following steps:

[0021] S211, screen a plurality of passenger flow fluctuation parameter combinations associated with the security check area from historical airport operation data, and correspond each passenger flow fluctuation parameter combination to a Markov chain;

[0022] S212, derive a posterior probability of each passenger flow fluctuation parameter combination by using Bayes' theorem, intercept the Markov chain at a steady state based on the posterior probability, and integrate to generate a Markov chain sample set;

[0023] S213, substitute the Markov chain sample set into a pre-constructed forward model, and output a forward simulation result corresponding to each passenger flow fluctuation parameter combination through the forward model;

[0024] S214, analyze an error of the forward simulation result corresponding to each passenger flow fluctuation parameter combination, and select a passenger flow fluctuation parameter combination with the smallest error as a current optimal population;

[0025] S215, take the Markov chain sample set as an initial population of a grey wolf algorithm, take the current optimal population as an optimal position of prey, and select an optimal passenger flow fluctuation parameter combination from the passenger flow fluctuation parameter combinations through the grey wolf algorithm;

[0026] S216, analyze an abnormal passenger flow fluctuation parameter combination based on the optimal passenger flow fluctuation parameter combination, and trace a passenger flow abnormal fluctuation event corresponding to the abnormal passenger flow fluctuation parameter combination.

[0027] Preferably, taking the Markov chain sample set as an initial population of a grey wolf algorithm, taking the current optimal population as an optimal position of prey, and selecting an optimal passenger flow fluctuation parameter combination from the passenger flow fluctuation parameter combinations through the grey wolf algorithm include:

[0028] S2151, taking each passenger flow fluctuation parameter combination in the Markov chain sample as a grey wolf individual, and taking the current optimal population as an optimal position of prey;

[0029] S2152, calculating a distance between the grey wolf individual and the optimal position of prey according to a predefined grey wolf algorithm rule, and updating a position of the grey wolf individual based on the distance, taking a passenger flow fluctuation parameter combination corresponding to the grey wolf individual after the position is updated as a new candidate parameter combination;

[0030] S2153, calculating a posterior probability of the new candidate parameter combination, comparing the posterior probability with a preset threshold, if the posterior probability is greater than or equal to the preset threshold, updating a Markov chain state by combining a random walk update strategy, otherwise, keeping the original Markov chain state unchanged, and obtaining an updated Markov chain sample set;

[0031] S2154, input the updated Markov chain sample set to step S213 to calculate a new error, and iteratively perform steps S213-S215 until the error meets a preset value, and stop iteration, and take the passenger flow fluctuation parameter set with the final minimum error as the optimal passenger flow fluctuation parameter set.

[0032] Preferably, the priority correction order is divided based on a predefined rule engine mechanism, and the initial passenger flow is corrected by the regulation factor function according to the priority correction order to obtain the corrected passenger flow of the security check area, which includes:

[0033] S231, integrate the passenger flow abnormal fluctuation events to obtain a passenger flow abnormal fluctuation event list, match the passenger flow abnormal fluctuation event list with the rule library, and identify the correction rules triggered for execution;

[0034] S232, according to the priority values predefined in the rule library, sort the correction rules triggered for execution to obtain an ordered rule list arranged in descending order of priority;

[0035] S233, apply the regulation factor function corresponding to the correction rule with the highest priority in the ordered rule list to the initial passenger flow matrix integrated from the initial passenger flow to obtain an initial corrected passenger flow matrix;

[0036] S234, apply the regulation factor function corresponding to the correction rule with the next priority in the ordered rule list to the initial corrected passenger flow matrix until all the correction rules in the ordered rule list are executed to generate a corrected passenger flow matrix of the security check area.

[0037] Preferably, the initial resource allocation scheme is deployed into the security check simulation scene to simulate the passenger passing behavior in the security check area in the future period, and the initial resource allocation scheme is optimized based on the simulation result of the passing behavior, which includes:

[0038] S41, respectively establish passenger agents and staff agents in the airport visualization model, and construct a security check simulation scene for interaction between the passenger agents and the staff agents;

[0039] S42, deploy the initial resource allocation scheme into the security check simulation scene, simulate the interaction behavior of the passenger agents and the staff agents in the future period according to the initial resource allocation scheme, continuously collect evaluation indexes in the interaction behavior simulation, and integrate the evaluation indexes to generate a simulation running result data set;

[0040] S43, identify the congestion density of the security check area according to the simulation running result data set, use a causal reasoning algorithm to analyze the congestion density, and optimize the initial resource allocation scheme based on the analysis result.

[0041] Preferably, the congestion density of the security check area is identified according to the simulation running result dataset, the congestion density is analyzed by using a causal inference algorithm, and the initial resource allocation scheme is optimized based on the analysis result, including:

[0042] S431, obtaining the corrected passenger flow of each security check channel in the security check area from the simulation running result dataset, and calculating the three-dimensional space volume of each security check channel based on the corrected passenger flow;

[0043] S432, based on the three-dimensional space volume of the security check channel, the three-dimensional space of the security check channel is divided into equal cubes, and the volume of each equal cube is determined by Boolean operation, and the volume ratio of the voxel volume to the equal cube volume is taken as the congestion density of the security check area;

[0044] S433, the congestion density of the security check area is analyzed by using a causal inference model, and the analysis result is matched with the passenger flow abnormal fluctuation event list to obtain the congestion source analysis result of each security check channel;

[0045] S434, based on the congestion source analysis result, a dynamic prediction mechanism is introduced in the initial resource allocation scheme to optimize the initial resource allocation scheme.

[0046] Preferably, the congestion density of the security check area is analyzed by using a causal inference model, and the analysis result is matched with the passenger flow abnormal fluctuation event list to obtain the congestion source analysis result of each security check channel, including:

[0047] S4331, taking the congestion density of the security check area as the dependent variable, and taking the instantaneous passenger flow input of the passenger agent or the service rate of the staff agent as the independent variable, a causal inference model is constructed to represent the causal relationship between the dependent variable and the independent variable through a directed edge;

[0048] S4332, using the causal inference model to analyze the estimation required for the causal effect of the independent variable on the congestion density, analyzing the average treatment effect of each independent variable based on the estimation, and obtaining a driving factor list of the security check area sorted by causal effect intensity;

[0049] S4333, the driving factor list of the security check area is analyzed in association with the passenger flow abnormal fluctuation event list, and the current congestion behavior is traced back to the passenger flow abnormal fluctuation event associated with it according to the association analysis result.

[0050] According to another aspect of the present application, a security check resource allocation optimization system is also provided, which comprises:

[0051] The passenger flow prediction module is configured to construct a time series prediction model based on historical airport operation data collected in advance, and predict the initial passenger flow of the security check area in a future period of time through the time series prediction model.

[0052] The passenger flow correction module is configured to inverse the passenger flow abnormal fluctuation event of the security check area, correct the initial passenger flow according to the passenger flow abnormal fluctuation event, and obtain the corrected passenger flow of the security check area in the future period of time.

[0053] The resource allocation scheme formulation module is configured to construct a security check resource allocation model with the minimum waiting time of passengers in the security check area as a target and the carrying capacity of the security check area as a constraint condition, input the corrected passenger flow into the security check resource allocation model, and output the initial resource allocation scheme of the security check area.

[0054] The resource allocation scheme optimization module is configured to construct a security check simulation scene, deploy the initial resource allocation scheme into the security check simulation scene, simulate the passing behavior of passengers in the security check area in the future period of time, and optimize the initial resource allocation scheme based on the simulation result of the passing behavior.

[0055] The present application has the following advantages:

[0056] 1. The present application accurately predicts and corrects the security check passenger flow in the future period of time by combining time series prediction of historical data and inversion of passenger flow abnormal fluctuation events, thereby providing accurate basis for resource allocation of the security check area. By constructing an optimization model and combining the minimum waiting time target and the carrying capacity constraint, the optimal resource allocation can be achieved, the passenger waiting time can be reduced, the security check efficiency can be improved, and the actual passenger passing behavior can be simulated through the application of the security check simulation scene, and the resource allocation scheme can be dynamically adjusted according to the simulation result to ensure that the scheme can adapt to the changing passenger flow demand in actual operation and improve the overall operation efficiency of the security check area.

[0057] 2. The present application optimizes the passenger flow fluctuation parameter combination by using the grey wolf algorithm, corrects the initial passenger flow by combining the control factor function, can effectively deal with various passenger flow abnormal fluctuations, can ensure that the most critical factor is adjusted in time through the priority sorting mechanism under the action of multiple factors, finally obtain the corrected passenger flow that meets the actual situation, thereby improving the accuracy of security check resource allocation, reducing congestion, optimizing security check efficiency, and improving the passing experience of passengers and the flexibility of the security check process.

[0058] 3、The application can determine the influence of different factors on the crowd density by identifying the crowd density of the security check area and applying the causal inference algorithm for in-depth analysis, provide data support for resource optimization, and combine the correlation analysis of inference analysis results and passenger flow abnormal fluctuation events to accurately locate the congestion source, and introduce a dynamic prediction mechanism in the initial resource allocation scheme accordingly, this closed-loop optimization method not only can real-time adjust the security check resource configuration, effectively alleviate the congestion condition, but also can significantly improve the response speed and adaptability of the security check system. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0060] Figure 1 is a flow chart of a security check resource allocation optimization method according to an embodiment of the present application;

[0061] Figure 2 is a principle block diagram of a security check resource allocation optimization system according to an embodiment of the present application.

[0062] In the drawings:

[0063] 1, passenger flow prediction module; 2, passenger flow correction module; 3, resource allocation scheme formulation module; 4, resource allocation scheme optimization module. DETAILED DESCRIPTION

[0064] To further illustrate the embodiments, the present application provides drawings, which are part of the disclosure of the present application, mainly used to illustrate the embodiments, and can explain the operating principle of the embodiments in conjunction with the related description of the specification. With reference to these contents, those skilled in the art should understand other possible embodiments and advantages of the present application. The components in the drawings are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0065] According to an embodiment of the present application, a security check resource allocation optimization method and system are provided.

[0066] The present application will be further described in conjunction with the drawings and specific embodiments, as shown in Figure 1 The security check resource allocation optimization method according to an embodiment of the present application comprises:

[0067] S1, based on the historical airport operation data collected in advance, a time series prediction model is constructed, and the initial passenger flow of the security check area in the future period is predicted through the time series prediction model.

[0068] The time series prediction model is constructed based on historical airport operation data collected in advance, and the initial passenger flow of the security check area in the future period is predicted through the time series prediction model.

[0069] Collect airport point cloud data, filter and denoise the airport point cloud data in sequence to obtain optimized point cloud data; and establish an airport visualization model according to the optimized point cloud data;

[0070] Identify geometric feature points associated with the security check area in the airport visualization model, and randomly select seed points for region growing based on the geometric feature points;

[0071] Taking the seed point as the starting point, gradually expanding and aggregating the geometric feature points according to the preset geometric similarity condition to divide the security check area in the airport visualization model.

[0072] It should be noted that the time series prediction model is constructed based on historical airport operation data collected in advance, and the initial passenger flow of the security check area in the future period is predicted through the time series prediction model.

[0073] The historical airport operation data includes daily or hourly passenger volume, security check passage time, security check passage efficiency, flight arrival and departure time, boarding gate distribution, security personnel number and service rate, airport resource usage, weather conditions, and external influencing factors such as holidays or special events.

[0074] The historical airport operation data is input into the time series prediction model, and the time series prediction model includes ARIMA, SARIMA or LSTM. The trained time series prediction model is used to predict the security check passenger flow in the future period, thereby providing a basis for resource allocation and area optimization.

[0075] S2, reverse the passenger flow abnormal fluctuation event of the security check area, set the resource control factor according to the passenger flow abnormal fluctuation event to correct the initial passenger flow, and obtain the corrected passenger flow of the security check area in the future period.

[0076] The time series prediction model is constructed based on historical airport operation data collected in advance, and the initial passenger flow of the security check area in the future period is predicted through the time series prediction model.

[0077] S21, screen the passenger flow fluctuation parameters associated with the security check area from the historical airport operation data, and reverse the passenger flow abnormal fluctuation event of the security check area based on the passenger flow fluctuation parameters.

[0078] wherein the passenger flow fluctuation parameters associated with the security check area are screened from historical airport operation data, and the passenger flow abnormal fluctuation events of the security check area are inversely derived based on the passenger flow fluctuation parameters, including:

[0079] S211, a plurality of passenger flow fluctuation parameter combinations associated with the security check area are screened from historical airport operation data, and each passenger flow fluctuation parameter combination is corresponded to a Markov chain;

[0080] S212, the posterior probability of each passenger flow fluctuation parameter combination is derived using Bayes' theorem, the Markov chain at the stable state is intercepted based on the posterior probability, and a Markov chain sample set is integrated and generated;

[0081] S213, the Markov chain sample set is substituted into the pre-constructed forward model, and the forward simulation result corresponding to each passenger flow fluctuation parameter combination is output by the forward model;

[0082] S214, the error of the forward simulation result corresponding to each passenger flow fluctuation parameter combination is analyzed, and the passenger flow fluctuation parameter combination with the smallest error is selected as the current optimal population;

[0083] S215, the Markov chain sample set is used as the initial population of the grey wolf algorithm, the current optimal population is used as the optimal position of the prey, and the optimal passenger flow fluctuation parameter combination is selected from the passenger flow fluctuation parameter combinations by the grey wolf algorithm.

[0084] It should be noted that by corresponding the passenger flow fluctuation parameter combinations associated with the security check area to the Markov chain, and combining the Bayes' theorem to intercept the stable chain and generate the sample set, the stability and reliability of the parameter sample can be ensured. Substituting the forward model to output the simulation result and selecting the combination with the smallest error as the current optimal population can preliminarily lock the high-quality parameter direction. Iterative optimization is performed with the Markov chain sample set as the initial population of the grey wolf algorithm and the current optimal population as the optimal position of the prey, which can accurately locate the optimal parameter combination with the help of the global search capability of the algorithm. Through multiple iterations until the error meets the standard, the optimal fluctuation parameter combination suitable for passenger flow analysis of the security check area can be efficiently obtained, and the accuracy and efficiency of parameter screening are improved.

[0085] wherein the Markov chain sample set is used as the initial population of the grey wolf algorithm, the current optimal population is used as the optimal position of the prey, and the optimal passenger flow fluctuation parameter combination is selected from the passenger flow fluctuation parameter combinations by the grey wolf algorithm, including:

[0086] S2151, each passenger flow fluctuation parameter combination in the Markov chain sample is used as a grey wolf individual, and the current optimal population is used as the optimal position of the prey;

[0087] S2152, calculate the distance between the gray wolf individual and the optimal position of the prey according to the predefined gray wolf algorithm rule, and update the position of the gray wolf individual based on the distance, and the passenger flow fluctuation parameter combination corresponding to the position of the gray wolf individual after the position is updated as a new candidate parameter combination;

[0088] S2153, calculate the posterior probability of the new candidate parameter combination, compare the posterior probability with a preset threshold, if the posterior probability is greater than or equal to the preset threshold, update the Markov chain state combined with the random walk update strategy, otherwise, keep the original Markov chain state unchanged, and obtain an updated Markov chain sample set;

[0089] S2154, input the updated Markov chain sample set into step S213 to calculate a new error, and iteratively execute steps S213-S215 until the error meets a preset value, and stop iteration, and the passenger flow fluctuation parameter with the minimum final error is taken as the optimal passenger flow fluctuation parameter combination.

[0090] It should be noted that by combining the gray wolf algorithm and the Markov chain model, the optimal solution can be found in multiple possible passenger flow fluctuation parameter combinations, so as to accurately predict and control the passenger flow fluctuation in the security check area. As an optimization algorithm simulating natural hunting behavior, the gray wolf algorithm effectively explores and optimizes the combination of passenger flow fluctuation parameters, while the Markov chain can model the state transition process of passenger flow. Through continuous updating and optimization, more accurate prediction results can be provided in uncertain environment. Through repeated iteration, the passenger flow fluctuation parameter combination with the minimum error can be finally found, so that the passenger flow fluctuation in the security check area is accurately controlled and predicted.

[0091] The combination of gray wolf algorithm and Markov chain can thus back-propagate the abnormal fluctuation event in the security check area through the predicted passenger flow fluctuation parameter combination, and timely adjust the resource allocation in the security check area, so as to improve the security efficiency, and further reduce congestion and optimize resource allocation.

[0092] S216, analyze abnormal passenger flow fluctuation parameter combinations based on the optimal passenger flow fluctuation parameter combination, and trace back the passenger flow abnormal fluctuation event corresponding to the abnormal passenger flow fluctuation parameter combination.

[0093] It should be noted that after obtaining the optimal passenger flow fluctuation parameter combination, it can be used as a reference to compare with other passenger flow fluctuation parameter combinations to analyze abnormal passenger flow fluctuation parameter combinations. Specifically, the distance or difference between other parameter combinations and the optimal parameter combination can be calculated, such as using Euclidean distance, Mahalanobis distance and other measurement methods. When the distance or difference exceeds a threshold, the parameter combination is considered to be an abnormal passenger flow fluctuation parameter combination.

[0094] For tracing the abnormal passenger flow fluctuation event corresponding to the abnormal passenger flow fluctuation parameter combination, the timestamp, related event record and other information in the historical data are combined. When the abnormal passenger flow fluctuation parameter combination is determined, the time point corresponding to the parameter combination is searched, and the airport operation log, weather condition, special activity arrangement and other records near the time point are viewed to determine the specific event causing the abnormal passenger flow fluctuation, such as a large-scale activity, extreme weather influence, large-area flight delay and the like.

[0095] S22, a control factor function is set according to the correction rule of the abnormal passenger flow fluctuation event, and the control factor includes a time offset factor, a passenger flow distribution factor and a passenger flow amplification factor.

[0096] It should be noted that the initial passenger flow is a result of prediction based on historical regular data, and the influence of real-time / sudden abnormal events on passenger flow is not considered (for example, a holiday will expand the passenger flow scale, a flight delay will change the passenger flow arrival time, and a channel closure will cause passenger flow redistribution); and the control factor function can counteract the influence of abnormal events. The time offset factor adjusts the time dimension of passenger flow, the passenger flow distribution factor adjusts the distribution dimension of passenger flow in space, and the passenger flow amplification factor adjusts the scale dimension of passenger flow. The three can combine to correct the regular prediction value to the actual value that fits the abnormal scene.

[0097] S23, a priority correction order is divided based on a pre-defined rule engine mechanism, and the initial passenger flow is corrected by the control factor function according to the priority correction order to obtain the corrected passenger flow of the security check area.

[0098] The priority correction order is divided based on a pre-defined rule engine mechanism, and the initial passenger flow is corrected by the control factor function according to the priority correction order to obtain the corrected passenger flow of the security check area.

[0099] S231, a passenger flow abnormal fluctuation event list is obtained by integrating passenger flow abnormal fluctuation events, the passenger flow abnormal fluctuation event list is matched with a rule library, and a correction rule triggered to be executed is identified;

[0100] S232, the correction rule triggered to be executed is sorted according to the priority value pre-defined in the rule library to obtain an ordered rule list arranged in descending order of priority;

[0101] S233, the control factor function corresponding to the correction rule with the highest priority in the ordered rule list is applied to the initial passenger flow matrix integrated from the initial passenger flow to obtain an initial corrected passenger flow matrix;

[0102] S234. Apply the control factor function corresponding to the second priority correction rule in the ordered rule list to the initially corrected passenger flow matrix until all correction rules in the ordered rule list have been executed, and generate the corrected passenger flow matrix of the security check area.

[0103] The following section, in conjunction with specific implementation methods, further explains how the priority correction order is divided based on the predefined rule engine mechanism, and how the initial passenger flow is corrected according to the priority correction order through the adjustment factor function to obtain the corrected passenger flow in the security check area.

[0104] Step 1: Abnormal Event Integration and Rule Matching: Collect all traceable abnormal passenger flow fluctuation events, such as the pre-Spring Festival travel rush, the end of large-scale exhibitions near the airport, and concentrated flight delays caused by thunderstorms; organize the events into a structured event list according to their occurrence time and impact range; simultaneously call the pre-built rule library, which needs to pre-enter related rules such as those triggered by holiday peaks leading to passenger flow amplification factor correction; those triggered by flight delays leading to passenger flow allocation factor correction; and those triggered by temporary traffic control during the morning peak hours leading to time offset factor correction. Filter out all correction rules that need to be executed through keyword matching or feature matching.

[0105] Step 2: Correct the priority sorting of rules: Preset a priority value for each rule in the rule base. Sort the identified triggering rules from high to low priority value to generate an ordered rule list. For example, the priority of large-scale flight delays causing passenger backlog is set to 10, the priority of weekend short-distance travel peak is set to 5, and the priority of temporary closure of individual security checkpoints is set to 3.

[0106] Step 3: Priority Matrix Correction: Organize the initial passenger flow into an initial passenger flow matrix by time period (e.g., every 15 minutes) × security checkpoint (e.g., matrix A: rows = time periods such as 08:00-08:15, 08:15-08:30, columns = security checkpoints 1-6, cell value = initial passenger flow of that time period and that checkpoint). Then, take the control factor function corresponding to the highest priority rule in the ordered rule list and substitute it into matrix A to calculate the initial correction matrix. Use the control factor function of the second priority rule to apply to the initial correction matrix, and repeat this process until all rules are executed. The final matrix is ​​the corrected passenger flow matrix.

[0107] It should be noted that the expression for the time offset factor function (for abnormal passenger flow time distribution, such as flight delays and temporary air traffic control) is as follows:

[0108] ;

[0109] In the formula, for t Time period c Adjusted passenger flow at security checkpoint number 1; fort - k Period c Initial passenger flow of the c-th channel in the t-th period k Offset period number k =0 corresponds to the original period k =1 corresponds to the previous 1 period m Total offset period number (determined by the duration of the abnormal event) ω ( e , k ) Offset weight of the abnormal event e under the t-th period, satisfying k , Rule base preset example: flight delay e corresponds to ω ( e ,0)=0.05、 ω ( e ,1)=0.15、 ω ( e ,2)=0.4、 ω ( e ,3)=0.3、 ω ( e ,4)=0.1。

[0110] Passenger flow allocation factor function (for passenger flow spatial distribution abnormalities such as security channel closure and regional flow limitation) is expressed as:

[0111] ;

[0112] In the formula, is the corrected passenger flow of the c-th security channel in the t-th period t is the initial passenger flow of the c-th channel in the t-th period c is the set of channels affected by the anomaly and requiring diversion (such as closed channels) ( t , c , D ) is the diversion coefficient of the c-th channel to the c'-th channel under the abnormal event γ c d e e d c Rule base preset example: when the channel c is closed, the diversion coefficients to adjacent channels c1 and c2 are 0.6 and 0.4 respectively d

[0113] ​​​​​​​​S3, targeting the minimum waiting time of passengers in the security area, and taking the carrying capacity of the security area as a constraint condition, a security resource allocation model is constructed; the corrected passenger flow is input into the security resource allocation model to output an initial resource allocation scheme of the security area, which specifically includes:

[0114] The passenger flow data corrected by the regulatory factor function is arranged into a passenger flow matrix divided by time period and security channel, an optimization model with the objective function of minimizing the average waiting time of passengers in the security area is constructed, the carrying capacity in the security area, the processing capacity of the security channel, the number of security personnel and the like are set as constraint conditions in the model, and an optimization method is selected to solve the model, and the optimal configuration mode of resources in different time periods is obtained through the solving process.

[0115] Among them, the type principle of the security resource allocation model belongs to a queuing optimization model with resource constraints, and the local optimal or global optimal resource allocation solution is sought under complex time-varying passenger flow through mathematical programming or meta-heuristic algorithms, and the principle is based on abstracting the security process into a multi-service node queuing system, minimizing the total waiting time under the premise of ensuring system stability, which can be used for queuing theory models or linear analysis models in the prior art and the like.

[0116] The implementation content of the initial resource allocation scheme includes assigning the number of security channels to be opened for each specific time period, the number of security personnel to be allocated for each channel, the passenger flow proportion carried by each channel, and whether to enable the fast channel or the guiding mechanism and the like specific scheduling strategies as the basic configuration scheme for subsequent dynamic regulation and real-time adjustment of resources.

[0117] S4, a security simulation scene is constructed, the initial resource allocation scheme is deployed into the security simulation scene, the passenger passing behavior in the future period is simulated, and the initial resource allocation scheme is optimized based on the simulation result of the passing behavior.

[0118] It should be noted that in this step, the influencing factors of the congestion density of the security area are determined, the congestion density is taken as the dependent variable, the instantaneous passenger flow input and the service rate of the staff are taken as the independent variables, a causal reasoning model is established through directed edges, and the causal relationship between each variable and the congestion density is clearly presented; the estimation of the causal effect of the independent variables on the congestion density is calculated in the model, the average treatment effect of each independent variable (for example, the increase of 10 people in the instantaneous passenger flow input in a period of time, the decrease of 5 people / hour in the service rate) is analyzed, and a driving factor list is obtained by sorting the effect intensity; the driving factor list is associated with the existing passenger flow abnormal fluctuation event list, for example, the sudden increase of instantaneous passenger flow in the driving factor corresponds to the flight delay caused by the concentrated arrival of passenger flow in the event list, and the service rate decreases correspond to the temporary absence of security personnel, so as to trace the current congestion behavior to the specific associated abnormal event.

[0119] wherein the security check simulation scene is constructed, the initial resource allocation scheme is deployed into the security check simulation scene, the passenger's passing behavior in the security check area in the future period is simulated, and the initial resource allocation scheme is optimized based on the simulation result of the passing behavior includes:

[0120] S41, a passenger agent and a staff agent are respectively established in an airport visualization model, and a security check simulation scene for interaction between the passenger agent and the staff agent is constructed;

[0121] S42, the initial resource allocation scheme is deployed into the security check simulation scene, the interaction behavior of the passenger agent and the staff agent in the future period is simulated according to the initial resource allocation scheme, and evaluation indexes are continuously collected in the interaction behavior simulation, and the evaluation indexes are integrated to generate a simulation running result data set;

[0122] S43, the crowded density of the security check area is identified according to the simulation running result data set, the crowded density is analyzed by using a causal reasoning algorithm, and the initial resource allocation scheme is optimized based on the reasoning analysis result.

[0123] wherein the crowded density of the security check area is identified according to the simulation running result data set, the crowded density is analyzed by using a causal reasoning algorithm, and the initial resource allocation scheme is optimized based on the reasoning analysis result includes:

[0124] S431, the corrected passenger flow of each security check channel in the security check area is obtained from the simulation running result data set, and the three-dimensional space volume of each security check channel is calculated based on the corrected passenger flow;

[0125] S432, based on the three-dimensional space volume of the security check channel, the three-dimensional space of the security check channel is divided into equal cubes, the body elements contained in each equal cube are determined by Boolean operation, and the volume ratio of the body element volume to the equal cube volume is taken as the crowded density of the security check area.

[0126] It should be noted that by obtaining the corrected passenger flow from the simulation data and calculating the three-dimensional space volume of each security check channel, the space utilization of each channel is more accurately quantified, and the fine-grained space density value is obtained through space segmentation and Boolean operation, so as to reveal the crowded degree in the security check area. This method makes the resource allocation and optimization of the security check area more scientific, can timely identify potential congestion points, and provides data support for further dynamic regulation and security check process optimization, which helps to improve security efficiency, reduce waiting time and improve passenger experience.

[0127] Among them, Boolean operation is an operation mode based on Boolean algebra, mainly dealing with two logical values: true (True) and false (False). Its basic operations include AND, OR, NOT, etc., through these operations to input data for logical judgment and combination, get the final result. The working principle of Boolean operation is to return a result representing true or false value according to the predetermined logic rule under the given input condition.

[0128] S433, the crowded density of the security check area is analyzed by using the causal reasoning model, and the analysis result is matched with the passenger flow abnormal fluctuation event list to obtain the crowded source reasoning analysis result of each security check channel.

[0129] Among them, the crowded density of the security check area is analyzed by using the causal reasoning model, and the analysis result is matched with the passenger flow abnormal fluctuation event list to obtain the crowded source reasoning analysis result of each security check channel.

[0130] S4331, taking the crowded density of the security check area as the dependent variable, and the instantaneous passenger flow input of the passenger agent or the service rate of the staff agent as the independent variable, a causal reasoning model is constructed through a directed edge to represent the causal relationship between the dependent variable and the independent variable;

[0131] S4332, the causal effect of the independent variable on the crowded density is analyzed by using the causal reasoning model, the average treatment effect of each independent variable is analyzed based on the estimator, and a driving factor list of the security check area is obtained according to the causal effect intensity;

[0132] S4333, the driving factor list of the security check area is analyzed in association with the passenger flow abnormal fluctuation event list, and the current crowded behavior is traced back to the passenger flow abnormal fluctuation event associated with it according to the association analysis result.

[0133] It should be noted that according to the association analysis result, the current crowded behavior is traced back to the associated abnormal event, which can accurately locate the root cause of the congestion and avoid the blindness of resource allocation; and based on the tracing result, a dynamic prediction mechanism is introduced to optimize the initial scheme, which can change the resource adjustment from after-the-event response to pre-event prevention, reducing the long waiting time caused by insufficient resources.

[0134] S434, a dynamic prediction mechanism is introduced in the initial resource allocation scheme based on the crowded source reasoning analysis result, so as to realize the optimization of the initial resource allocation scheme.

[0135] It needs to be explained that the dynamic pre-judgment mechanism is introduced in the initial resource allocation scheme, based on the reasoning analysis result of the congestion source, the corresponding law of different abnormal events and congestion driving factors is analyzed, for example, flight delay event often leads to instantaneous passenger flow input surge this high impact driving factor, channel equipment failure often accompanied by service rate decline, then according to these laws, the pre-judgment rule is preset, when the flight delay event record is detected, it is predicted that the instantaneous passenger flow input of the corresponding security channel will rise in the next 1-2 hours, when the temporary job transfer of the staff is found, the service rate will be reduced.

[0136] These pre-judgment rules are embedded in the execution process of the initial resource allocation scheme, the update of the passenger flow abnormal fluctuation event list is monitored in real time during the execution process, once the preset associated event is monitored, the corresponding pre-judgment rule is triggered immediately, the resource configuration is adjusted in advance, specifically including pre-judging instantaneous passenger flow rise to open 1-2 security channels in advance, pre-judging service rate decline to timely deploy standby personnel to work, so as to realize the optimization of the initial resource allocation scheme.

[0137] According to another embodiment of the present application, as Figure 2 shown, a security check resource allocation optimization system is also provided, the system comprises:

[0138] The passenger flow prediction module 1 is used for constructing a time series prediction model based on the pre-collected historical airport operation data, and predicting the initial passenger flow of the security check area in the future period through the time series prediction model;

[0139] The passenger flow correction module 2 is used for inverting the passenger flow abnormal fluctuation event of the security check area, correcting the initial passenger flow according to the resource regulation factor set according to the passenger flow abnormal fluctuation event, and obtaining the corrected passenger flow of the security check area in the future period;

[0140] The resource allocation scheme formulation module 3 is used for constructing a security check resource allocation model with the minimum waiting time of the passengers in the security check area as the target and the carrying capacity of the security check area as the constraint condition; the corrected passenger flow is input into the security check resource allocation model to output the initial resource allocation scheme of the security check area;

[0141] The resource allocation scheme optimization module 4 is used for constructing a security check simulation scene, deploying the initial resource allocation scheme into the security check simulation scene, simulating the passing behavior of the passengers in the security check area in the future period, and optimizing the initial resource allocation scheme based on the simulation result of the passing behavior.

[0142] The above only describes the preferred embodiments of the present application and does not limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A security resource allocation optimization method, characterized in that, The method comprises the following steps: S1, constructing a time series prediction model based on pre-acquired historical airport operation data, and predicting the initial passenger flow of the security check area in the future period through the time series prediction model; S2, inverting the passenger flow abnormal fluctuation event of the security check area, setting the resource control factor according to the passenger flow abnormal fluctuation event to correct the initial passenger flow, and obtaining the corrected passenger flow of the security check area in the future period; S3, constructing a security check resource allocation model with the minimum waiting time of passengers in the security check area as the target and the carrying capacity of the security check area as the constraint condition; inputting the corrected passenger flow into the security check resource allocation model to output the initial resource allocation scheme of the security check area; S4, constructing a security check simulation scene, deploying the initial resource allocation scheme to the security check simulation scene, simulating the passing behavior of passengers in the security check area in the future period, and optimizing the initial resource allocation scheme based on the simulation result of the passing behavior; Wherein, S4 comprises: S41, establishing passenger agents and staff agents in the airport visualization model, and constructing a security check simulation scene for interaction between the passenger agents and the staff agents; S42, deploying the initial resource allocation scheme to the security check simulation scene, simulating the interaction behavior of the passenger agents and the staff agents in the future period according to the initial resource allocation scheme, and continuously collecting evaluation indexes in the interaction behavior simulation, and integrating the evaluation indexes to generate a simulation running result data set; S43, identifying the congestion density of the security check area according to the simulation running result data set, using a causal reasoning algorithm to analyze the congestion density, and optimizing the initial resource allocation scheme based on the analysis result; Wherein, S43 comprises: S431, obtaining the corrected passenger flow of each security check channel in the security check area from the simulation running result data set, and calculating the three-dimensional space volume of each security check channel based on the corrected passenger flow; S432, dividing the three-dimensional space of the security check channel into equal cubes based on the three-dimensional space volume of the security check channel, determining the voxels contained in each equal cube by Boolean operation, taking the volume ratio of the voxels to the equal cube as the congestion density of the security check area; S433, using a causal reasoning model to analyze the congestion density of the security check area, and matching the analysis result with the list of passenger flow abnormal fluctuation events to obtain the congestion source analysis result of each security check channel; S434, introducing a dynamic prediction mechanism in the initial resource allocation scheme based on the congestion source analysis result to optimize the initial resource allocation scheme. The S433 comprises constructing a causal inference model by taking the congestion density of the security check area as the dependent variable and taking the instantaneous passenger flow input of the passenger agent or the service rate of the staff agent as the independent variable; analyzing the required estimation of the causal effect of the independent variable on the congestion density by using the causal inference model; obtaining a driving factor list of the security check area sorted by causal effect intensity based on the average treatment effect of each independent variable analyzed based on the estimation; and performing correlation analysis on the driving factor list of the security check area and the passenger flow abnormal fluctuation event list, and tracing the current congestion behavior to the passenger flow abnormal fluctuation event associated therewith according to the correlation analysis result. 2.The security check resource allocation optimization method of claim 1, wherein, Before the step of predicting the initial passenger flow of the security check area in the future period by using the time series prediction model, the method further comprises: collecting airport point cloud data, sequentially filtering and denoising the airport point cloud data to obtain optimized point cloud data, and establishing an airport visualization model based on the optimized point cloud data; identifying geometric feature points associated with the security check area in the airport visualization model, and randomly selecting seed points for region growing based on the geometric feature points; taking the seed points as the starting point, gradually expanding and aggregating the geometric feature points according to a preset geometric similarity condition, and dividing the security check area in the airport visualization model.

3. The method of claim 1, wherein, The passenger flow abnormal fluctuation event of the security check area is inverted, and the initial passenger flow is corrected according to the resource control factor set according to the passenger flow abnormal fluctuation event to obtain the corrected passenger flow of the security check area in the future period, which comprises: S21, screening passenger flow fluctuation parameters associated with the security check area from historical airport operation data, and inversely deriving passenger flow abnormal fluctuation events of the security check area based on the passenger flow fluctuation parameters; S22, setting a control factor function according to the correction rule of the passenger flow abnormal fluctuation event, wherein the control factor comprises a time offset factor, a passenger flow distribution factor and a passenger flow amplification factor; S23, dividing a priority correction order based on a pre-defined rule engine mechanism, and correcting the initial passenger flow by using the control factor function according to the priority correction order to obtain the corrected passenger flow of the security check area.

4. The security resource allocation optimization method of claim 3, wherein, The passenger flow abnormal fluctuation event of the security check area is inversely derived based on the passenger flow fluctuation parameters screened from the historical airport operation data, which comprises: S211, screening a plurality of passenger flow fluctuation parameter combinations associated with the security check area from the historical airport operation data, and corresponding each passenger flow fluctuation parameter combination to a Markov chain; S212, deriving the posterior probability of each passenger flow fluctuation parameter combination by using Bayes' theorem, and generating a Markov chain sample set by integrating the Markov chain in the stable state based on the posterior probability; S213, substituting the Markov chain sample set into a pre-constructed forward model, and outputting the forward simulation result corresponding to each passenger flow fluctuation parameter combination by using the forward model; S214, analyzing the error of the forward simulation result corresponding to each passenger flow fluctuation parameter combination, and selecting the passenger flow fluctuation parameter combination with the smallest error as the current optimal population. S215, taking the Markov chain sample set as the initial population of the grey wolf algorithm, taking the current optimal population as the optimal position of the prey, and selecting the optimal passenger flow fluctuation parameter combination from the passenger flow fluctuation parameter combination through the grey wolf algorithm; S216, analyzing the abnormal passenger flow fluctuation parameter combination based on the optimal passenger flow fluctuation parameter combination, and tracing the passenger flow abnormal fluctuation event corresponding to the abnormal passenger flow fluctuation parameter combination.

5. The method of claim 4, wherein, Taking the Markov chain sample set as the initial population of the grey wolf algorithm, taking the current optimal population as the optimal position of the prey, and selecting the optimal passenger flow fluctuation parameter combination from the passenger flow fluctuation parameter combination through the grey wolf algorithm includes: S2151, taking each passenger flow fluctuation parameter combination in the Markov chain sample as a grey wolf individual, and taking the current optimal population as the optimal position of the prey; S2152, calculating the distance between the grey wolf individual and the optimal position of the prey according to the predefined grey wolf algorithm rule, and updating the position of the grey wolf individual based on the distance, taking the passenger flow fluctuation parameter combination corresponding to the grey wolf individual after the position is updated as a new candidate parameter combination; S2153, calculating the posterior probability of the new candidate parameter combination, comparing the posterior probability with a preset threshold, if the posterior probability is greater than or equal to the preset threshold, updating the Markov chain state in combination with a random walk update strategy, otherwise, keeping the original Markov chain state unchanged, obtaining an updated Markov chain sample set; S2154, inputting the updated Markov chain sample set into step S213 to calculate a new error, and iteratively executing steps S213-S215 until the error meets a preset value, and stopping iteration, taking the passenger flow fluctuation parameter with the minimum final error as the optimal passenger flow fluctuation parameter combination.

6. The security resource allocation optimization method of claim 5, wherein, The priority correction order is divided based on the predefined rule engine mechanism, the initial passenger flow is corrected through the regulation factor function according to the priority correction order, and the corrected passenger flow of the security check area is obtained, including: S231, integrating the passenger flow abnormal fluctuation events to obtain a passenger flow abnormal fluctuation event list, matching the passenger flow abnormal fluctuation event list with the rule library, and identifying the correction rules triggered for execution; S232, according to the priority values predefined in the rule library, sorting the correction rules triggered for execution to obtain an ordered rule list arranged in descending order of priority; S233, applying the regulation factor function corresponding to the correction rule with the highest priority in the ordered rule list to the initial passenger flow matrix integrated from the initial passenger flow, to obtain an initial corrected passenger flow matrix; S234, applying the regulation factor function corresponding to the correction rule with the second priority in the ordered rule list to the initial corrected passenger flow matrix, until all the correction rules in the ordered rule list are executed, and generating the corrected passenger flow matrix of the security check area.

7. A security resource allocation optimization system for implementing the security resource allocation optimization method of any one of claims 1-6, characterized in that, It includes: A passenger flow prediction module is configured to construct a time series prediction model based on pre-collected historical airport operation data, and predict the initial passenger flow of the security check area in a future period through the time series prediction model; The passenger flow correction module is configured to inverse the passenger flow abnormal fluctuation event of the security check area, correct the initial passenger flow according to the passenger flow abnormal fluctuation event, and obtain the corrected passenger flow of the security check area in the future period; The resource allocation scheme formulation module is configured to construct a security check resource allocation model with the minimum waiting time of passengers in the security check area as a target and with the carrying capacity of the security check area as a constraint condition; input the corrected passenger flow into the security check resource allocation model to output an initial resource allocation scheme of the security check area; The resource allocation scheme optimization module is configured to construct a security check simulation scene, deploy the initial resource allocation scheme to the security check simulation scene, simulate the passing behavior of passengers in the security check area in the future period, and optimize the initial resource allocation scheme based on the simulation result of the passing behavior.

Images