CN120875468A - Airport public transportation preallocation processing system and method based on big data model - Google Patents
Airport public transportation preallocation processing system and method based on big data modelInfo
- Publication number
- CN120875468A CN120875468A CN202511375081.8A CN202511375081A CN120875468A CN 120875468 A CN120875468 A CN 120875468A CN 202511375081 A CN202511375081 A CN 202511375081A CN 120875468 A CN120875468 A CN 120875468A
- Authority
- CN
- China
- Prior art keywords
- data
- time
- model
- passenger
- result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention belongs to the technical field of airport public transportation distribution and discloses an airport public transportation pre-distribution processing system and method based on a big data model. The method comprises the steps of obtaining historical data, utilizing a machine learning and deep learning model to infer passenger demands and public transportation service conditions in the historical data based on a big data processing technology, generating a data deduction model, inputting real-time weather, time and passenger data into the data deduction model, adopting an intelligent optimization algorithm to dynamically adjust public transportation resource allocation in combination with real-time traffic conditions, monitoring public transportation resource allocation results in real time, continuously optimizing dynamic adjustment correction results in combination with a feedback mechanism, and pushing the latest correction results to a downstream third-party system. The invention utilizes real-time data analysis and intelligent optimization algorithm to realize a more efficient traffic scheduling scheme, and has important practical significance and wide application prospect.
Description
Technical Field
The invention belongs to the technical field of airport public transportation distribution, and particularly relates to an airport public transportation pre-distribution processing system and method based on a big data model.
Background
With the rapid development of the global air transportation industry, the passenger flow of airports is continuously increased, and the dispatching and allocation of public transportation systems face great challenges. The airport is used as a city comprehensive transportation hub, and reasonable distribution of public transportation systems has important significance for improving the travel experience of passengers, reducing traffic jams and improving the operation efficiency. However, existing airport public transportation distribution systems suffer from a number of deficiencies, mainly represented by the following:
The information lag is that the public transportation scheduling of most airports at present depends on a fixed schedule and manual allocation, and the real-time response of flight dynamics, passenger flow changes and emergencies is difficult, so that the resource allocation is unreasonable.
The existing allocation mode usually depends on historical experience or static rules, big data analysis and machine learning technology are not fully utilized, prediction accuracy is low, and a scheduling scheme is not flexible enough.
The passengers experience poor, and due to unbalanced traffic resource allocation, the passengers can face the problems of long waiting time, overload of vehicles, resource waste and the like at the airport, and the travel satisfaction is affected.
The resource utilization rate is low, public transportation means (such as taxis, buses, subways and the like) around the airport cannot be matched with actual demands dynamically, and the situation of excessive or shortage of resources exists, so that the overall operation efficiency is reduced.
In recent years, the development of big data technology and artificial intelligence has provided new possibilities for airport public transportation scheduling optimization. Through the deep analysis of flight information, passenger flow, traffic tool states and historical data, intelligent prediction and dynamic allocation of public traffic resources can be realized, the operation efficiency of an airport traffic system is improved, the waiting time of passengers is reduced, and the overall travel experience is improved.
Disclosure of Invention
In order to overcome the problems in the related art, the disclosed embodiments of the invention provide an airport public transportation pre-allocation processing system and method based on a big data model. The invention aims to solve the problems of information lag, lack of intelligent optimization, poor passenger experience, low resource utilization rate and the like in the prior art.
The technical scheme is that the airport public traffic pre-allocation processing method based on the big data model comprises the following steps:
s1, acquiring historical data of flight information, passenger flow, weather conditions, vehicle states, passenger separation information and passenger arrival intention multi-source data;
S2, based on a big data processing technology, a machine learning and deep learning model is utilized to infer the passenger demand and the public transportation service condition in historical data, and a data deduction model is generated;
S3, inputting real-time weather, time and passenger data into a data deduction model, and dynamically adjusting public transportation resource allocation by adopting an intelligent optimization algorithm and combining real-time traffic conditions;
and S4, monitoring the public transportation resource allocation result in real time, continuously optimizing the dynamic adjustment correction result by combining a feedback mechanism, and pushing the latest correction result to a downstream third-party system.
In step S1, historical data of flight information, passenger flow, weather conditions, vehicle states, passenger departure information and passenger arrival intention multisource data are acquired, wherein the historical data comprise data acquisition, data preprocessing, data clustering and data distribution;
The data acquisition comprises the steps of obtaining flight information, passenger flow, passenger departure information, passenger arrival intention, weather conditions of an urban area where an airport is located, traffic condition data of the urban area where the airport is located through platforms such as butt joint production operation, passenger service, comprehensive management of airport environments, and traffic management of the urban area where the airport is located;
The data preprocessing is to perform preliminary preprocessing on the data acquired by the data, and filter useless data and error data;
The data clustering is to group the data according to the occurrence time period and the belonging passenger association information;
the data distribution comprises the step of distributing the processed and clustered data to the downstream for processing.
In step S2, based on big data processing technology, a large amount of historical data is collected, cleaned and analyzed, and disturbance data of abnormal conditions is eliminated, so that a set of historical data set which can be suitable for machine learning and deep learning is finally obtained, and then the machine learning and deep learning model is utilized to infer the passenger demands and the public transportation use condition in the historical data, so as to generate a data deduction model;
Analyzing all historical data by utilizing machine learning and deep learning models to analyze data and analyze traffic speculation service conditions of airports in different times, different weather and different passenger numbers;
setting the time influencing factor as the current time ,The influencing factors of (a) vary with time, and the expression is:
(1)
In the formula, As a factor of the influence of time,Half the difference between the highest and lowest population on the same day,Representing average people flow;
from the historical data, the formula evolves:
(2)
In the formula, For the amount of use of the vehicle,Weights for influencing factors (e.g. a class 4F airport in China, the influencing weight is thatAt the level of 0.42 of the total weight of the product,The total number of the components is 0.12,At the level of 0.09 c,The total number of the components is 0.11,At the level of 0.43 of the total number of the components,0.29),As a matter of the type of weather,In order to be a degree of humidity,In order to be able to determine the temperature,As a value of the error it is,For the volume of the passenger flow,Indicating whether or not to save holidays;
the model deduction and correction comprises comparing the estimated use condition of public traffic in historical data with the actual condition at the time, substituting the actual weather and time data into formula (2), judging whether the result Y is similar to the deduction result, if so, not adjusting, otherwise, adjusting Repeating the steps, and when the deduction result is always similar to the actual result, considering that the deduction model deduction is successful and generating a data deduction model, wherein the model realizes the formula by writing codes and comprises the realization logic of the formula and a historical data set for deduction.
Further, generating the data deduction model includes:
The collected data is subjected to preliminary processing, and repeated data and null data in the data are preliminarily removed in a computer mode;
extracting features of the processed data;
constructing a data set, and constructing a mapping comparison set with the public transportation service condition according to the processed data;
Training a model, namely using a big data combination model of a long-short-term memory network LSTM plus a model transducer based on an attention mechanism as a data deduction model, and respectively processing short-term time dependence and long-term trend to obtain a prediction result;
Training and optimizing, inputting all constructed data sets into LSTM and transducer models, and outputting corresponding results.
Further, the collected data is subjected to preliminary processing, repeated data and null data in the data are preliminarily cleared in a computer mode, wherein the steps of compiling a python script for the data obtained from excel, introducing pandas libraries, reading all the data in the excel by a pd.read_excel method, traversing each data, deleting the current data, storing all the undeleted data, compiling a corresponding sql script for the data obtained from the database, judging that key fields columns are not null or null character strings, judging 'columns is not null or columns =', finally obtaining all the data, and then manually checking the preliminarily clear data, and processing and correcting the abnormal data in the data;
The feature extraction of the processed data comprises the steps of removing useless feature values, marking key features, extracting whether the current peak time is the coverage area of the peak time, the influence of corresponding weather on riding traffic and extracting the subsequent riding traffic of passengers.
Further, the LSTM is configured to process short-term time series patterns, respectively including:
Forgetting door Determining how much past information is forgotten, wherein the formula is as follows:
;
In the formula, For the activation function in the LSTM,The current input weight is the value between 0 and 1 according to the input content; the input data comprises current hour flight information and weather conditions; For the weight entered at the previous time instant, The data input for the previous time comprises the flight information and weather condition of the previous time,As the offset, manually intervening the numerical value when the result is calculated finally; for the current moment of time, Is the previous time;
Input door Determining how the currently input information is stored in the cell state, remembering the influence of severe weather on traffic, and adopting the following formula:
;
In the formula, The weight input for the current input gate is manually judged according to the input content, and is manually adjusted, wherein the value is between 0 and 1; the weight input for the previous input gate, Input gate offset;
cell status Is responsible for storing long-term memory information, remembers past peak period modes, and has the following formula:
;
In the formula, The result value of the cell state data at the last moment; is a mathematical hyperbolic tangent function;
Output door Determining how much information of the current cell state is output, and only outputting information affecting future passenger flow, wherein the formula is as follows:
;
In the formula, For the weight of the current output gate output,The weight output by the gate is output for the last moment,To output a gate offset;
final result Representing the predicted content of public transportation at the next moment, the formula is:
;
In the formula, And obtaining the numerical value for the cell state formula at the current moment.
Further, the transducer is used for capturing long-term trends and global patterns, and learning short-term passenger flow changes in combination with LSTM, and the formula is as follows:
;
In the formula, The calculation result of the calculation formula is calculated; Converting the numerical value into probability distribution as a normalization function; For the transposition of the matrix, Is a matrix, transposed and connected withMultiplying and calculating the similarity; Is that I.e. the length of its vector, here defaulting to 64-dimensional data processing; Is the characteristic vector of the current moment; is the actual passenger flow demand at the historical time point.
In the step S3, real-time weather, time and passenger data are input into a data deduction model, an intelligent optimization algorithm is adopted, real-time traffic conditions are combined, public traffic resource allocation is dynamically adjusted, and comprises real-time data analysis intelligent adjustment and resource allocation recommendation (such as 7 months, 9 days, sunny days, late peak hours, historical data are analyzed according to historical data, the passenger flow is large in the historical data for the late peak hours of sunny days of summer transportation peak hours, 230 taxis, net jockey cars and 12 airport buses are needed at the moment according to the deduction of the past data, the rapid evacuation of passengers at the moment can be met in a subway matching mode, the number of taxi net jockey cars and airport buses is insufficient at the moment, and a small number of taxi net jockey cars and airport buses can be output outwards in a difference calculating mode;
The real-time data analysis intelligent adjustment comprises the steps of inputting real-time weather, time and passenger number key data into a data deduction model, dynamically adjusting and analyzing a current scene, and predicting the most likely public transportation resource allocation situation in a future scene;
The resource allocation recommendation includes outputting the prediction results and pushing the prediction results to a downstream third party system.
In step S4, monitoring public transportation resource allocation results in real time, continuously optimizing dynamic adjustment correction results by combining a feedback mechanism, and pushing the latest correction results to a downstream third-party system to acquire monitoring data, correct allocation results and feed back the data;
The monitoring data acquisition comprises real-time monitoring of public transportation at an airport, distinguishing the condition that the real-time data output result is inconsistent with the presumed content in time in a manual mode, and adjusting the weight value of each part Error value;
The distribution result correction comprises that when deduction is carried out, special data is analyzed again, and the weight is readjustedError valueAnd correcting the data result;
the data feedback comprises the steps of outputting the latest correction result and pushing the latest correction result to a downstream third-party system.
Another object of the present invention is to provide an airport public transportation pre-allocation processing system based on big data model, which implements the airport public transportation pre-allocation processing method based on big data model, the system comprises:
The data acquisition module is used for acquiring historical data of flight information, passenger flow, weather conditions, vehicle states, passenger departure information and passenger arrival intention multi-source data;
The data analysis and speculation module is used for speculating the passenger demands and the public transportation service conditions in the historical data by utilizing a machine learning and deep learning model based on a big data processing technology, and generating a data deduction model;
the prediction scheduling module is used for inputting real-time weather, time and passenger data into the data deduction model, and dynamically adjusting public traffic resource allocation by adopting an intelligent optimization algorithm and combining real-time traffic conditions;
The intelligent monitoring and feedback module is used for monitoring the public transportation resource allocation result in real time, continuously optimizing the dynamic adjustment correction result by combining a feedback mechanism, and pushing the latest correction result to a downstream third-party system.
By combining all the technical schemes, the invention has the following beneficial effects:
Firstly, the invention utilizes real-time data analysis and intelligent optimization algorithm to realize more efficient traffic scheduling scheme, and has important practical significance and wide application prospect. The method also has the advantages of accurately predicting, based on multidimensional data analysis, and improving the accuracy of traffic demand prediction. And (3) dynamically scheduling, namely adjusting traffic resources in real time, reducing waiting time of passengers and improving traffic efficiency. And intelligent optimization, namely adopting an advanced algorithm LSTM and a transducer to improve the resource utilization rate and reduce no-load operation. The system architecture supports multiple data source access, and can adapt to the requirements of different airports.
Secondly, the invention can reduce the pressure of public transportation near the airport, improve the travel efficiency of passengers and reduce the detention rate of passengers at the airport, thereby reducing the potential safety hazard caused by excessive personnel at the airport. The invention can optimize the public transportation scheduling scheme and improve the utilization rate of public transportation, thereby reducing the waste degree of public transportation. The method solves the management of most airports on arriving passengers, and greatly solves the secondary problem caused by the passengers staying at the airports.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure;
FIG. 1 is a flow chart of an airport public transportation pre-allocation processing method based on a big data model provided by an embodiment of the invention;
FIG. 2 is a graph finally formed in a historical data analysis provided by an embodiment of the present invention;
FIG. 3 is a schematic diagram of an airport public transportation pre-allocation processing method based on a big data model provided by an embodiment of the invention;
Fig. 4 is a schematic diagram of an airport public transportation pre-allocation processing system based on a big data model according to an embodiment of the present invention.
Detailed Description
In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be embodied in many other forms than those described herein, and similar modifications may be made by those skilled in the art without departing from the spirit of the invention, so that the invention is not limited to the specific embodiments disclosed below.
The invention has the innovation points that the historical data is analyzed to form a deduction model capable of processing big data of airport public traffic, and a more efficient traffic scheduling scheme is generated by utilizing real-time data analysis and intelligent optimization algorithm.
Embodiment 1 of the present invention provides an airport public transportation pre-allocation processing method based on big data model, which has the following technical characteristics.
The data acquisition and real-time analysis are realized by interfacing with platforms such as an airport production operation system, a passenger service system, an airport environment integrated management system, and local city traffic management, so as to acquire data such as flight information, passenger flow, passenger departure information, passenger arrival intention, weather conditions of an urban area where an airport is located, traffic conditions of the urban area where the airport is located, and the like. The integrated multi-dimensional historical data and instant data of flight dynamics, passenger flow trends, vehicle states and the like are analyzed by using a big data combination model of LSTM (long short term memory network) and a transducer (attention mechanism-based model), and an analysis result of the instant data is output, so that the accurate prediction of airport public transportation demands is realized.
And the intelligent scheduling optimization is realized by utilizing a machine learning algorithm and an optimization model based on LSTM (long short time memory network) and a transducer (attention mechanism-based model), analyzing historical data, generating an algorithm model based on airport multidimensional data, dynamically adjusting traffic resource configuration according to real-time data, and improving the scientificity and flexibility of a scheduling scheme.
And the travel experience of passengers is improved, namely the generated algorithm model based on the multidimensional data of the airports is used for analyzing the instant data, outputting the forecast arrangement of public traffic of the airports, and scheduling and adjusting the quantity of various public traffic, so that the time for the passengers to wait for the public traffic at the airports is reduced, inconvenience caused by uneven resource allocation is avoided, and the convenience and satisfaction of travel are improved.
The resource utilization rate is improved, the prediction arrangement of public transportation at the airport is output based on big data analysis, and transportation means such as taxis, buses, subways and the like can be reasonably allocated, so that idle and overload phenomena are reduced, and the overall operation efficiency is improved.
By introducing the intelligent and data driving methods, the invention provides a more accurate, efficient and flexible allocation scheme for the public transportation system of the airport, and meets the double requirements of efficient operation of modern airports and comfortable travel of passengers.
Specifically, as shown in fig. 1, the airport public traffic pre-allocation processing method based on the big data model provided by the embodiment of the invention includes:
s1, acquiring historical data of flight information, passenger flow, weather conditions, vehicle states, passenger separation information and passenger arrival intention multi-source data;
the method comprises data acquisition, data preprocessing, data clustering and data distribution.
Exemplary data collection includes obtaining data such as flight information, passenger flow, passenger departure information, passenger arrival intent, weather conditions in urban areas where airports are located, traffic conditions in urban areas where airports are located, and the like through platforms such as docking production operation, passenger service, comprehensive management of airport environments, and traffic management in local cities.
The data preprocessing comprises the steps of carrying out preliminary preprocessing on data acquired by data, and filtering useless data and error data.
The data clustering comprises the step of grouping the data according to the occurrence time period, the association information of the belonged passengers and the like.
The data distribution comprises the step of distributing the processed and clustered data to a downstream data analysis module for processing.
Still another exemplary, to build an accurate airport public transportation pre-allocation processing system, historical public transportation data of an airport needs to be collected first, including but not limited to, the number of flights at each time node, the amount of passengers, weather conditions, airport traffic pressure conditions, airport public transportation (buses, subways, network buses, taxis), and the like. The data can be acquired from platforms such as an airport production operation platform, a passenger service platform, an airport integrated service management platform, airport city traffic management and the like, and other platforms. And uniformly summarizing the data in the data pool by butting the data of each system, and waiting for subsequent processing. As shown in fig. 4;
S2, based on a big data processing technology, a machine learning and deep learning model is utilized to infer the passenger demand and the public transportation service condition in historical data, and a data deduction model is generated;
the method comprises data deep processing, historical data analysis, model deduction and correction.
Data depth processing, by way of example, involves converting all historical data into machine-readable form, such as weather conditions including weather conditions such as weather, rain, etc., into binary data that can be read by a computer. For example, the data at the same time, such as the data of weather, passengers, flight volume and the like at the time of 1 month, 7 days and 17 minutes, are integrated, and are put together and converted into binary data which can be read by a computer.
The historical data analysis comprises the steps of analyzing all historical data by using models such as machine learning, deep learning and the like, and analyzing traffic estimation use conditions of airports at different times, different weather and different passenger numbers.
Setting the time influencing factors as,The influencing factors of (2) vary over time. According to the analysis of the historical data, the time and the passenger volume can show a certain proportion, the early morning people flow is less, the early peak is more, the daytime is relatively stable, the late peak can appear at night, then the night can be reduced, and the finally formed curve is shown in figure 2. The formula is as follows:
(1)
In the formula, As a factor of the influence of time,Half the difference between the highest and lowest population on the same day,Representing the average flow of people and,Is the current time;
the number of passengers and the traffic influence condition are also larger in different weather. Weather is mainly divided into three types, one is weather type Such as clear or overcast, one is humidityThe last one is the temperature. From the historical data, the formula evolves:
(2)
In the formula, For the amount of usage of a vehicle such as a taxi,Weights for influencing factors (e.g. a class 4F airport in China, the influencing weight is thatAt the level of 0.42 of the total weight of the product,The total number of the components is 0.12,At the level of 0.09 c,The total number of the components is 0.11,At the level of 0.43 of the total number of the components,0.29),As a matter of the type of weather,In order to be a degree of humidity,In order to be able to determine the temperature,As a value of the error it is,For the volume of the passenger flow,Indicating whether or not to save holidays;
Exemplary model deduction and correction include comparing the estimated use condition of public transportation in historical data with the actual condition at the time by analyzing the actual weather, time and other data to be substituted into the formula (2), judging whether the result Y is similar to the deduction result, if so, not adjusting, otherwise, adjusting And (3) a value, so that the deduction result is similar to the final actual result. Repeating the steps, when the deduction result is always similar to the actual result, the deduction model can be considered to be successful in deduction, and a data deduction model is generated, wherein the model realizes the formula by writing codes. The method comprises the realization logic of the formula and a historical data set for deduction.
As shown in fig. 3 and 4, generating the data deduction model (data processing and model construction) further includes:
(1) And carrying out preliminary processing on the acquired data, and firstly, carrying out preliminary clearing on repeated data and null data in the data in a computer mode. For the data obtained from excel, a python script can be written, a pandas library is introduced, all data in excel is read through a pd.read_excel method, all the data are traversed, the current data are deleted if the key field is 'None', 'NULL' or empty character strings, and finally all the undeleted data are stored, for the data obtained from the database, a corresponding sql script can be written, the key field column is judged not to be empty or empty character strings, namely 'columns is not NULL or columns |=', and finally all the data are obtained.
And then, the preliminary clear data is examined in a manual mode, and the abnormal data in the data are processed and corrected, so that the data quality is ensured.
And extracting the characteristics of the processed data. Useless characteristic values such as "flight number", "passenger name", etc. are first removed. The key features are marked and extracted, for example, whether the current time is peak time or not can be extracted according to the flight arrival time, so that the coverage range of the peak time can be extracted, for example, the influence of corresponding weather on riding traffic can be extracted according to outdoor weather, rainfall condition and the like, and the possibility of the following riding of the passenger can be extracted according to the intention of the passenger.
The data set is constructed, the processed data is constructed, and a mapping comparison set of the parameters and public transportation use cases, such as flight peak time, public transportation use cases, bad weather such as rainy weather, strong wind and the like, public transportation use cases, travelers comprise travel groups, public transportation use cases and the like are constructed. The parameters also need to be combined to influence parameters such as off-peak time and rainy day, public transportation use condition when the local city has great activities and weather is good, public transportation use condition when off-peak time and subway is not present, and the like. According to the construction, the time characteristics (the time of entering a port of a flight, the delay time of the flight, the stop time of a bus subway and the like), the flight information (the number of flights, the arrival/departure flight ratio, the delay condition and the like), the weather characteristics (temperature, precipitation, wind speed and the like) and the historical traffic flow (the passenger flow of the past N hours and the taxi/subway utilization rate) are marked as a main data characteristic set.
Training model for the invention, large data combination model of LSTM (long short time memory network) and transducer (model based on attention mechanism) is used as data model (data deduction model) of the system, short time dependence and long time trend are processed respectively, so as to obtain more accurate prediction result.
Wherein LSTM is primarily used to handle short-term time series patterns, such as how flight changes over the past 1 hour affect passenger traffic. In LSTM, each includes:
Forget door (Forget Gate), ) Deciding how much past information was forgotten, such as forgetting night time data, focusing on peak daytime demands. The formula is as follows:
;
In the formula, For the activation function in the LSTM,The input data comprises current hour flight information and weather conditions; For the weight entered at the previous time instant, The data input for the previous time comprises the flight information and weather condition of the previous time,As the offset, manually intervening the numerical value when the result is calculated finally; for the current moment of time, Is the previous time; The importance of the current weight is manually judged according to the input content, the importance is manually adjusted, the numerical value is between 0 and1, if no flight exists at night, the required weight is smaller, and the value is set to be 0.1; As an offset, the numerical value is manually intervened in the final calculation result.
The Input Gate (Input Gate,) Determining how the currently entered information should be stored in the cell state, such as remembering the impact of bad weather on traffic. The formula is as follows:
;
In the formula, The weight input for the current input gate is manually judged according to the input content, and is manually adjusted, wherein the value is between 0 and 1; The input gate offset and the rest parameters have the same meaning as forgetting gate.
Cell status (CELL STATE,) Is responsible for the storage of long-term memory information, such as remembering past peak patterns. The formula is as follows:
;
The parameters are the same as the input gate and the forget gate.
The Output Gate (Output Gate,) It is decided how much information of the current cell state is output, such as only information affecting future passenger flows. The formula is as follows:
;
final result The final result is represented by the predicted content of public transportation at the next moment
;
The long-short-term memory network model uses Python for encoding, and LSTM is realized by introducing a torch.nn library, and the encoding is as follows:
import torch.nn as nn
class LSTMBlock(nn.Module):
def __init__(self, input_size, hidden_size, num_layers):super(LSTMBlock, self).__init__() self.lstm = nn.LSTM(input_size, hidden_size, num_layers, batch_first=1)
def forward(self, x): output, _ = self.lstm(x) return output
The input parameter input_size is data of an input gate, the hidden_size is data of a forgetting gate, and num_layers is data of each time period.
Aiming at the Transformer, the method is mainly used for capturing long-term trend and global mode, and learning short-term passenger flow change by combining LSTM, so as to realize accurate prediction. The formula is as follows:
;
In the formula, The calculation result of the calculation formula is calculated; Converting the numerical value into probability distribution as a normalization function; For the transposition of the matrix, Is a matrix, transposed and then connected withThe degree of similarity is calculated by multiplying,Is thatThe latitude values of (1), i.e. the length of its vector, are defaulted to 64-dimensional data processing, Q (Query) represents the feature vector (e.g. number of flights of 12:00, weather conditions) at the current time, K (Key) represents a matrix (e.g. flights and passenger flows of 10:00, 11:00), and V (Value) represents the actual passenger flow demand at the historical time point.
The attention mechanism based model uses Python for coding, implemented by introducing a torch. Nn library, which codes as follows:
class TransformerEncoder(nn.Module):
def __init__(self, d_model, nhead, num_layers): super(TransformerEncoder, self).__init__() self.encoder_layer = nn.TransformerEncoderLayer(d_model=d_model, nhead=nhead) self.transformer_encoder = nn.TransformerEncoder(self.encoder_layer, num_layers=num_layers)
def forward(self, x): return self.transformer_encoder(x)
Training and optimizing, namely inputting all constructed data sets into LSTM and transducer models, and outputting corresponding results. And (3) manually checking the output result, wherein if the check result is unknown from the actual public transportation use condition, the input parameters can be adjusted, and if the output result is close to the actual result, the data is checked. Calculating a variance value of the data, wherein, Data representing predictions from LSTM and transducer,Representing actual real data, n representing the number of times. And when the variance value is smaller than a reasonable value, the training optimization result is considered to be successful.
;
S3, inputting real-time weather, time and passenger data into a data deduction model, and dynamically adjusting public transportation resource allocation by adopting an intelligent optimization algorithm and combining real-time traffic conditions;
Mature intelligent optimization algorithms such as Genetic Algorithm (GA) and ant colony Algorithm (ACO) can be adopted, and public transportation resource allocation can be dynamically adjusted in combination with real-time traffic conditions. Including real-time data analysis intelligent adjustment and resource allocation recommendation.
The intelligent real-time data analysis and adjustment comprises the steps of inputting real-time weather, time, passenger number and other key data into a data deduction model, dynamically adjusting and analyzing a current scene, and predicting the most likely public transportation resource allocation situation in the future scene.
The resource allocation recommendation includes outputting the prediction results and pushing the prediction results to a downstream third party system.
Still further exemplary, future data may be input into the trained data inference model, and the prediction results may be output. And issues the results to other third party systems.
When the actual time is reached, if the difference between the actual result and the predicted result deviation value is larger, the environmental parameter in the current time period is recorded, the reason for the current deviation value is analyzed, whether interference items exist or not is checked, and therefore parameter values in an LSTM (long short time memory network) model and a transducer (model based on an attention mechanism) model are adjusted, as shown in fig. 3 and 4.
And S4, monitoring the public transportation resource allocation result in real time, continuously optimizing the dynamic adjustment correction result by combining a feedback mechanism, and pushing the latest correction result to a downstream third-party system.
The method comprises the steps of monitoring data acquisition, distribution result correction and data feedback.
Exemplary monitoring data acquisition includes monitoring public transportation and other data at an airport in real time. By manual mode, timely distinguishing the inconsistent condition of real-time data output result and presumed content, adjusting the weight value of each partError value。
The allocation result correction includes, for example, that when deduction from step S2 to step S3 occurs, the influence of tornado is not considered, but very special conditions such as tornado are present in the actual result. Then the special data is re-analyzed and the weight is readjustedError valueAnd corrects the data result.
The data feedback illustratively includes outputting the latest correction results and pushing them to downstream third party systems.
In another exemplary embodiment, the prediction result of the data deduction model and the numerical value required by the actual public transportation are compared and checked through a proper monitoring mode (such as manual spot check recording, video analysis and the like), the situation that the data deviation value is larger and is inconsistent with the ideal data result is obtained, and the collection record is performed. When the data to be collected reaches a certain value and can form a data set, analyzing according to the steps to check whether the influence of uncorrelated environmental factors exists or not, so as to optimize the whole model.
Embodiment 2, as shown in fig. 4, the schematic diagram of the airport public transportation pre-allocation processing system based on the big data model provided by the embodiment of the invention;
specifically, the airport public traffic pre-allocation processing system based on the big data model provided by the embodiment of the invention comprises:
The data acquisition module is used for acquiring multi-source data such as flight information, passenger flow, weather conditions, vehicle states, passenger departure information, passenger arrival intention and the like. The system comprises sub-modules of data acquisition, data preprocessing, data clustering, data distribution and the like. The data acquisition sub-module acquires data such as flight information, passenger flow, passenger departure information, passenger arrival intention, weather conditions of an urban area where an airport is located, traffic conditions of the urban area where the airport is located and the like through platforms such as butt joint production operation, passenger service, airport environment comprehensive management, local city traffic management and the like. And the data preprocessing sub-module is used for carrying out preliminary preprocessing on the data acquired by the data and filtering useless data and error data. And the data clustering sub-module is used for grouping the data according to the occurrence time period, the association information of the belonged passengers and the like. And the data distribution sub-module distributes the processed and clustered data to the downstream data analysis module for processing.
The data analysis and speculation module is used for speculating the passenger demands and the public transportation service conditions in the historical data by utilizing models such as machine learning, deep learning and the like based on big data processing technology. The method comprises the sub-modules of data deep processing, historical data analysis, model deduction, correction and the like. The historical data analysis submodule analyzes all historical data by using models such as machine learning, deep learning and the like, and analyzes traffic estimation use conditions of airports at different times, different weather and different passenger numbers. The model deduction and correction sub-module is used for comparing the estimated use condition of public transportation in the historical data with the actual condition at the moment by analyzing, namely substituting the actual weather, time and other data into a formula, judging whether the result Y is similar to the deduction result or not, if so, not adjusting, if not, adjusting the weight value to enable the deduction result to be similar to the final actual result. Repeating the steps, and when the deduction result is always similar to the actual result, the deduction model can be considered to be successful in deduction, and a data deduction model is generated.
And the prediction scheduling module is used for dynamically adjusting the allocation of public traffic resources by adopting an intelligent optimization algorithm and combining with real-time traffic conditions. The system comprises sub-modules of real-time data analysis intelligent adjustment, resource allocation recommendation and the like. The real-time data analysis intelligent adjustment sub-module is used for analyzing the current scene by inputting real-time weather, time, the number of passengers and other key data into the deduction model, and predicting the most probably used public transportation condition in the future scene. And the resource allocation recommendation sub-module is used for outputting the prediction result and pushing the prediction result to a downstream third-party system.
And the intelligent monitoring and feedback module is used for monitoring the traffic distribution effect in real time and continuously optimizing the scheduling strategy by combining a feedback mechanism. The system comprises sub-modules for monitoring data acquisition, distribution result correction, data feedback and the like.
The monitoring data acquisition sub-module is used for monitoring public transportation and other data of the airport in real time. And distinguishing the situation that the real-time data output result is inconsistent with the presumed content in time by a manual mode, and adjusting the weight value and the error value of each part. The distribution result correction submodule does not consider the influence of tornadoes when deducting, but extremely special conditions such as tornadoes appear in actual results. Then the special data is re-analyzed, the weight and error value are readjusted, and the data result is corrected. And the data feedback sub-module outputs the latest speculated result and pushes the latest speculated result to a downstream third-party system.
To further illustrate the effects associated with the embodiments of the present invention, the following experiments were performed.
Application example 1, airport bus scheduling situation is predicted based on flight data and passenger traffic.
The parameters that need to be entered include:
Flight information, flight arrival time, flight number, delay, etc.
Passenger flow information, namely real-time passenger quantity in an airport and travel mode preference (such as driving, taking buses and the like).
The traffic demands can be influenced by special weather such as heavy rain, heavy snow and the like.
Traffic state, which is the current available quantity of taxis and buses around the airport and traffic jam condition.
The invention can acquire the number of flights and the number of passengers borne by the passengers through the passenger service and production operation platform in a short period. Based on outdoor weather conditions and the same weather as in the previous period, the passengers need more or less airport buses to provide service. And then, according to the related equipment of the Internet of things, acquiring how many passengers are detained in the current terminal building and the traffic jam situation around the airport, finally transmitting the data into an LSTM and a Transformer model, and outputting the data to predict how many buses are needed to meet the requirements of the passengers under the environment.
Application example 2, when predicting airport shift schedule change based on long-term data, corresponding bus adjustment condition.
The parameters to be entered include flight information, the next flight schedule, the adjustment of the quarterly flight in the history, etc. Passenger information, passenger information from past years. Weather conditions are weather conditions in the quarter of the year. Holidays and major activities-holidays and major activities included in this quarter.
The invention can infer through the past year passenger information, the schedule of the airlines, the historical number of flights, and predict the number of passengers per day in the next quarter. And judging whether other influences exist according to weather and whether holidays or important activities exist. The data is finally transferred to LSTM and transducer models, which output the next demands of airport buses, net buses, etc. required every day throughout the quarter.
While the invention has been described with respect to what is presently considered to be the most practical and preferred embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications, equivalents, and alternatives falling within the spirit and scope of the invention.
Claims (10)
1. An airport public transportation pre-allocation processing method based on a big data model is characterized by comprising the following steps:
s1, acquiring historical data of flight information, passenger flow, weather conditions, vehicle states, passenger separation information and passenger arrival intention multi-source data;
S2, based on a big data processing technology, a machine learning and deep learning model is utilized to infer the passenger demand and the public transportation service condition in historical data, and a data deduction model is generated;
S3, inputting real-time weather, time and passenger data into a data deduction model, and dynamically adjusting public transportation resource allocation by adopting an intelligent optimization algorithm and combining real-time traffic conditions;
and S4, monitoring the public transportation resource allocation result in real time, continuously optimizing the dynamic adjustment correction result by combining a feedback mechanism, and pushing the latest correction result to a downstream third-party system.
2. The airport public transportation preallocation processing method based on big data model of claim 1, wherein in step S1, the historical data of flight information, passenger flow, weather conditions, vehicle status, passenger separation information, passenger arrival intention multisource data is obtained, including data acquisition, data preprocessing, data clustering, data distribution;
The data acquisition is to acquire flight information, passenger flow, passenger departure information, passenger arrival intention, weather conditions of urban areas where airports are located and traffic condition data of the urban areas where airports are located through a butt joint production operation, passenger service, airport environment comprehensive management and local city traffic management platform;
The data preprocessing is to perform preliminary preprocessing on the data acquired by the data, and filter useless data and error data;
The data clustering is to group the data according to the occurrence time period and the belonging passenger association information;
The data distribution is to distribute the processed and clustered data to the downstream for processing.
3. The airport public traffic pre-allocation processing method based on the big data model of claim 1, wherein in step S2, based on big data processing technology, the historical data is collected, cleaned and analyzed, the interference data of abnormal conditions is eliminated, the historical data set suitable for machine learning and deep learning is obtained, and then the machine learning and deep learning model is utilized to infer the passenger demand and the public traffic use condition in the historical data, so as to generate a data deduction model;
Analyzing all historical data by utilizing machine learning and deep learning models to analyze data and analyze traffic speculation service conditions of airports in different times, different weather and different passenger numbers;
setting the time influencing factor as the current time ,The influencing factors of (a) vary with time, and the expression is:
(1)
In the formula, As a factor of the influence of time,Half the difference between the highest and lowest population on the same day,Representing average people flow;
from the historical data, the formula evolves:
(2)
In the formula, For the amount of use of the vehicle,For the weight of each influencing factor,As a matter of the type of weather,In order to be a degree of humidity,In order to be able to determine the temperature,As a value of the error it is,For the volume of the passenger flow,Indicating whether or not to save holidays;
The model deduction and correction comprises comparing the estimated use condition of public traffic in historical data with the actual condition at the time, substituting the actual weather and time data into formula (2), and judging the result If the result is similar to the deduction result, if so, the adjustment is not performed, if not, the adjustment is performedA value that causes the deduction result to be similar to the final actual result;
Repeating the steps, when the deduction result is always similar to the actual result, the deduction model is considered to be successful in deduction, and a data deduction model is generated, wherein the model realizes the formula by writing codes and comprises the realization logic of the formula and a historical data set for deduction.
4. The airport mass transit pre-allocation process method based on big data model of claim 3, wherein generating the data deduction model comprises:
The collected data is subjected to preliminary processing, and repeated data and null data in the data are preliminarily removed in a computer mode;
extracting features of the processed data;
constructing a data set, and constructing a mapping comparison set with the public transportation service condition according to the processed data;
Training a model, namely using a big data combination model of a long-short-term memory network LSTM plus a model transducer based on an attention mechanism as a data deduction model, and respectively processing short-term time dependence and long-term trend to obtain a prediction result;
Training and optimizing, inputting all constructed data sets into LSTM and transducer models, and outputting corresponding results.
5. The airport public transportation preallocation processing method based on big data model of claim 4, wherein the preliminary processing of the collected data, the preliminary clearing of the repeated data and null data in the data by means of computer, comprises writing python script for the data obtained from excel, introducing pandas library, reading all data in excel by pd.read_excel method, traversing each data, deleting the current data, and storing all undeleted data; aiming at the data obtained from the database, writing a corresponding sql script, judging that a key field column is not empty or empty character strings, and judging 'columns is not null or columns +|=', through a coding mode, obtaining all data;
The feature extraction of the processed data comprises the steps of removing useless feature values, marking key features, extracting whether the current peak time is the peak time, the coverage range of the peak time, the influence of corresponding weather on riding traffic, and extracting the subsequent riding traffic of passengers.
6. The airport mass transit pre-allocation method based on big data model of claim 4, wherein LSTM is configured to process short-term time series patterns, each comprising:
Forgetting door Determining how much past information is forgotten, wherein the formula is as follows:
;
In the formula, For the activation function in the LSTM,The current input weight is the value between 0 and 1 according to the input content; the input data comprises current hour flight information and weather conditions; For the weight entered at the previous time instant, The data input for the previous time comprises the flight information and weather condition of the previous time,As the offset, manually intervening the numerical value when the result is calculated finally; for the current moment of time, Is the previous time;
Input door Determining how the currently input information is stored in the cell state, remembering the influence of severe weather on traffic, and adopting the following formula:
;
In the formula, For the weight input by the current input gate,The weight input for the previous input gate,Input gate offset;
cell status Is responsible for storing long-term memory information, remembers past peak period modes, and has the following formula:
;
In the formula, The result value of the cell state data at the last moment; is a mathematical hyperbolic tangent function;
Output door Determining how much information of the current cell state is output, and only outputting information affecting future passenger flow, wherein the formula is as follows:
;
In the formula, For the weight of the current output gate output,The weight output by the gate is output for the last moment,To output a gate offset;
final result Representing the predicted content of public transportation at the next moment, the formula is:
;
In the formula, And obtaining the numerical value for the cell state formula at the current moment.
7. The airport public transportation pre-allocation method based on big data model of claim 4, wherein the transform is used for capturing long-term trend and global pattern, and learning short-term passenger flow change in combination with LSTM, the formula is:
;
In the formula, Is the calculation result; Converting the numerical value into probability distribution as a normalization function; For the transposition of the matrix, In the form of a matrix of the matrix,Is thatIs used for the latitude value of (c),Is the characteristic vector of the current moment; is the actual passenger flow demand at the historical time point.
8. The airport public transportation pre-allocation processing method based on the big data model according to claim 1, wherein in step S3, real-time weather, time and passenger data are input into the data deduction model, an intelligent optimization algorithm is adopted, and by combining with real-time traffic conditions, dynamic adjustment of public transportation resource allocation comprises real-time data analysis intelligent adjustment and resource allocation recommendation;
The real-time data analysis intelligent adjustment comprises the steps of inputting real-time weather, time and passenger number key data into a data deduction model, dynamically adjusting and analyzing a current scene, and predicting public traffic resource allocation conditions used in future scenes;
The resource allocation recommendation includes outputting the prediction results and pushing the prediction results to a downstream third party system.
9. The airport public transportation pre-allocation processing method based on the big data model according to claim 1, wherein in step S4, the public transportation resource allocation result is monitored in real time, the dynamic adjustment correction result is continuously optimized by combining with the feedback mechanism, and the latest correction result is pushed to the downstream third party system to comprise the steps of monitoring data acquisition, allocation result correction and data feedback;
The monitoring data acquisition comprises real-time monitoring of public transportation at an airport, distinguishing the condition that the real-time data output result is inconsistent with the presumed content in time in a manual mode, and adjusting the weight value of each part Error value;
The distribution result correction comprises that when deduction is carried out, special data is analyzed again, and the weight is readjustedError valueAnd correcting the data result;
the data feedback comprises the steps of outputting the latest correction result and pushing the latest correction result to a downstream third-party system.
10. An airport public traffic pre-allocation processing system based on a big data model, characterized in that the system implements the airport public traffic pre-allocation processing method based on the big data model according to any one of claims 1-9, the system comprises:
The data acquisition module is used for acquiring historical data of flight information, passenger flow, weather conditions, vehicle states, passenger departure information and passenger arrival intention multi-source data;
The data analysis and speculation module is used for speculating the passenger demands and the public transportation service conditions in the historical data by utilizing a machine learning and deep learning model based on a big data processing technology, and generating a data deduction model;
the prediction scheduling module is used for inputting real-time weather, time and passenger data into the data deduction model, and dynamically adjusting public traffic resource allocation by adopting an intelligent optimization algorithm and combining real-time traffic conditions;
The intelligent monitoring and feedback module is used for monitoring the public transportation resource allocation result in real time, continuously optimizing the dynamic adjustment correction result by combining a feedback mechanism, and pushing the latest correction result to a downstream third-party system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202511375081.8A CN120875468A (en) | 2025-09-25 | 2025-09-25 | Airport public transportation preallocation processing system and method based on big data model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202511375081.8A CN120875468A (en) | 2025-09-25 | 2025-09-25 | Airport public transportation preallocation processing system and method based on big data model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN120875468A true CN120875468A (en) | 2025-10-31 |
Family
ID=97465377
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202511375081.8A Pending CN120875468A (en) | 2025-09-25 | 2025-09-25 | Airport public transportation preallocation processing system and method based on big data model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120875468A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103761585A (en) * | 2014-01-27 | 2014-04-30 | 北京市交通运行监测调度中心 | Airport continuous transport dynamic matching, transport capacity monitoring and early warning and intelligent dispatching method |
| US20160350770A1 (en) * | 2015-06-01 | 2016-12-01 | Xerox Corporation | Method, system and processor-readable media for estimating airport usage demand |
| CN111798165A (en) * | 2020-07-31 | 2020-10-20 | 湖南汽车工程职业学院 | Airport taxi scheduling system based on big data |
| CN112598923A (en) * | 2020-12-11 | 2021-04-02 | 湖南汽车工程职业学院 | Airport taxi passenger intelligence joint debugging system based on big data |
| CN114493009A (en) * | 2022-01-27 | 2022-05-13 | 青岛海信网络科技股份有限公司 | Airport vehicle scheduling method, device and equipment based on double-layer model |
| CN114943356A (en) * | 2022-06-06 | 2022-08-26 | 南京航空航天大学 | Airport arrival passenger taxi short-time demand integrated prediction method |
| CN117933612A (en) * | 2023-12-28 | 2024-04-26 | 宁波大学 | Airport taxi demand prediction method |
-
2025
- 2025-09-25 CN CN202511375081.8A patent/CN120875468A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103761585A (en) * | 2014-01-27 | 2014-04-30 | 北京市交通运行监测调度中心 | Airport continuous transport dynamic matching, transport capacity monitoring and early warning and intelligent dispatching method |
| US20160350770A1 (en) * | 2015-06-01 | 2016-12-01 | Xerox Corporation | Method, system and processor-readable media for estimating airport usage demand |
| CN111798165A (en) * | 2020-07-31 | 2020-10-20 | 湖南汽车工程职业学院 | Airport taxi scheduling system based on big data |
| CN112598923A (en) * | 2020-12-11 | 2021-04-02 | 湖南汽车工程职业学院 | Airport taxi passenger intelligence joint debugging system based on big data |
| CN114493009A (en) * | 2022-01-27 | 2022-05-13 | 青岛海信网络科技股份有限公司 | Airport vehicle scheduling method, device and equipment based on double-layer model |
| CN114943356A (en) * | 2022-06-06 | 2022-08-26 | 南京航空航天大学 | Airport arrival passenger taxi short-time demand integrated prediction method |
| CN117933612A (en) * | 2023-12-28 | 2024-04-26 | 宁波大学 | Airport taxi demand prediction method |
Non-Patent Citations (2)
| Title |
|---|
| XIAOJIA GUO ET AL.: "Forecasting Airport Transfer Passenger Flow Using Real-Time Data and Machine Learning", 《MANUFACTURING & SERVICE OPERATIONS MANAGEMENT》, 12 July 2021 (2021-07-12) * |
| 钱鹏程;朱家明;: "基于信息智能化的机场出租车供求匹配系统优化", 高师理科学刊, no. 07, 30 July 2020 (2020-07-30) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | Spatiotemporal evolution of ridesourcing markets under the new restriction policy: A case study in Shanghai | |
| CN111915073A (en) | Short-term prediction method for intercity passenger flow of railway by considering date attribute and weather factor | |
| CN110390349A (en) | Bus passenger flow volume based on XGBoost model predicts modeling method | |
| CN110503104B (en) | A Short-term Remaining Parking Space Prediction Method Based on Convolutional Neural Network | |
| CN111915059A (en) | Seq2seq berth occupancy prediction method based on attention mechanism | |
| CN118314728B (en) | A method for dynamic prediction and accuracy optimization of traffic flow based on one-hot coding | |
| CN114943356B (en) | An integrated prediction method for short-term taxi demand of passengers arriving at the airport | |
| CN117332996A (en) | A collaborative optimization method for ship scheduling and berth allocation based on demand forecasting | |
| CN109344991B (en) | Method for predicting passenger flow of highest section of bus line | |
| CN115359659B (en) | A lane opening and closing configuration method and system | |
| CN116663742B (en) | Regional capacity prediction method based on multi-factor and model fusion | |
| CN110147919A (en) | A kind of public bicycles automatic scheduling method based on price competition mechanism | |
| CN114066503A (en) | Method for forecasting taxi demand based on virtual service area construction | |
| CN109344990A (en) | A short-term load forecasting method and system based on DFS and SVM feature selection | |
| CN115619038A (en) | Data-driven highway network congestion prediction and cause analysis method and system | |
| CN113537569B (en) | A short-term bus passenger flow prediction method and system based on weighted stacked decision trees | |
| CN115269758A (en) | Passenger-guidance-oriented road network passenger flow state deduction method and system | |
| CN118659336A (en) | Novel full-time-period multi-element load prediction method with load participation | |
| CN111798275B (en) | Domestic flight price prediction method | |
| CN120746220B (en) | An Intelligent Emergency Resource Scheduling Optimization Method and System Based on Large Model Analysis | |
| CN120875468A (en) | Airport public transportation preallocation processing system and method based on big data model | |
| CN118313638B (en) | Method, device, equipment and medium for predicting long-term travel demand of inter-city network about car based on SRFE-BLP-converter | |
| CN120280901A (en) | Electric vehicle charging load prediction method based on user behavior analysis | |
| CN119962716A (en) | A method for predicting taxi origin and destination points in urban road network based on event effect | |
| Wang et al. | Research on parking space prediction based on long short-term memory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |