CN112785044B - Real-time full-load rate prediction method, device, equipment and medium for public transport means - Google Patents
Real-time full-load rate prediction method, device, equipment and medium for public transport means Download PDFInfo
- Publication number
- CN112785044B CN112785044B CN202011641120.1A CN202011641120A CN112785044B CN 112785044 B CN112785044 B CN 112785044B CN 202011641120 A CN202011641120 A CN 202011641120A CN 112785044 B CN112785044 B CN 112785044B
- Authority
- CN
- China
- Prior art keywords
- real
- station
- time
- sample data
- public transport
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012549 training Methods 0.000 claims description 28
- 238000004590 computer program Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 8
- 238000010219 correlation analysis Methods 0.000 description 7
- 238000002790 cross-validation Methods 0.000 description 6
- 238000011156 evaluation Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR 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)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of artificial intelligence, and provides a real-time full load rate prediction method and device of a multi-station public transport, computer equipment and a storage medium. The method and the device can accurately predict the real-time full load rate of the public transport means. The method comprises the following steps: the method comprises the steps of obtaining the number of passengers getting on the bus and the arrival time period of the bus corresponding to the public transport to be predicted when the public transport arrives at a current station and a pre-positioned station, determining the time type corresponding to the current running service time of the public transport, taking the number of passengers getting on the bus, the arrival time period of the station and the time type as the influence factors of the real-time full load rate of the public transport, outputting the real-time passenger load predicted value of the public transport corresponding to the current station according to the input characteristic data of the influence factors by using a pre-established real-time passenger load prediction model, and obtaining the real-time full load rate of the public transport according to the ratio of the real-time passenger load predicted value and the rated passenger load corresponding to the public transport.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a real-time full load rate prediction method and device of a multi-station public transport means, a computer device and a storage medium.
Background
In the context of the vigorous development of urban public transport, the number of public transport lines and public transport means continues to increase, but at the same time the passenger flow volume continues to decline. The full load rate is an index for measuring the degree of congestion of the bus during travel and an index for measuring the running benefit of the bus route, and therefore the full load rate is an important parameter for bus route planning, vehicle scheduling and service evaluation. Further, under the background of epidemic situation prevention and control, the real-time full load rate is also an important index and reference basis for measuring the crowd gathering degree in public transport.
However, with current technology, it is not possible to acquire or deduce the real-time loading rate of public transportation.
Disclosure of Invention
In view of the above, it is necessary to provide a real-time full-load rate prediction method, apparatus, computer device and storage medium for multi-station public transportation.
A method of real-time full-load rate prediction for a multi-station public transportation, the method comprising:
acquiring the number of first passengers getting on a bus when a public transport means to be predicted arrives at a current station and the arrival time period of the first station when the public transport means arrives at the current station, and acquiring the number of second passengers getting on the bus when the public transport means arrives at a preorder station and the arrival time period of the second station when the public transport means arrives at the preorder station;
determining a time type corresponding to the current running service time of the public transport means;
obtaining influence factor characteristic data of the real-time full load rate of the public transport means according to the number of the first passengers getting on the bus, the arrival time period of the first stop, the number of the second passengers getting on the bus, the arrival time period of the second stop and the time type;
inputting the characteristic data of the influence factors into a pre-constructed real-time passenger capacity prediction model so that the real-time passenger capacity prediction model outputs a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the characteristic data of the influence factors;
and obtaining the real-time full load rate of the public transport means according to the real-time passenger load predicted value and the rated passenger load corresponding to the public transport means.
A real-time full-load rate prediction apparatus of a multi-station public transportation, comprising:
the passenger number and time interval acquisition module is used for acquiring the number of first passengers getting on the bus when the public transport means to be predicted arrives at the current stop and the arrival time interval of the first stop when the public transport means arrives at the current stop, and acquiring the number of second passengers getting on the bus when the public transport means arrives at the preorder stop and the arrival time interval of the second stop when the public transport means arrives at the preorder stop;
the time type determining module is used for determining the time type corresponding to the current running service time of the public transport means;
the characteristic data acquisition module is used for acquiring influence factor characteristic data of the real-time full load rate of the public transport means according to the first number of passengers getting on the bus, the arrival time period of the first stop, the number of passengers getting on the bus, the arrival time period of the second stop and the time type;
the model prediction module is used for inputting the characteristic data of the influence factors into a pre-constructed real-time passenger capacity prediction model so as to enable the real-time passenger capacity prediction model to output a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the characteristic data of the influence factors;
and the full load rate calculation module is used for obtaining the real-time full load rate of the public transport means according to the real-time passenger load predicted value and the rated passenger load corresponding to the public transport means.
A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
acquiring the number of first passengers getting on a bus when a public transport means to be predicted arrives at a current station and the arrival time period of the first station when the public transport means arrives at the current station, and acquiring the number of second passengers getting on the bus when the public transport means arrives at a preorder station and the arrival time period of the second station when the public transport means arrives at the preorder station; determining a time type corresponding to the current running service time of the public transport means; obtaining influence factor characteristic data of the real-time full load rate of the public transport means according to the number of the first passengers getting on the bus, the arrival time period of the first stop, the number of the second passengers getting on the bus, the arrival time period of the second stop and the time type; inputting the characteristic data of the influence factors into a pre-constructed real-time passenger capacity prediction model so that the real-time passenger capacity prediction model outputs a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the characteristic data of the influence factors; and obtaining the real-time full load rate of the public transport means according to the real-time passenger load predicted value and the rated passenger load corresponding to the public transport means.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring the number of first passengers getting on a bus when a public transport means to be predicted arrives at a current station and the arrival time period of the first station when the public transport means arrives at the current station, and acquiring the number of second passengers getting on the bus when the public transport means arrives at a preorder station and the arrival time period of the second station when the public transport means arrives at the preorder station; determining a time type corresponding to the current running service time of the public transport means; obtaining influence factor characteristic data of the real-time full load rate of the public transport means according to the number of the first passengers getting on the bus, the arrival time period of the first stop, the number of the second passengers getting on the bus, the arrival time period of the second stop and the time type; inputting the characteristic data of the influence factors into a pre-constructed real-time passenger capacity prediction model so that the real-time passenger capacity prediction model outputs a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the characteristic data of the influence factors; and obtaining the real-time full load rate of the public transport means according to the real-time passenger load predicted value and the rated passenger load corresponding to the public transport means.
The real-time full-load rate prediction method, the device, the computer equipment and the storage medium of the multi-station public transport means acquire the number of passengers getting on the bus and the arrival time period of the station corresponding to the arrival of the public transport means to be predicted at the current station and the pre-positioned station thereof, and determining the time type corresponding to the current running service time of the public transport means, taking the number of passengers getting on the bus, the arrival time period of the station and the time type as the influence factors of the real-time full load rate of the public transport means, inputting the corresponding characteristic data of the influence factors into a pre-constructed real-time passenger load prediction model, outputting a real-time passenger load prediction value corresponding to the current station of the public transport means by the real-time passenger load prediction model according to the input characteristic data of the influence factors, and finally obtaining the real-time full load rate of the public transport means according to the ratio of the real-time passenger load prediction value and the rated passenger load corresponding to the public transport means. According to the scheme, the number of passengers getting on the bus, the arrival time period of the bus and the time type corresponding to the current bus stop and the preorder bus stop are used as main influence factors of the real-time full load rate of the public transport means, the real-time passenger load prediction model is used for accurately predicting the real-time full load rate of the public transport means, and the technical problem that the real-time full load rate of the public transport means cannot be acquired or calculated in the traditional technology is solved.
Drawings
FIG. 1 is a schematic flow chart diagram illustrating a method for real-time full rate prediction for a multi-stop mass transit vehicle in one embodiment;
FIG. 2(a) is a graph of the results of an analysis of the relevance of influencing factors in one embodiment;
FIG. 2(b1) is a diagram illustrating the results of another correlation analysis of influencing factors in an embodiment;
FIG. 2(b2) is a graph showing the results of a peak-to-flat correlation analysis in one embodiment;
FIG. 2(c) is a graph showing the correlation analysis result of another influencing factor in an embodiment;
FIG. 3 is a functional block diagram of a real-time full rate prediction method for a multi-station public transportation vehicle in one embodiment;
FIG. 4 is a functional block diagram of model training in one embodiment;
FIG. 5(a) is a graph of the results of a prediction error analysis in one embodiment;
FIG. 5(b) is a graph showing the results of another prediction error analysis in one embodiment;
FIG. 5(c) is a graph showing the result of analysis of measurement errors according to still another embodiment;
FIG. 6(a) is a graph of the results of a prediction error ratio analysis in one embodiment;
FIG. 6(b) is a graph of the results of another prediction error ratio analysis in one embodiment;
FIG. 6(c) is a graph illustrating the results of a prediction error ratio analysis in an embodiment;
FIG. 7 is a block diagram of a real-time full rate prediction apparatus of a multi-station public transportation vehicle in one embodiment;
FIG. 8 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In one embodiment, as shown in fig. 1, a real-time full-load rate prediction method for a multi-station public transportation vehicle is provided, and the method can be implemented by a computer device such as a server, a terminal, and the like, wherein the terminal can be, but is not limited to, various personal computers, laptops and tablet computers, and the server can be implemented by an independent server or a server cluster consisting of a plurality of servers. The real-time full-load rate prediction method of the multi-station public transport means can comprise the following steps:
step S101, obtaining the number of first passengers getting on the bus when the public transport means to be predicted arrives at the current station and the arrival time period of the first station when the public transport means arrives at the current station, and obtaining the number of second passengers getting on the bus when the public transport means arrives at the preorder station and the arrival time period of the second station when the public transport means arrives at the preorder station.
The public transport means to be predicted can be buses, subways and the like. The public transport vehicles usually travel according to a specific route, the route comprises a plurality of stations, the public transport vehicles stop at each station, passengers in the vehicles can get off the vehicles, passengers outside the vehicles can get on the vehicles, the number of passengers getting on the vehicles at a certain station is the number of passengers getting on the vehicles at the station, and the time period when the public transport vehicles arrive at the certain station is the station arrival time period. Specifically, a station that a public transportation means currently arrives at while traveling on a specific route is referred to as a current station, and a station that precedes the current station is referred to as a preceding station. In the step, the computer equipment obtains the number of passengers getting on the bus, which arrive at the current station, of the public transport means to be predicted, records the number as a first passenger getting on the bus, obtains the arrival time period of the station where the current station is located, and records the arrival time period as a first station arrival time period; in addition, the computer equipment also acquires the number of passengers getting on the bus arriving at the preorder station of the public transport means, and records the number as the second passenger getting on the bus, and acquires the arrival time of the station arriving at the preorder station, and records the arrival time as the second station arrival time.
In some embodiments, the number of preamble stations may be multiple. In this regard, the second boarding passenger number needs to include the number of boarding passengers corresponding to each preorder station, and the second station arrival time period includes the station arrival time period corresponding to each preorder station, that is, the number of boarding passengers and the station arrival time period of each preorder station need to be obtained.
Step S102, determining a time type corresponding to the current running service time of the public transport means;
in this step, the operation service time of the public transportation means may be divided in units of days, that is, each day is used as the operation service time of the public transportation means, and when the public transportation means provides services on the travel route, the current day is the operation service time of the public transportation means. The time type can be used for representing that the operation service time is a working day or a non-working day, namely in the step, the computer device can judge whether the current day of the operation service provided by the public transport means is the working day or the non-working day.
In some embodiments, each day may be divided into a plurality of time periods in minutes, for example, each day may be divided into 144 time periods in 10 minutes, and thus, when a public transport arrives at any station, the station arrival time period may be determined from the 144 time periods.
The above steps S101 and S102 are mainly to acquire the number of passengers getting on the bus, the station arrival time period, and whether it is a working day as main factors affecting the real-time full-load rate of the public transportation. The correlation between the influence factors and the actual passenger number of the public transport means is analyzed by combining the images in. Wherein, as shown in FIG. 2(a), the Spearman correlation coefficient of 0.76 is high intensity correlation; as shown in fig. 2(b1), the time-interval correlation analysis shows that the Spearman correlation coefficient is 0.65, which is high intensity correlation, further, the time interval shows a certain variation trend with the number of actual passengers in the bus compartment, and has two peaks in the morning and evening similar to the variation trend of urban traffic volume with the time interval, wherein the early peak is obviously larger than the late peak, as shown in fig. 2(b2), the peak average passenger number is obviously higher than the flat peak, wherein the flat peak average passenger number is 15.35, the peak average passenger number is 22.91, and the Kendall correlation coefficient is 0.12, which is low intensity correlation; as shown in fig. 2(c), whether the correlation analysis is a weekday correlation analysis or not, the Kendall correlation coefficient of 0.02 is low intensity correlation, and it can be seen that the average number of passengers is significantly higher in weekdays than in non-weekdays, where the average number of passengers is 19.66 and the average number of passengers is 15.82.
Step S103, obtaining influence factor characteristic data of the real-time full load rate of the public transport means according to the number of the first passengers getting on the bus, the arrival time period of the first stop, the number of the second passengers getting on the bus, the arrival time period of the second stop and the time type;
by performing correlation analysis on the three influence factors, the computer device can use the influence factors as variables for predicting the real-time passenger capacity, and the corresponding data variables are input into a pre-constructed real-time passenger capacity prediction model in the subsequent step for predicting the real-time passenger capacity. The computer device converts the number of passengers getting on the bus, the arrival time period of the first stop, the number of passengers getting on the bus, the arrival time period of the second stop and the time type (namely whether the time is a working day) which are main influence factors of the real-time full load rate of the public transport means into corresponding influence factor characteristic data.
In some embodiments, the computer device may obtain the characteristic data of the influence factors of the real-time full load rate of the public transportation vehicle by adopting the following modes:
obtaining a station boarding passenger number sequence of the public transport means as first influence factor characteristic data according to the first boarding passenger number and the second boarding passenger number; obtaining a station arrival time interval sequence of the public transport means as second influence factor characteristic data according to the first station arrival time interval and the second station arrival time interval; taking the time type as third influence factor characteristic data; and taking the first influence factor characteristic data, the second influence factor characteristic data and the third influence factor characteristic data as the influence factor characteristic data of the real-time full load rate of the public transport means.
Taking 30 stations and taking 10 minutes as a unit to divide a day into 144 time intervals as an example, each station has a corresponding number of passengers getting on the bus, the number of passengers getting on the bus at a station which does not arrive or has no passengers getting on the bus can be marked as 0, the number of passengers getting on the bus at the station corresponding to the 30 stations is arranged according to the sequence of the stations to form a station passenger number sequence of the public transportation means, the station passenger number sequence can be obtained according to the first number of passengers getting on the bus and the second number of passengers getting on the bus, and the station passenger number sequence getting on the bus is used as first influence factor characteristic data and also used as an input variable of a real-time passenger capacity prediction model in the subsequent step; the computer equipment can also form the arrival time periods corresponding to the 30 stations into a station arrival time period sequence of the public transport means in such a way, similarly, the station arrival time period sequence can be obtained according to the arrival time period of the first station and the arrival time period of the second station, the stations which do not arrive can be marked as empty, and the station arrival time period sequence is used as second influence factor characteristic data and also used as a second input variable of a real-time passenger capacity prediction model; the time type can be represented by 0 or 1, for example, 0 represents working day, 1 represents non-working day, etc., and the time type can be used as the third influence factor characteristic data and also as the third input variable of the real-time passenger capacity prediction model. Specifically, the attribute categories of the three influencing factor characteristic data/input variables are shown in table 1 below:
TABLE 1
In some embodiments, the computer device may preprocess data of an arrival time period to form second influence factor characteristic data, and the specific steps include:
obtaining a station arrival time interval sequence according to the first station arrival time interval and the second station arrival time interval; and carrying out one-hot encoding operation on the data of the station arrival time interval sequence to obtain second influence factor characteristic data.
In this embodiment, after obtaining a station arrival time period sequence according to a first station arrival time period and a second station arrival time period, the computer device performs a one-hot coding operation on data of the station arrival time period sequence belonging to a category variable, converts a value range of the variable into 0 or 1, and performs the one-hot coding operation on the time period, so that finally formed influence factor characteristic data is as shown in table 2 below:
TABLE 2
Step S104, inputting the characteristic data of the influence factors into a pre-constructed real-time passenger capacity prediction model so that the real-time passenger capacity prediction model outputs a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the characteristic data of the influence factors;
the three influence factor characteristic data are input into a pre-constructed real-time passenger capacity prediction model for prediction, the real-time passenger capacity prediction model can be constructed based on an XGboost model, the real-time passenger capacity prediction model can output a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the input influence factor characteristic data, namely the real-time passenger capacity prediction model predicts the current actual passenger capacity of the public transport means according to the three influence factor characteristic data to obtain the corresponding prediction value.
Referring to fig. 3, when the real-time full load factor prediction is performed on the public transportation, the number of passengers getting on the bus at the current station and the station corresponding to the preorder station, the station arrival time period, and the time type (i.e., whether it is a working day) corresponding to the current operation service time of the public transportation are obtained, then the corresponding influence factor characteristic data are input to the pre-constructed real-time passenger capacity prediction model, and the real-time passenger capacity prediction value is output by the real-time passenger capacity prediction model.
And S105, obtaining the real-time full load rate of the public transport means according to the real-time passenger load predicted value and the rated passenger load corresponding to the public transport means.
The real-time passenger capacity prediction value is obtained by the computer device, and then the real-time full load rate of the public transport means is determined according to the ratio of the real-time passenger capacity prediction value to the rated passenger capacity corresponding to the public transport means.
The real-time full load rate prediction method of the multi-station public transport means obtains the number of passengers getting on the bus and the arrival time period of the station corresponding to the arrival of the public transport means to be predicted at the current station and the preorder station thereof, and determining the time type corresponding to the current running service time of the public transport means, taking the number of passengers getting on the bus, the arrival time period of the station and the time type as the influence factors of the real-time full load rate of the public transport means, inputting the corresponding characteristic data of the influence factors into a pre-constructed real-time passenger load prediction model, outputting a real-time passenger load prediction value corresponding to the current station of the public transport means by the real-time passenger load prediction model according to the input characteristic data of the influence factors, and finally obtaining the real-time full load rate of the public transport means according to the ratio of the real-time passenger load prediction value and the rated passenger load corresponding to the public transport means. According to the scheme, the number of passengers getting on the bus, the arrival time period of the bus and the time type corresponding to the current bus stop and the preorder bus stop are used as main influence factors of the real-time full load rate of the public transport means, the real-time passenger load prediction model is used for accurately predicting the real-time full load rate of the public transport means, and the technical problem that the real-time full load rate of the public transport means cannot be acquired or calculated in the traditional technology is solved.
In one embodiment, the computer device may construct the real-time passenger capacity prediction model based on the XGBoost model by the following steps:
firstly, training sample data and a real passenger capacity value for model training are obtained.
The training sample data can comprise first boarding passenger number sample data, first station arrival time period sample data, second boarding passenger number sample data, second station arrival time period sample data and time type sample data. Specifically, the training sample data can be acquired according to historical OD data, real-time bus payment data and other data of the public transport means, the real passenger capacity value can be collected and counted manually to form training sample data and a corresponding real passenger capacity value, and in the sample data acquisition process, the information such as the number of passengers getting on the bus, the arrival time period of the bus, whether the bus is a working day and the actual passenger capacity when the bus arrives at each bus station when the bus arrives at each station of a specific line can be counted, so that first passenger number sample data, first bus arrival time period sample data, second passenger number sample data, second bus arrival time period sample data, time type sample data and real passenger capacity value corresponding to each station are formed.
Secondly, obtaining a station boarding passenger number sequence sample as a first influence factor characteristic sample data according to the first boarding passenger number sample data and the second boarding passenger number sample data; obtaining a station arrival time period sequence sample according to the first station arrival time period sample data and the second station arrival time period sample data, and performing one-hot encoding operation on the data of the station arrival time period sequence sample to obtain second influence factor characteristic sample data; taking time type sample data as third influence factor characteristic sample data;
for each site, the first, second, and third influencing factor feature sample data may be formed, and the forming process of the corresponding influencing factor feature sample data may refer to the forming process of the influencing factor feature data in the above embodiments. And for the sample data of the arrival time period of the station, the one-hot encoding operation is required to be carried out, so that the value range of the variable of the arrival time period of the station is converted into 0 or 1.
And finally, taking the first influence factor characteristic sample data, the second influence factor characteristic sample data and the third influence factor characteristic sample data as influence factor characteristic sample data, training the XGboost model by using the influence factor characteristic sample data and the passenger capacity real value, and constructing to obtain a real-time passenger capacity prediction model. The XGboost model can effectively solve the problem of high dimensionality, can specify the default direction of branches for sparse data, and improves algorithm efficiency.
Further, in some embodiments, the training of the XGBoost model by using the impact factor characteristic sample data and the passenger capacity real value to construct a real-time passenger capacity prediction model specifically includes:
firstly, initializing a general parameter, a learning task parameter set and a lifting parameter set of an XGboost model; the parameters contained in the general parameters and the learning task parameter set are not updated in the model training process; and then updating parameters contained in a lifting parameter set of the XGboost model based on a network search cross-validation method by using the influence factor characteristic sample data and the real passenger capacity value, and constructing to obtain a real-time passenger capacity prediction model.
Specifically, the model training process is described as a whole with reference to fig. 4, and first, the model training process is converted into an input data set through feature engineering, that is, the serial sample of the number of passengers getting on the station is obtained according to the sample data of the number of passengers getting on the vehicle in the first embodiment and the sample data of the number of passengers getting on the vehicle in the second embodiment, the serial sample of the number of passengers getting on the station is used as the first influence factor feature sample data, the serial sample of the arrival time period of the station is obtained according to the sample data of the arrival time period of the first station and the sample data of the arrival time period of the second station, the second influence factor feature sample data and the time type are used as the third influence factor feature sample data, and finally, the attribute of the obtained corresponding influence feature sample data may refer to table 2 above.
TABLE 3
After the influence factor characteristic sample data is obtained, the XGboost model can be trained, and the model is trained, namely, a proper value is given to the parameters of the model. The general parameters are used for setting the overall function, the lifting parameter set is used for setting the parameters of the regression tree of each step, the learning task parameter set guides the XGboost model to execute the optimization task, and specific parameter indexes, definitions and usage methods corresponding to the three types of parameters are shown in the table 3. Based on the method, the process of updating the parameters of the XGboost model comprises the following steps:
firstly, parameter initialization is performed, that is, a general parameter and a learning task parameter set are determined, the general parameter and the learning task parameter set are not updated in the process of updating the parameters of the XGBoost model, and the parameter initialization process also needs to initialize a lifting parameter set.
Then a parameter update procedure is performed. In this embodiment, the computer device performs parameter update on the parameters included in the lifting parameter set by using a network search cross-validation method. The network search cross-validation method returns evaluation index scores under all parameter combinations in a cross-validation mode by traversing all permutation combinations of the transmitted parameters, and the specific parameter updating process of the network search cross-validation method comprises the following steps:
1. the initialized parameter combination W is set. In the parameter combination W, the parameters included in the general parameter and the learning task parameter set are both marked, and the parameters included in the boosting parameter set are not marked.
2. Selecting one of the parameters w which is not marked and the value theta of the parameter w
3. And introducing a plurality of values near theta to form a permutation combination D, traversing candidate values in the permutation combination D to give a parameter w, and returning the evaluation index scores of XGboost model training under all parameter combinations in a cross validation mode, so that the value theta of the parameter w is updated to be the candidate value with the best evaluation index score.
4. If the optimal parameter value of the parameter w needs to be further adjusted, returning to the step 3; otherwise, go to step 5.
5. The adjusted parameter w is marked. If all the parameters in the parameter combination W are marked, saving the current XGboost model structure and parameters; otherwise, go back to step 2 to continue marking.
After all parameters of the XGboost model are determined through the process, the construction of the real-time passenger capacity prediction model based on the XGboost is completed.
Before the real-time passenger capacity prediction model is constructed and applied to real-time passenger capacity prediction, the effectiveness of the real-time passenger capacity prediction model can be verified. Specifically, 61000 pieces of example data of one month are used for effectiveness analysis, wherein the samples are randomly extracted and divided into 48800 training sets and 12200 testing sets. Based on this, the errors and error ratios of the test set data were analyzed separately:
firstly, for error analysis:
12200 pieces of data in the test set have an average error of 2.74 and a standard deviation of 3.34, which shows that the constructed real-time passenger capacity prediction model has certain precision on prediction of real-time passenger capacity and has small prediction volatility. Specifically, referring to fig. 5(a) to 5(c), wherein fig. 5(a) shows an error with respect to a time period, which is divided into two peaks, but the early peak is much smaller than the late peak, mainly because of a large number of passengers during the early peak and the high prediction accuracy, and further, as shown in fig. 5(b), the flat peak error is slightly lower than the high peak error with respect to the prediction errors of the flat peak and the high peak; as shown in fig. 5(c), the prediction error between the working day and the non-working day is substantially equivalent to the error between the working day and the non-working day, and the error between the working day and the non-working day is slightly higher than the error between the working day and the non-working day.
Secondly, analyzing error proportion:
the average error ratio is 0.33, the standard deviation is 0.61, and the constructed real-time passenger capacity prediction model has certain precision on the prediction of the real-time passenger capacity and has small prediction volatility. Specifically, referring to fig. 6(a) to 6(c), wherein fig. 6(a) shows error ratios with respect to time periods, it can be seen that the error ratios of the time periods are different from the errors, and the distribution thereof is substantially independent of the time periods; for the peak with flat peak, as shown in fig. 6(b), the error ratio between the peak and the flat peak is substantially equivalent, and the peak is slightly higher; for weekdays and non-weekdays, as shown in fig. 6(c), the error ratios between weekdays and non-weekdays are substantially equivalent, with non-weekdays being slightly higher.
Therefore, the XGboost-based real-time passenger capacity prediction model adopted by the application, has certain precision and small calculation fluctuation in the effectiveness analysis, can be applied to the real-time passenger load calculation/prediction of public transport means, in practical application, the number of passengers getting on the bus at the preceding station and the current station, the arrival time period of the station and whether the station is a working day are only required to be taken as influence factor characteristics and converted into corresponding influence factor characteristic data, the influence factor characteristic data are input into a pre-constructed real-time passenger capacity prediction model, the method can obtain the predicted value of the real-time passenger capacity output by the real-time passenger capacity prediction model, thereby obtaining the predicted value of the real-time full load rate based on the ratio of the predicted value of the real-time passenger capacity to the rated passenger capacity, realizing the prediction of the real-time full load rate of the public transport means, and solving the technical problem that the real-time full load rate of the public transport means can not be acquired or calculated in the prior art.
It should be understood that although the various steps in the flowcharts of fig. 1 and 4 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 1 and 4 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.
In one embodiment, as shown in fig. 7, there is provided a real-time full-load rate prediction apparatus of a multi-station public transportation, the apparatus 700 may include:
a passenger number and time period obtaining module 701, configured to obtain the number of first passengers getting on a bus arriving at a current stop and the arrival time period of the first stop when the public transportation arrives at the current stop, and obtain the number of second passengers getting on a bus arriving at a preorder stop and the arrival time period of the second stop when the public transportation arrives at the preorder stop;
a time type determining module 702, configured to determine a time type corresponding to a current running service time of the public transportation;
the characteristic data acquisition module 703 is used for acquiring influence factor characteristic data of the real-time full load rate of the public transportation means according to the first number of passengers getting on the bus, the arrival time period of the first stop, the number of passengers getting on the bus, the arrival time period of the second stop and the time type;
the model prediction module 704 is used for inputting the characteristic data of the influence factors into a pre-constructed real-time passenger capacity prediction model so that the real-time passenger capacity prediction model outputs a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the characteristic data of the influence factors;
and a full load rate calculation module 705, configured to obtain a real-time full load rate of the public transportation vehicle according to the real-time passenger load predicted value and a rated passenger load corresponding to the public transportation vehicle.
In one embodiment, the number of preamble stations is multiple; the second number of passengers getting on the train comprises the number of passengers getting on the train corresponding to each preorder station; the second station arrival time period includes station arrival time periods corresponding to the preamble stations, respectively.
In one embodiment, the time type is used to characterize the operational service time as a weekday or a non-weekday.
In one embodiment, the characteristic data obtaining module 703 is further configured to obtain a station boarding passenger number sequence of the public transportation means as first influencing factor characteristic data according to the first boarding passenger number and the second boarding passenger number; obtaining a station arrival time interval sequence of the public transport means as second influence factor characteristic data according to the first station arrival time interval and the second station arrival time interval; taking the time type as third influence factor characteristic data; and taking the first influence factor characteristic data, the second influence factor characteristic data and the third influence factor characteristic data as the influence factor characteristic data of the real-time full load rate of the public transport means.
In an embodiment, the characteristic data obtaining module 703 is further configured to obtain the station arrival time interval sequence according to the first station arrival time interval and the second station arrival time interval; and carrying out one-hot encoding operation on the data of the station arrival time interval sequence to obtain the second influence factor characteristic data.
In one embodiment, the real-time passenger capacity prediction model is constructed on the basis of an XGboost model; the apparatus 700 may further include: the model construction module is used for acquiring training sample data and a passenger capacity real value for model training; the training sample data comprises first boarding passenger number sample data, first station arrival time period sample data, second boarding passenger number sample data, second station arrival time period sample data and time type sample data; obtaining a station boarding passenger number sequence sample serving as a first influence factor characteristic sample data according to the first boarding passenger number sample data and the second boarding passenger number sample data; obtaining a station arrival time period sequence sample according to the first station arrival time period sample data and the second station arrival time period sample data, and performing one-hot encoding operation on the data of the station arrival time period sequence sample to obtain second influence factor characteristic sample data; taking the time type sample data as third influence factor characteristic sample data; and taking the first influence factor characteristic sample data, the second influence factor characteristic sample data and the third influence factor characteristic sample data as influence factor characteristic sample data, training an XGboost model by using the influence factor characteristic sample data and the passenger capacity real value, and constructing to obtain the real-time passenger capacity prediction model.
In one embodiment, the model construction module is further configured to initialize a general parameter, a learning task parameter set, and a lifting parameter set of the XGBoost model; the parameters contained in the general parameter and the learning task parameter set are not updated in the model training process; and updating parameters contained in the lifting parameter set of the XGboost model based on a network search cross verification method by using the influence factor characteristic sample data and the real passenger capacity value, and constructing to obtain the real-time passenger capacity prediction model.
The specific limitations of the real-time full rate prediction device for multi-station public transportation means can be referred to the limitations of the real-time full rate prediction method for multi-station public transportation means, and will not be described herein again. The modules in the real-time full-load-rate prediction device of the multi-station public transportation vehicle can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 8. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer equipment is used for storing data such as the number of passengers getting on the bus, the arrival time of a station, the time type, the characteristic data of influence factors, the predicted value of real-time passenger capacity, the real-time full load rate and the like. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a real-time full rate prediction method for a multi-station public transportation.
Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method for real-time full-load rate prediction for a multi-station public transportation, the method comprising:
acquiring the number of first passengers getting on a bus when a public transport means to be predicted arrives at a current station and the arrival time period of the first station when the public transport means arrives at the current station, and acquiring the number of second passengers getting on the bus when the public transport means arrives at a preorder station and the arrival time period of the second station when the public transport means arrives at the preorder station;
determining a time type corresponding to the current running service time of the public transport means;
taking the first number of passengers getting on the bus, the arrival time period of the first stop, the number of passengers getting on the bus, the arrival time period of the second stop and the time type as the characteristic data of the influence factors of the real-time full load rate of the public transport means;
inputting the characteristic data of the influence factors into a pre-constructed real-time passenger capacity prediction model so that the real-time passenger capacity prediction model outputs a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the characteristic data of the influence factors;
obtaining the real-time full load rate of the public transport means according to the real-time passenger load predicted value and the rated passenger load corresponding to the public transport means;
the method further comprises the following steps:
acquiring training sample data and a real passenger capacity value for model training; the training sample data comprises first boarding passenger number sample data, first station arrival time period sample data, second boarding passenger number sample data, second station arrival time period sample data and time type sample data;
obtaining a station boarding passenger number sequence sample serving as a first influence factor characteristic sample data according to the first boarding passenger number sample data and the second boarding passenger number sample data;
obtaining a station arrival time period sequence sample according to the first station arrival time period sample data and the second station arrival time period sample data, and performing one-hot encoding operation on the data of the station arrival time period sequence sample to obtain second influence factor characteristic sample data;
taking the time type sample data as third influence factor characteristic sample data;
and taking the first influence factor characteristic sample data, the second influence factor characteristic sample data and the third influence factor characteristic sample data as influence factor characteristic sample data, training a real-time passenger capacity prediction model to be trained by using the influence factor characteristic sample data and the real passenger capacity value, and constructing to obtain the real-time passenger capacity prediction model.
2. The method of claim 1 wherein said preamble sites are plural in number; the second number of passengers getting on the train comprises the number of passengers getting on the train corresponding to each preorder station; the second station arrival time period includes station arrival time periods corresponding to the preamble stations, respectively.
3. The method of claim 1, wherein the time type is used to characterize the operational service time as a weekday or a non-weekday.
4. The method of claim 1, wherein said characterizing said first number of passengers boarding, first stop arrival period, second number of passengers boarding, second stop arrival period, and type of time as contributor characteristics of real-time full-load rate of said mass-transit vehicle comprises:
obtaining a station boarding passenger number sequence of the public transport means as first influence factor characteristic data according to the first boarding passenger number and the second boarding passenger number;
obtaining a station arrival time interval sequence of the public transport means as second influence factor characteristic data according to the first station arrival time interval and the second station arrival time interval;
taking the time type as third influence factor characteristic data;
and taking the first influence factor characteristic data, the second influence factor characteristic data and the third influence factor characteristic data as the influence factor characteristic data of the real-time full load rate of the public transport means.
5. The method according to claim 4, wherein the obtaining of the station arrival time interval sequence of the public transportation means as the second influence factor characteristic data according to the first station arrival time interval and the second station arrival time interval comprises:
obtaining the station arrival time interval sequence according to the first station arrival time interval and the second station arrival time interval;
and carrying out one-hot encoding operation on the data of the station arrival time interval sequence to obtain the second influence factor characteristic data.
6. The method according to any one of claims 1 to 5, wherein the real-time passenger capacity prediction model is constructed based on an XGboost model.
7. The method according to claim 6, wherein the training of the real-time passenger capacity prediction model to be trained by using the influencing factor feature sample data and the real passenger capacity value to construct the real-time passenger capacity prediction model comprises:
initializing a general parameter, a learning task parameter set and a lifting parameter set of the XGboost model; the parameters contained in the general parameter and the learning task parameter set are not updated in the model training process;
and updating parameters contained in the lifting parameter set of the XGboost model based on a network search cross verification method by using the influence factor characteristic sample data and the real passenger capacity value, and constructing to obtain the real-time passenger capacity prediction model.
8. A real-time full-load rate prediction apparatus for a multi-station public transportation, comprising:
the passenger number and time interval acquisition module is used for acquiring the number of first passengers getting on the bus when the public transport means to be predicted arrives at the current stop and the arrival time interval of the first stop when the public transport means arrives at the current stop, and acquiring the number of second passengers getting on the bus when the public transport means arrives at the preorder stop and the arrival time interval of the second stop when the public transport means arrives at the preorder stop;
the time type determining module is used for determining the time type corresponding to the current running service time of the public transport means;
the characteristic data acquisition module is used for taking the first number of passengers getting on the bus, the arrival time period of a first stop, the number of passengers getting on the bus, the arrival time period of a second stop and the time type as the characteristic data of the influence factors of the real-time full load rate of the public transport means;
the model prediction module is used for inputting the characteristic data of the influence factors into a pre-constructed real-time passenger capacity prediction model so as to enable the real-time passenger capacity prediction model to output a real-time passenger capacity prediction value corresponding to the current station of the public transport means according to the characteristic data of the influence factors;
the full load rate calculation module is used for obtaining the real-time full load rate of the public transport means according to the real-time passenger load predicted value and the rated passenger load corresponding to the public transport means;
the model construction module is used for acquiring training sample data and a passenger capacity real value for model training; the training sample data comprises first boarding passenger number sample data, first station arrival time period sample data, second boarding passenger number sample data, second station arrival time period sample data and time type sample data; obtaining a station boarding passenger number sequence sample serving as a first influence factor characteristic sample data according to the first boarding passenger number sample data and the second boarding passenger number sample data; obtaining a station arrival time period sequence sample according to the first station arrival time period sample data and the second station arrival time period sample data, and performing one-hot encoding operation on the data of the station arrival time period sequence sample to obtain second influence factor characteristic sample data; taking the time type sample data as third influence factor characteristic sample data; and taking the first influence factor characteristic sample data, the second influence factor characteristic sample data and the third influence factor characteristic sample data as influence factor characteristic sample data, training a real-time passenger capacity prediction model to be trained by using the influence factor characteristic sample data and the real passenger capacity value, and constructing to obtain the real-time passenger capacity prediction model.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011641120.1A CN112785044B (en) | 2020-12-31 | 2020-12-31 | Real-time full-load rate prediction method, device, equipment and medium for public transport means |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011641120.1A CN112785044B (en) | 2020-12-31 | 2020-12-31 | Real-time full-load rate prediction method, device, equipment and medium for public transport means |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112785044A CN112785044A (en) | 2021-05-11 |
| CN112785044B true CN112785044B (en) | 2021-11-23 |
Family
ID=75753469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011641120.1A Active CN112785044B (en) | 2020-12-31 | 2020-12-31 | Real-time full-load rate prediction method, device, equipment and medium for public transport means |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112785044B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114415610B (en) * | 2021-12-28 | 2023-08-15 | 深圳市优必选科技股份有限公司 | Scheduling method and device for robot, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533214A (en) * | 2019-07-12 | 2019-12-03 | 北京航空航天大学 | A kind of subway passenger flow Forecasting Approach for Short-term based on XGBoost algorithm |
| CN111259714A (en) * | 2019-09-27 | 2020-06-09 | 华南理工大学 | Subway train carriage passenger flow detection and prediction and platform waiting guidance system |
| CN111723871A (en) * | 2020-07-09 | 2020-09-29 | 广州市公共交通数据管理中心 | Estimation method for real-time carriage full load rate of bus |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102085593B1 (en) * | 2019-09-16 | 2020-03-06 | 포항공과대학교 산학협력단 | Method and device for detecting posting bot for blockchain SNS based on machine learning |
-
2020
- 2020-12-31 CN CN202011641120.1A patent/CN112785044B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533214A (en) * | 2019-07-12 | 2019-12-03 | 北京航空航天大学 | A kind of subway passenger flow Forecasting Approach for Short-term based on XGBoost algorithm |
| CN111259714A (en) * | 2019-09-27 | 2020-06-09 | 华南理工大学 | Subway train carriage passenger flow detection and prediction and platform waiting guidance system |
| CN111723871A (en) * | 2020-07-09 | 2020-09-29 | 广州市公共交通数据管理中心 | Estimation method for real-time carriage full load rate of bus |
Non-Patent Citations (1)
| Title |
|---|
| 利用XGBoost和SVR算法的地铁站客流量模型研究;李蕙萱、吴瑞溢;《三明学院学报》;20191220;第36卷(第6期);第56-64页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112785044A (en) | 2021-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112700072B (en) | Traffic condition prediction method, electronic device, and storage medium | |
| Yang et al. | Stochastic skyline route planning under time-varying uncertainty | |
| CN111126668B (en) | Spark operation time prediction method and device based on graph convolution network | |
| CN109920248B (en) | Bus arrival time prediction method based on GRU neural network | |
| CN111612281A (en) | Method and device for predicting pedestrian flow peak value of subway station and computer equipment | |
| Cools et al. | Assessment of the effect of micro-simulation error on key travel indices: Evidence from the activity-based model Feathers | |
| CN109740818A (en) | A kind of probability density forecasting system applied to en-route sector traffic | |
| CN112785044B (en) | Real-time full-load rate prediction method, device, equipment and medium for public transport means | |
| CN113361690A (en) | Water quality prediction model training method, water quality prediction device, water quality prediction equipment and medium | |
| CN113838303B (en) | Parking lot recommendation method and device, electronic equipment and storage medium | |
| CN109583086A (en) | Distribution transformer heavy-overload prediction technique and terminal device | |
| CN112419711A (en) | Closed parking lot parking demand prediction method based on improved GMDH algorithm | |
| Xu et al. | Network-flow-based efficient vehicle dispatch for city-scale ride-hailing systems | |
| CN109978241B (en) | Method and device for determining charging load of electric automobile | |
| Büchel et al. | Modeling conditional dependencies for bus travel time estimation | |
| CN113837383B (en) | Model training method and device, electronic equipment and storage medium | |
| JP2004023114A (en) | Method and system for predicting communication traffic | |
| CN111667689B (en) | Method, device and computer device for predicting vehicle travel time | |
| Rodríguez-Rueda et al. | Origin–Destination matrix estimation and prediction from socioeconomic variables using automatic feature selection procedure-based machine learning model | |
| CN114548463A (en) | Line information prediction method, line information prediction device, computer equipment and storage medium | |
| CN109615187B (en) | OD matrix evaluation method, bus load simulation method and device | |
| CN114048886A (en) | Effective berthage combination prediction method based on neural network, terminal and storage medium | |
| CN114492905A (en) | Customer appeal rate prediction method and device based on multi-model fusion and computer equipment | |
| CN113077079A (en) | Data-driven rail transit new line access passenger flow prediction method | |
| Karim et al. | Long-Term Traffic Prediction Based on Stacked GCN Model |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |