CN110378522A - Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information - Google Patents
Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information Download PDFInfo
- Publication number
- CN110378522A CN110378522A CN201910590756.9A CN201910590756A CN110378522A CN 110378522 A CN110378522 A CN 110378522A CN 201910590756 A CN201910590756 A CN 201910590756A CN 110378522 A CN110378522 A CN 110378522A
- Authority
- CN
- China
- Prior art keywords
- preset
- data
- area
- historical
- time
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012549 training Methods 0.000 claims description 54
- 230000006870 function Effects 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 description 44
- 238000010586 diagram Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 4
- 238000004880 explosion Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000002596 correlated effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 235000012054 meals Nutrition 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
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
- 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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Telephonic Communication Services (AREA)
Abstract
This disclosure relates to a kind of method, apparatus, storage medium and the electronic equipment of prediction dispatching status information, available target area includes time identifier data, the first weather data and the region identity data for characterizing corresponding region in the corresponding provincial characteristics data of the delivery service data of predetermined time and the target area, the provincial characteristics data；According to the delivery service data and the provincial characteristics data by dispatching status information prediction model, dispatching status information of the target area in the preset duration from the predetermined time is obtained.
Description
Technical Field
The present disclosure relates to the field of internet technologies, and in particular, to a method and an apparatus for predicting delivery status information, a storage medium, and an electronic device.
Background
The instant distribution business is a business mode which is started in recent years and is rapidly developed, if a distribution area is in an order explosion condition, the area is in a state of unbalanced supply and demand, and at the moment, corresponding measures should be taken to inhibit ordering willingness of a user, so that distribution pressure of the area is relieved, and user experience is improved.
In general, distribution state information (such as average distribution time length, order quantity or average load) in a future preset time length can be used as a main basis for triggering regional explosive orders, in the related art, a global regression model is mainly used for predicting the distribution state information in the future preset time length, but the global regression model is obtained by training according to distribution service data of all regions, so that the problem of global optimization can be solved, namely the overall accuracy of all regions is ensured, and when the global regression model is used for predicting the distribution state information of extreme regions (such as high-load regions, regions with severe weather and the like), the predicted accuracy is poor.
Disclosure of Invention
The invention aims to provide a method, a device, a storage medium and an electronic device for predicting delivery state information.
In a first aspect, a method for predicting delivery status information is provided, the method comprising: acquiring distribution service data of a target area at a preset time and area characteristic data corresponding to the target area, wherein the area characteristic data comprises time identification data, first day gas data and area identification data for representing the corresponding area; and obtaining the distribution state information of the target area within a preset time from the preset moment according to the distribution service data and the area characteristic data through a distribution state information prediction model.
Optionally, the delivery status information includes average delivery duration information, order quantity information, and average order load information; the average order load information is used to characterize the amount of orders evenly distributed to each of the dispatchers over the preset length of time.
Optionally, the delivery service data comprises one or more of the following data items: the data of the regional order quantity of the target region, the regional order load data of the target region, the regional distributor data of the target region, the order service duration data of the target region in a preset historical duration taking the preset time as an ending time and the historical order statistical data of the target region in a preset historical time period.
Optionally, the delivery status information prediction model is trained by: acquiring historical distribution service data of a preset sample region at a preset historical moment, historical region characteristic data corresponding to the preset sample region and historical distribution state information of the preset sample region within the preset duration from the preset historical moment; and training by taking the historical distribution service data corresponding to the preset sample region at the preset historical moment, the historical region feature data corresponding to the preset sample region and the historical distribution state information of the preset sample region within the preset duration from the preset historical moment as model training samples to obtain the distribution state information prediction model.
Optionally, the training, with the historical distribution service data corresponding to the preset sample region at the preset historical time, the historical region feature data corresponding to the preset sample region, and the historical distribution status information of the preset sample region within the preset duration from the preset historical time as model training samples, to obtain the distribution status information prediction model includes: and taking the historical distribution service data corresponding to the preset sample region at the preset historical moment, the historical region feature data corresponding to the preset sample region and the historical distribution state information of the preset sample region within the preset time from the preset historical moment as model training samples, and training a multitask model to obtain the distribution state information prediction model, wherein the average distribution time information is used as a main task of the multitask model, and the order quantity information and the average order load information are used as auxiliary tasks of the multitask model.
Optionally, the loss function of the main task is a piecewise function; the piecewise function includes a first sub-function and a second sub-function; a first subsection interval corresponding to the first sub-function is that the actual average distribution time length of the orders in the preset time length from the preset historical moment is smaller than a preset time threshold, and the first sub-function is obtained according to a preset first parameter adjusting coefficient and the actual average distribution time length in the first subsection interval; the second subsection interval corresponding to the second sub-function is that the actual average distribution time length is greater than or equal to the preset time threshold, and the second sub-function is obtained according to a preset second parameter adjusting coefficient and the actual average distribution time length in the second subsection interval; the preset time threshold value represents a critical value of average delivery duration when a preset order abnormal event occurs in a preset area; the first parameter adjustment coefficient and the second parameter adjustment coefficient are different, and the first parameter adjustment coefficient is smaller than the second parameter adjustment coefficient.
Optionally, the area feature data further includes area type feature data, the delivery status information prediction model includes a plurality of delivery status information prediction models, and different area type feature data correspond to different delivery status information prediction models; the obtaining, according to the distribution service data and the area characteristic data, distribution state information of the target area within a preset time period from the preset time through the distribution state information prediction model includes: determining a target prediction model corresponding to the region type feature data from the plurality of delivery state information prediction models; and obtaining the distribution state information of the target area within a preset time from the preset moment through the target prediction model according to the distribution service data and the area characteristic data.
Optionally, the region type feature data includes any one of: second weather data used for representing that the weather information of the corresponding area at the preset moment accords with severe weather conditions; the order load data are used for representing order load information of the corresponding area at the preset moment and meeting high load conditions, and the order load data are used for representing order quantity averagely distributed to each distributor at the preset moment; the second weather data and the order load data; and abnormal area identification data used for representing that the corresponding area is an order abnormal area.
In a second aspect, an apparatus for predicting delivery status information is provided, the apparatus comprising: the system comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is used for acquiring distribution service data of a target area at a preset moment and area characteristic data corresponding to the target area, and the area characteristic data comprises time identification data, first day air data and area identification data for representing the corresponding area; and the prediction module is used for obtaining the distribution state information of the target area within the preset time from the preset moment according to the distribution service data and the area characteristic data through a distribution state information prediction model.
In a third aspect, a computer readable storage medium is provided, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the method according to the first aspect of the disclosure.
In a fourth aspect, an electronic device is provided, comprising: a memory having a computer program stored thereon; a processor for executing the computer program in the memory to implement the steps of the method of the first aspect of the disclosure.
According to the technical scheme, distribution service data of a target area at a preset time and area characteristic data corresponding to the target area are obtained, wherein the area characteristic data comprise time identification data, first day air data and area identification data used for representing the corresponding area; according to the distribution service data and the area characteristic data, distribution state information of the target area in a preset time length from the preset time is obtained through a distribution state information prediction model, namely, according to the distribution service data of the target area, the area identification data of the target area and the first day data, the distribution state information of the target area is predicted according to the area characteristic data such as the time identification data, and therefore the area characteristic data of the target area can be learned, and accuracy of predicting the distribution state information of the target area can be improved.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow diagram illustrating a method of predicting delivery status information in accordance with an exemplary embodiment;
FIG. 2 is a block diagram illustrating a delivery status information prediction model according to an exemplary embodiment;
FIG. 3 is a flow diagram illustrating a delivery status information prediction model training method in accordance with an exemplary embodiment;
FIG. 4 is a block diagram illustrating an apparatus for predicting delivery status information in accordance with an exemplary embodiment;
FIG. 5 is a block diagram illustrating an electronic device in accordance with an example embodiment.
Detailed Description
The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
The present disclosure is directed to a method for predicting delivery status information such as an average delivery duration, an order amount, an average order load, etc. of a specific target area (the target area may be any one of an extreme area such as a high load area, a severe weather area, etc. or a non-extreme area) in a future period of time, and the method is used as a main reference for measuring a situation of transport capacity tension of the target area.
In order to solve the existing problems, the present disclosure provides a method, an apparatus, a storage medium, and an electronic device for predicting delivery status information of a target area, wherein in the process of predicting the delivery status information of the target area, delivery service data of the target area at a preset time and area characteristic data corresponding to the target area may be obtained first, and then the delivery status information of the target area within a preset time period from the preset time is obtained through a delivery status information prediction model according to the delivery service data and the area characteristic data, wherein the area characteristic data includes time identification data, first day data, and area identification data for characterizing the corresponding area, and the delivery status information prediction model is obtained by training according to area characteristic data of a plurality of preset sample areas and the delivery service data, so that the delivery status information prediction model can perform personalized learning on the area characteristic data of the target area, therefore, the accuracy of predicting the distribution state information of the target area can be improved, and in addition, the target area can be any one of an extreme area or a non-extreme area such as a high-load area, a severe weather area and the like.
FIG. 1 is a flow chart illustrating a method of predicting delivery status information, as shown in FIG. 1, according to an exemplary embodiment, the method comprising the steps of:
in step 101, distribution service data of a target area at a preset time and area feature data corresponding to the target area are obtained, where the area feature data includes time identification data, first day data, and area identification data used for representing the corresponding area.
The target area may be any one of a plurality of areas in which the delivery status information needs to be predicted, the target area may be any one of an extreme area or a non-extreme area, such as a high load area or a severe weather area, the preset time may be a current time, and the delivery service data may include one or more of the following data items: the data of the regional order quantity of the target region, the regional order load data of the target region, the regional distributor data of the target region, the order service duration data of the target region within a preset historical duration taking the preset time as an end time, and the historical order statistics data of the target region within a preset historical time period, further, the regional order quantity data can be the order quantity generated by the target region within a period of time, the order quantity which is not taken at the current time, the order quantity which is taken but not delivered at the current time, the cumulative order completion quantity on the day, the load data can be understood as the order quantity averagely distributed to each distributor, the regional order load data can comprise the regional load of the target region, the average load of the target region within a period of time, and the like, the regional distributor data can comprise the current on-duty rider quantity of the target region, the total order quantity of the target region, The total number of riders in the target area, and the like.
The order service duration data is described below by way of example, assuming that the preset time is a current time, the preset historical duration is ten minutes in the past, and the current time is an end time, and the order service duration data may include an average actual store-to duration (arrival time minus order-taking time) of ten minutes past taken orders in the target area, an average actual waiting duration (meal-taking time minus arrival time) of ten minutes past taken orders in the target area, an average actual delivery duration (actual delivery time minus meal-taking time) of ten minutes past delivered orders in the target area, an average actual delivery duration (actual delivery time minus order-taking time) of ten minutes past delivered orders in the target area, an average expected delivery duration (expected delivery time minus order-taking time) of ten minutes past delivered orders in the target area, and a total time of ten minutes in the target area, The target area has taken the average actual order taking duration of the order (order taking time minus order placing time) of ten minutes in the past, and the like.
Assuming that the preset historical time period is the past 24 hours with the current time as the end time, the historical order statistics may include the total order volume of the target area in the past 24 hours, the average person load volume of the target area in the past 24 hours, the average meal taking time, the average delivery time, and the like of the order of the target area in the past 24 hours.
The above description of the distribution service data is only an example, and the present disclosure does not limit the distribution service data.
In addition, the time identification data is used to represent the time period of the preset time in the preset time periods, in a possible implementation manner, 24 hours of a day may be divided into the time periods, for example, one time period may be divided every 10 minutes, specifically, 0:00-0:10 (representing zero to zero tenths) is a first time period, 0:10-0:20 is a second time period, and so on, 24 hours of the day may be divided into 144 time periods, so that the time identification data may be identification information of the time period in which the current time is located, for example, the current time is 0 to 15 minutes, and the time period in which the time is located may be determined to be the second time period 0:10-0:20, which is merely an example and is not limited by this disclosure.
The first weather data is used to characterize weather information of the target area at the preset time, for example, the first weather data may include weather data such as temperature, wind speed, rainfall and the like of the target area at the preset time, and the area identification data may include identification information such as a city ID and an area ID.
In step 102, the delivery status information of the target area within a preset duration from the preset time is obtained through a delivery status information prediction model according to the delivery service data and the area characteristic data.
The preset duration may be a duration preset by a user according to needs of the user in an actual prediction scenario, for example, the preset duration may be 20 minutes in the future with the current time as an initial time, one hour in the future with the current time as an initial time, and the like, the delivery status information may include average delivery duration information, order quantity information, and average order load information, and in a general case, machine learning models (such as the delivery status information prediction model in the present disclosure) are all single-task learning, for example, only one of three data, namely, average delivery duration, order quantity, and average order load is predicted, but considering that tasks are also correlated with one another, the correlated multi-task learning can achieve a better generalization effect than the single-task learning The order quantity and the average order load are in a strong correlation relationship, and based on the correlation relationship, the present disclosure may predict multiple information (which may be understood as multitask learning) such as the average delivery time length, the order quantity, and the average order load of the target area within a preset time length through one delivery state information prediction model, so as to improve the prediction accuracy of the model, where the average order load information is used to represent the order quantity evenly distributed to each deliverer within the preset time length.
The delivery status information prediction model may include a deep fm model, and the following describes a specific implementation of this step with the delivery status information prediction model as the deep fm model:
for example, fig. 2 is a schematic structural diagram of a deep fm model, as shown in fig. 2, the deep fm model includes a feature data input layer, a word embedding layer, a deep fm layer, and an output layer, in this step, the time identification data, the first day air data, and the region identification data of the target region may be input into the word embedding layer of the deep fm model for embedding characterization learning, so as to convert the region feature data of the target region into a word embedding vector, and then the word embedding vector is input into the deep fm layer, and the distribution service data belongs to continuous feature data, and without word embedding processing, the distribution service data may be directly input into the deep fm layer for feature learning, so that the two data, i.e., the region feature data and the distribution service data of the target region, may be simultaneously learned through the deep fm layer, and thus the average distribution duration of the target region within a preset duration, may be output through the output layer, The above examples are merely illustrative, and the present disclosure is not limited thereto.
After step 102 is executed, the delivery status information such as the average delivery duration, the order quantity, and the average order load of the target area within the preset duration may be obtained through the delivery status information prediction model based on the delivery service data and the area characteristic data of the target area, so that whether an order explosion occurs in the target area within the preset duration may be determined based on the delivery status information, so that when the order explosion is predicted to occur, effective measures may be taken in time to relieve the delivery pressure of the target area.
In another possible implementation manner of the present disclosure, the region characteristic data may further include region type characteristic data, the delivery status information prediction model may include a plurality of delivery status information prediction models, different ones of the area type characteristic data corresponding to different ones of the delivery status information prediction models, such that, in the process of obtaining the distribution state information of the target area within the preset time from the preset time through the distribution state information prediction model according to the distribution service data and the area characteristic data, a target prediction model corresponding to the area type feature data may be determined from among the plurality of delivery status information prediction models, and then obtaining the distribution state information of the target area within a preset time from the preset time according to the distribution service data and the area characteristic data through the target prediction model.
Wherein the region type feature data may include any one of: second weather data used for representing that the weather information of the corresponding area at the preset moment accords with severe weather conditions; the order load data are used for representing order load information of the corresponding area at the preset moment and meeting high load conditions, wherein the order load data are used for representing order quantity averagely distributed to each distributor at the preset moment; the second weather data and the order load data; and abnormal area identification data used for representing that the corresponding area is an order abnormal area.
For example, the second weather data may be identification information of severe weather (for example, 1 may be used to represent severe weather, and 0 may be used to represent non-severe weather), in one possible implementation, a plurality of weather thresholds meeting severe weather conditions may be preset for different weather types, specifically, a temperature threshold may be set for high-temperature severe weather, and if the temperature of the target area in the first weather data at the current time is greater than or equal to the temperature threshold, the target area may be determined as a high-temperature severe weather area; a rainfall threshold may be set for heavy rainfall severe weather, and if the rainfall of the target area in the first weather data at the current time is greater than or equal to the rainfall threshold, the target area may be determined to be a heavy rainfall severe weather area; a wind speed threshold may be set for severe stormy weather, and if the wind speed of the target area in the first weather data at the current time is greater than or equal to the wind speed threshold, the target area may be determined as a severe stormy weather area, it needs to be noted that when it is determined that at least one type of weather data (at least one of multiple weather data such as temperature, rainfall, wind speed, and the like) in the first weather data satisfies the corresponding threshold condition, the current weather of the target area may be determined as severe weather, and at this time, the area type of the target area may be determined as a severe weather area.
In addition, a load threshold meeting a high load condition may be preset, so that when the acquired order load data of the target area at the current time is greater than or equal to the load threshold, the area type of the target area may be determined as a high load area.
In a possible application scenario, if the target area is determined to be a severe weather area according to the first day data of the target area, and the target area is determined to be a high load area according to the order load data, at this time, the area type of the target area may be determined to be the severe weather area and the high load area.
In another possible application scenario, there are usually some long-tailed areas with abnormal orders due to unknown reasons, in a possible implementation manner, in order to improve the accuracy of estimation of the distribution state information of the long-tailed areas, a distribution state information prediction model for the long-tailed areas may be pre-established according to the distribution service data and the area characteristic data of a plurality of long-tailed areas, and area identification information may be set for the areas of this type in advance, so that, when the acquired area characteristic data includes the abnormal area identification data, the area type of the target area may be determined as an abnormal order area (also referred to as a long-tailed area), and thus, the distribution state information prediction model corresponding to the abnormal order area may predict and obtain the distribution state information of the target area.
In the practical application scenario of the present disclosure, other area types other than the four area types may also be included, which is not limited by the present disclosure, so that when the delivery status information of the target area is predicted, the area type corresponding to the target area may be determined according to the area type feature data, and then a model corresponding to the area type of the target area is determined as a target prediction model from a plurality of preset delivery status information prediction models, so that the delivery status information of the target area within a preset time period from the preset time may be obtained through the target prediction model according to the delivery service data and the area feature data, the specific prediction method has been described in the example shown in fig. 2, and is not described herein again.
It should be further noted that, if it is determined that the area type of the target area does not belong to any of the area types according to the area type feature data of the target area, the area type of the target area may be determined as a common area type, and then an object prediction model corresponding to the common area type is obtained, so as to predict the delivery status information of the target area through the object prediction model.
By adopting the method, the distribution state information of the target area can be predicted according to the distribution service data of the target area, the area identification data of the target area, the first day data, the time identification data and other area characteristic data, so that the area characteristic data of the target area can be learned, the accuracy of predicting the distribution state information of the target area can be improved, and the target area can be any one of an extreme area or a non-extreme area such as a high-load area, a severe weather area and the like.
In order to ensure the effectiveness and accuracy of model prediction, model training and updating are performed on the delivery status information prediction model according to a preset period (e.g. every day), a training process of the delivery status information prediction model is described below, fig. 3 is a flowchart of a training method of the delivery status information prediction model according to an exemplary embodiment, and as shown in fig. 3, the method includes the following steps:
in step 301, historical distribution service data of a preset sample region at a preset historical time, historical region feature data corresponding to the preset sample region, and historical distribution status information of the preset sample region within the preset duration from the preset historical time are obtained.
Wherein the preset sample region may include a plurality of sample regions, and the preset historical time may include a plurality of times, for example, after dividing 24 hours of a day into a plurality of time periods every ten minutes, the preset historical time may be any time within each of the time periods, and in addition, when the delivery status information prediction model is trained, the preset time period is equal to the preset time period in step 102 of embodiment 1 shown in fig. 1, so that the trained delivery status information prediction model can be used to predict the delivery status information of the target area within the preset time period from the preset time, the historical delivery service data, the historical regional characteristic data, and the historical delivery status information may be selected from data within a predetermined historical time period prior to the model training date, for example, the preset historical period of time may be the day before the model training date to the two weeks before the model training date.
In step 302, the historical distribution service data corresponding to the preset sample region at the preset historical time, the historical region feature data corresponding to the preset sample region, and the historical distribution status information of the preset sample region within the preset duration from the preset historical time are used as model training samples, and a multitask model is trained to obtain the distribution status information prediction model.
In an actual training scenario, a training set and a test set of model training need to be constructed first, and in a possible implementation manner, data in a first preset history sub-time period may be selected as a training sample, data in a second preset history sub-time period may be selected as a test sample, and the preset history time period includes the first preset history sub-time period and the second preset history sub-time period, for example, if a training date of the model is 6 month 15, the preset history time period is 6 month 1 # to 6 month 14 #, the training sample may be data acquired in 6 month 1 # to 6 month 13 #, the test sample may be data acquired in 6 month 14 #, which is just described by way of example, and this disclosure is not limited thereto, and in addition, in the training sample or the test sample, each training sample includes two parts, namely a feature part and a label part, where, the characteristics are the historical distribution service data and the historical regional characteristic data, and the label is the historical distribution status information, for example, the average distribution time length of all orders within the preset time length from the preset historical time, the order quantity within the preset time length from the preset historical time, and the order quantity (i.e., the average order load) equally distributed to each distributor within the preset time length are the labels.
The average distribution time length in each sample can be calculated by formula (1), and the average order load in each sample can be calculated by formula (2):
wherein,indicates the average delivery duration, Finish _ durationnIndicating the delivery duration of the nth order within a preset duration,indicates the average order loadtThe order load at the T-th historical time within the preset time length (namely, the order quantity averagely distributed to each distributor at the T time) is represented, N represents the total quantity of orders within the preset time length, and T represents the preset time length.
It should be noted that, when the training sample is selected, for a sample whose historical delivery status information belongs to abnormal data (for example, for a sample whose area average delivery duration is greater than 2 hours belongs to an extreme abnormal condition), the sample may be removed from the training sample, so as to ensure the prediction accuracy of the model obtained by training.
In addition, in the process of model training, a network structure of a prediction model may be constructed first, as shown in fig. 2, assuming that the delivery status information prediction model is the DeepFM model, the network structure may be constructed according to the structure shown in fig. 2, and a construction manner of a specific model structure may refer to descriptions in relevant documents, which is not described herein again.
After the model structure is constructed, the model structure may be initialized according to an empirical value, for example, the word embedding vector length of the word embedding layer may be set to 64, the DNN part of the deep FM model may be set to 3 fully-connected layers, and dropouts are respectively set for the first-order term and the second-order term of the FM part and each layer of the DNN, where the dropout coefficient of the FM part may be set to 0.85, the dropout coefficient of the DNN part may be set to 0.9, the initial learning rate of model training may be set to 0.015, and the learning rate may be attenuated once per 2000 training steps, and the attenuation coefficient may be set to 0.95.
In addition, since the prediction method provided by the present disclosure can perform prediction learning on multiple tasks such as the average delivery time length, the order quantity, the average order load, and the like of the target area within the preset time length from the preset time, therefore, in this step, the multitask model also needs to be trained, and considering that in an actual application scenario, the average distribution time length within the preset time length is mainly used as a main basis for measuring whether an area is exploded, therefore, in the present disclosure, the average delivery duration information may be used as the main task of the multitask model, the order quantity information and the average order load information are used as auxiliary tasks of the multitask model, model training is carried out, and, the loss function of the main task can adopt a piecewise function to carry out model optimization so as to solve the problem of low prediction accuracy of extreme regions such as high-load regions and severe weather regions.
The piecewise function may include a first sub-function and a second sub-function, where a first piecewise interval corresponding to the first sub-function is that an actual average distribution time length of the order within the preset time length from the preset historical time is less than a preset time threshold, and the first sub-function is obtained according to a preset first parameter adjustment coefficient and the actual average distribution time length located in the first piecewise interval; the second subsection interval corresponding to the second sub-function is that the actual average distribution time length is greater than or equal to the preset time threshold, and the second sub-function is obtained according to a preset second parameter adjusting coefficient and the actual average distribution time length in the second subsection interval; the preset time threshold value represents a critical value of average delivery duration when a preset order abnormal event occurs in a preset area; the first parameter adjustment coefficient and the second parameter adjustment coefficient are different, and the first parameter adjustment coefficient is smaller than the second parameter adjustment coefficient.
Illustratively, in one possible implementation, the piecewise function may be expressed as the following equation (3):
wherein, ylabelRepresents the actual average delivery duration, and T represents the preset time threshold, namely the preset time thresholdCritical value of the average distribution duration when a region is singled out (e.g., T may be equal to 2400 seconds), ylabel< T denotes the first segmentation interval, ylabelT or more represents the second segment interval, LmainValue of loss function, alpha, representing the main tasklowRepresenting the first parameter adjustment coefficient, alphahighRepresents the second parameter adjustment factor and is due to the actual average delivery time period being less than the predetermined time threshold (i.e., y)label< T), a scenario in which no flare usually occurs, and conversely, when the actual average delivery duration is greater than or equal to the preset time threshold (i.e., y)labelT) is generally more likely to occur, and therefore, to ensure the accuracy of the prediction of the mean delivery duration for a single burst scenario win, α is generally setlowLess than alphahighAnd Ω (W) represents the regularization coefficient of the main task loss function.
In addition, the loss function for the auxiliary task may be a squared loss function as shown in equation (4):
Laux＝||ypred-ylabel||2 (4)
wherein L isauxA loss function value representing the auxiliary task, when the auxiliary task is predicted order amount information, ylabelRepresenting the actual order quantity, ypredRepresenting the amount of orders predicted to be output by the model; when the auxiliary task is predicting average order load information, ylabelRepresenting actual average order load, ypredThe average order load of the model prediction output is represented.
After the loss function is constructed, a multitask model can be trained according to the historical distribution service data, the historical region feature data and the historical distribution state information, and in a possible implementation mode, model training can be performed in an alternating training mode.
For example, a random number in a range of 0 to 1 may be generated first, and then the random number is used as a basis for a training task of a current training round, a training task selection rule is shown in table 1, as can be seen from the table, different training probabilities may be set for a main task and an auxiliary task, for example, the main task may be set to 0.6, and the auxiliary task may be set to 0.2, so that the main task may be sufficiently trained under a limited number of training times, and thus an explosion warning may be issued more accurately, for example, a random number interval corresponding to a predicted average delivery duration of the main task may be set to [0,0.6], a random number interval corresponding to a predicted average order load of the auxiliary task may be set to (0.6,0.8], a random number interval corresponding to a predicted order amount of the auxiliary task may be set to (0.8,1], thus, when the generated random number is located in the [0,0.6] interval, the training task of the current training round is the predicted average distribution time length; the training task of the current training round is the predicted average order load when the generated random number is located in the (0.6, 0.8) interval, and the training task of the current training round is the predicted order amount when the generated random number is located in the (0.8, 1) interval.
| Training task | Training probabilities |
| Predicting average delivery duration | 0.6 |
| Predicting average order load | 0.2 |
| Predicting order volume | 0.2 |
TABLE 1
By adopting the model training method, a prediction model capable of multi-task learning is obtained through training, and the loss function of the main task of the model is customized as a piecewise function, so that the problem of low prediction accuracy of extreme regions such as high-load regions and severe weather regions can be solved.
Fig. 4 is a block diagram illustrating an apparatus for predicting delivery status information according to an exemplary embodiment, as shown in fig. 4, the apparatus including:
an obtaining module 401, configured to obtain distribution service data of a target area at a preset time and area feature data corresponding to the target area, where the area feature data includes time identification data, first day air data, and area identification data used for representing the corresponding area;
a predicting module 402, configured to obtain, according to the distribution service data and the area feature data, distribution state information of the target area within a preset time from the preset time through a distribution state information predicting model.
By adopting the device, the distribution state information of the target area can be predicted according to the distribution service data of the target area, the area identification data of the target area, the first day data, the time identification data and other area characteristic data, so that the area characteristic data of the target area can be learned, the accuracy of predicting the distribution state information of the target area can be improved, and the target area can be any one of an extreme area or a non-extreme area such as a high-load area, a severe weather area and the like.
Fig. 5 is a block diagram illustrating an electronic device 500 in accordance with an example embodiment. For example, the electronic device 500 may be provided as a server. Referring to fig. 5, the electronic device 500 comprises a processor 522, which may be one or more in number, and a memory 532 for storing computer programs executable by the processor 522. The computer programs stored in memory 532 may include one or more modules that each correspond to a set of instructions. Further, the processor 522 may be configured to execute the computer program to perform the above-described method of predicting delivery status information.
Additionally, the electronic device 500 may also include a power component 526 and a communication component 550, the power component 526 may be configured to perform power management of the electronic device 500, and the communication component 550 may be configured to enable communication, e.g., wired or wireless communication, of the electronic device 500. In addition, the electronic device 500 may also include input/output (I/O) interfaces 558. The electronic device 500 may operate based on an operating system stored in the memory 532, such as Windows Server, Mac OSXTM, UnixTM, LinuxTM, and the like.
In another exemplary embodiment, a computer readable storage medium comprising program instructions which, when executed by a processor, implement the steps of the above-described method of predicting shipping status information is also provided. For example, the computer readable storage medium may be the memory 532 including program instructions executable by the processor 522 of the electronic device 500 to perform the method of predicting shipping status information described above.
In another exemplary embodiment, a computer program product is also provided, which comprises a computer program executable by a programmable apparatus, the computer program having code portions for performing the above-described method of predicting delivery status information when executed by the programmable apparatus.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.
It should be noted that, in the foregoing embodiments, various features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various combinations that are possible in the present disclosure are not described again.
In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.
Claims (11)
1. A method of predicting delivery status information, the method comprising:
acquiring distribution service data of a target area at a preset time and area characteristic data corresponding to the target area, wherein the area characteristic data comprises time identification data, first day gas data and area identification data for representing the corresponding area;
and obtaining the distribution state information of the target area within a preset time from the preset moment according to the distribution service data and the area characteristic data through a distribution state information prediction model.
2. The method of claim 1, wherein the delivery status information comprises average delivery duration information, order volume information, average order load information; the average order load information is used to characterize the amount of orders evenly distributed to each of the dispatchers over the preset length of time.
3. The method of claim 1, wherein the delivery service data comprises one or more of the following items:
the data of the regional order quantity of the target region, the regional order load data of the target region, the regional distributor data of the target region, the order service duration data of the target region in a preset historical duration taking the preset time as an ending time and the historical order statistical data of the target region in a preset historical time period.
4. The method of claim 2, wherein the delivery status information prediction model is trained by:
acquiring historical distribution service data of a preset sample region at a preset historical moment, historical region characteristic data corresponding to the preset sample region and historical distribution state information of the preset sample region within the preset duration from the preset historical moment;
and training by taking the historical distribution service data corresponding to the preset sample region at the preset historical moment, the historical region feature data corresponding to the preset sample region and the historical distribution state information of the preset sample region within the preset duration from the preset historical moment as model training samples to obtain the distribution state information prediction model.
5. The method of claim 4, wherein the training of the historical distribution service data corresponding to the preset sample region at the preset historical time, the historical region feature data corresponding to the preset sample region, and the historical distribution status information of the preset sample region within the preset duration from the preset historical time as model training samples to obtain the distribution status information prediction model comprises:
and taking the historical distribution service data corresponding to the preset sample region at the preset historical moment, the historical region feature data corresponding to the preset sample region and the historical distribution state information of the preset sample region within the preset time from the preset historical moment as model training samples, and training a multitask model to obtain the distribution state information prediction model, wherein the average distribution time information is used as a main task of the multitask model, and the order quantity information and the average order load information are used as auxiliary tasks of the multitask model.
6. The method of claim 5, wherein the loss function of the primary task is a piecewise function;
the piecewise function includes a first sub-function and a second sub-function;
a first subsection interval corresponding to the first sub-function is that the actual average distribution time length of the orders in the preset time length from the preset historical moment is smaller than a preset time threshold, and the first sub-function is obtained according to a preset first parameter adjusting coefficient and the actual average distribution time length in the first subsection interval;
the second subsection interval corresponding to the second sub-function is that the actual average distribution time length is greater than or equal to the preset time threshold, and the second sub-function is obtained according to a preset second parameter adjusting coefficient and the actual average distribution time length in the second subsection interval;
the preset time threshold value represents a critical value of average delivery duration when a preset order abnormal event occurs in a preset area; the first parameter adjustment coefficient and the second parameter adjustment coefficient are different, and the first parameter adjustment coefficient is smaller than the second parameter adjustment coefficient.
7. The method according to any one of claims 1 to 6, wherein the area characteristic data further includes area type characteristic data, the delivery status information prediction model includes a plurality of delivery status information prediction models, and different ones of the area type characteristic data correspond to different ones of the delivery status information prediction models;
the obtaining, according to the distribution service data and the area characteristic data, distribution state information of the target area within a preset time period from the preset time through the distribution state information prediction model includes:
determining a target prediction model corresponding to the region type feature data from the plurality of delivery state information prediction models;
and obtaining the distribution state information of the target area within a preset time from the preset moment through the target prediction model according to the distribution service data and the area characteristic data.
8. The method according to claim 7, wherein the region type feature data comprises any of:
second weather data used for representing that the weather information of the corresponding area at the preset moment accords with severe weather conditions;
the order load data are used for representing order load information of the corresponding area at the preset moment and meeting high load conditions, and the order load data are used for representing order quantity averagely distributed to each distributor at the preset moment;
the second weather data and the order load data;
and abnormal area identification data used for representing that the corresponding area is an order abnormal area.
9. An apparatus for predicting delivery status information, the apparatus comprising:
the system comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is used for acquiring distribution service data of a target area at a preset moment and area characteristic data corresponding to the target area, and the area characteristic data comprises time identification data, first day air data and area identification data for representing the corresponding area;
and the prediction module is used for obtaining the distribution state information of the target area within the preset time from the preset moment according to the distribution service data and the area characteristic data through a distribution state information prediction model.
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 according to any one of claims 1 to 8.
11. An electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to carry out the steps of the method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910590756.9A CN110378522A (en) | 2019-07-02 | 2019-07-02 | Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910590756.9A CN110378522A (en) | 2019-07-02 | 2019-07-02 | Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110378522A true CN110378522A (en) | 2019-10-25 |
Family
ID=68251537
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910590756.9A Pending CN110378522A (en) | 2019-07-02 | 2019-07-02 | Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110378522A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111047152A (en) * | 2019-11-22 | 2020-04-21 | 口碑(上海)信息技术有限公司 | Work order processing method and device |
| CN111459675A (en) * | 2020-03-31 | 2020-07-28 | 拉扎斯网络科技（上海）有限公司 | Data processing method and device, readable storage medium and electronic equipment |
| CN111815050A (en) * | 2020-07-07 | 2020-10-23 | 北京嘀嘀无限科技发展有限公司 | Model training method, model training device, state prediction method, state prediction device, electronic equipment and storage medium |
| CN112036695A (en) * | 2020-07-28 | 2020-12-04 | 拉扎斯网络科技（上海）有限公司 | Weather information prediction method and device, readable storage medium and electronic equipment |
| CN112036788A (en) * | 2020-07-28 | 2020-12-04 | 拉扎斯网络科技（上海）有限公司 | Data processing method and device, readable storage medium and electronic equipment |
| CN112613809A (en) * | 2020-12-15 | 2021-04-06 | 拉扎斯网络科技（上海）有限公司 | Label data obtaining method and device, electronic equipment and storage medium |
| CN112819394A (en) * | 2019-11-15 | 2021-05-18 | 北京三快在线科技有限公司 | Waybill processing method and device, computer-readable storage medium and electronic equipment |
| CN112819266A (en) * | 2019-11-15 | 2021-05-18 | 北京三快在线科技有限公司 | Distribution parameter adjusting method and device, storage medium and electronic equipment |
| CN112884501A (en) * | 2019-11-29 | 2021-06-01 | 百度在线网络技术（北京）有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN112990521A (en) * | 2019-12-13 | 2021-06-18 | 顺丰科技有限公司 | Distribution time estimation method and device, computer equipment and storage medium |
| CN113077181A (en) * | 2021-04-22 | 2021-07-06 | 上海钧正网络科技有限公司 | Parking station setting method, device, medium and electronic equipment |
| CN113159281A (en) * | 2021-03-25 | 2021-07-23 | 拉扎斯网络科技（上海）有限公司 | Data processing method and data processing device |
| CN113379497A (en) * | 2021-06-11 | 2021-09-10 | 口碑(上海)信息技术有限公司 | Order regulation and control method and device, computer equipment and computer readable storage medium |
| CN113807630A (en) * | 2020-12-23 | 2021-12-17 | 京东科技控股股份有限公司 | Method, device, equipment and storage medium for acquiring requirements of robot service platform |
| CN115222109A (en) * | 2022-06-28 | 2022-10-21 | 浙江口碑网络技术有限公司 | Information processing method, device, medium, and apparatus |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060235557A1 (en) * | 2002-10-11 | 2006-10-19 | Thomas Knight | Associated systems and methods for improving planning, scheduling, and supply chain management |
| US8078485B1 (en) * | 2008-05-29 | 2011-12-13 | Accenture Global Services Limited | Postal, freight, and logistics industry high performance capability assessment |
| CN108491951A (en) * | 2018-01-25 | 2018-09-04 | 北京三快在线科技有限公司 | A kind of prediction technique, device and electronic equipment for taking out distribution time |
| CN108960283A (en) * | 2018-05-30 | 2018-12-07 | 北京市商汤科技开发有限公司 | Classification task incremental processing method and device, electronic equipment and storage medium |
| CN109146345A (en) * | 2017-06-16 | 2019-01-04 | 北京三快在线科技有限公司 | Information-pushing method and device, electronic equipment |
| CN109426885A (en) * | 2017-08-28 | 2019-03-05 | 北京小度信息科技有限公司 | Order allocation method and device |
| CN109685275A (en) * | 2018-12-27 | 2019-04-26 | 拉扎斯网络科技（上海）有限公司 | Distribution team load pressure prediction method and device, electronic equipment and storage medium |
-
2019
- 2019-07-02 CN CN201910590756.9A patent/CN110378522A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060235557A1 (en) * | 2002-10-11 | 2006-10-19 | Thomas Knight | Associated systems and methods for improving planning, scheduling, and supply chain management |
| US8078485B1 (en) * | 2008-05-29 | 2011-12-13 | Accenture Global Services Limited | Postal, freight, and logistics industry high performance capability assessment |
| CN109146345A (en) * | 2017-06-16 | 2019-01-04 | 北京三快在线科技有限公司 | Information-pushing method and device, electronic equipment |
| CN109426885A (en) * | 2017-08-28 | 2019-03-05 | 北京小度信息科技有限公司 | Order allocation method and device |
| CN108491951A (en) * | 2018-01-25 | 2018-09-04 | 北京三快在线科技有限公司 | A kind of prediction technique, device and electronic equipment for taking out distribution time |
| CN108960283A (en) * | 2018-05-30 | 2018-12-07 | 北京市商汤科技开发有限公司 | Classification task incremental processing method and device, electronic equipment and storage medium |
| CN109685275A (en) * | 2018-12-27 | 2019-04-26 | 拉扎斯网络科技（上海）有限公司 | Distribution team load pressure prediction method and device, electronic equipment and storage medium |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112819266B (en) * | 2019-11-15 | 2023-04-18 | 北京三快在线科技有限公司 | Distribution parameter adjusting method and device, storage medium and electronic equipment |
| CN112819394B (en) * | 2019-11-15 | 2023-03-24 | 北京三快在线科技有限公司 | Waybill processing method and device, computer-readable storage medium and electronic equipment |
| CN112819394A (en) * | 2019-11-15 | 2021-05-18 | 北京三快在线科技有限公司 | Waybill processing method and device, computer-readable storage medium and electronic equipment |
| CN112819266A (en) * | 2019-11-15 | 2021-05-18 | 北京三快在线科技有限公司 | Distribution parameter adjusting method and device, storage medium and electronic equipment |
| CN111047152A (en) * | 2019-11-22 | 2020-04-21 | 口碑(上海)信息技术有限公司 | Work order processing method and device |
| CN112884501B (en) * | 2019-11-29 | 2023-10-10 | 百度在线网络技术（北京）有限公司 | Data processing method, device, electronic equipment and storage medium |
| CN112884501A (en) * | 2019-11-29 | 2021-06-01 | 百度在线网络技术（北京）有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN112990521A (en) * | 2019-12-13 | 2021-06-18 | 顺丰科技有限公司 | Distribution time estimation method and device, computer equipment and storage medium |
| CN111459675A (en) * | 2020-03-31 | 2020-07-28 | 拉扎斯网络科技（上海）有限公司 | Data processing method and device, readable storage medium and electronic equipment |
| CN111459675B (en) * | 2020-03-31 | 2023-09-15 | 拉扎斯网络科技（上海）有限公司 | Data processing method and device, readable storage medium and electronic equipment |
| CN111815050B (en) * | 2020-07-07 | 2021-05-28 | 北京嘀嘀无限科技发展有限公司 | Model training method, model training device, state prediction method, state prediction device, electronic equipment and storage medium |
| CN111815050A (en) * | 2020-07-07 | 2020-10-23 | 北京嘀嘀无限科技发展有限公司 | Model training method, model training device, state prediction method, state prediction device, electronic equipment and storage medium |
| CN112036788B (en) * | 2020-07-28 | 2024-05-07 | 拉扎斯网络科技（上海）有限公司 | Data processing method, device, readable storage medium and electronic equipment |
| CN112036788A (en) * | 2020-07-28 | 2020-12-04 | 拉扎斯网络科技（上海）有限公司 | Data processing method and device, readable storage medium and electronic equipment |
| CN112036695A (en) * | 2020-07-28 | 2020-12-04 | 拉扎斯网络科技（上海）有限公司 | Weather information prediction method and device, readable storage medium and electronic equipment |
| CN112613809A (en) * | 2020-12-15 | 2021-04-06 | 拉扎斯网络科技（上海）有限公司 | Label data obtaining method and device, electronic equipment and storage medium |
| CN112613809B (en) * | 2020-12-15 | 2024-03-22 | 拉扎斯网络科技（上海）有限公司 | Tag data obtaining method and device, electronic equipment and storage medium |
| CN113807630A (en) * | 2020-12-23 | 2021-12-17 | 京东科技控股股份有限公司 | Method, device, equipment and storage medium for acquiring requirements of robot service platform |
| CN113807630B (en) * | 2020-12-23 | 2024-03-05 | 京东科技控股股份有限公司 | Method, device, equipment and storage medium for acquiring requirements of robot service platform |
| CN113159281A (en) * | 2021-03-25 | 2021-07-23 | 拉扎斯网络科技（上海）有限公司 | Data processing method and data processing device |
| CN113077181A (en) * | 2021-04-22 | 2021-07-06 | 上海钧正网络科技有限公司 | Parking station setting method, device, medium and electronic equipment |
| CN113379497B (en) * | 2021-06-11 | 2023-12-26 | 口碑(上海)信息技术有限公司 | Order regulation method, order regulation device, computer equipment and computer readable storage medium |
| CN113379497A (en) * | 2021-06-11 | 2021-09-10 | 口碑(上海)信息技术有限公司 | Order regulation and control method and device, computer equipment and computer readable storage medium |
| CN115222109B (en) * | 2022-06-28 | 2023-06-16 | 浙江口碑网络技术有限公司 | Information processing method, device, medium and equipment |
| CN115222109A (en) * | 2022-06-28 | 2022-10-21 | 浙江口碑网络技术有限公司 | Information processing method, device, medium, and apparatus |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110378522A (en) | Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information | |
| CN110888788B (en) | Abnormality detection method, abnormality detection device, computer device, and storage medium | |
| US9235620B2 (en) | Updating cached database query results | |
| AU2013304366B2 (en) | Updating cached database query results | |
| US9508041B2 (en) | Method for predicting user operation and mobile terminal | |
| CN107844848A (en) | A kind of region flow of the people Forecasting Methodology and system | |
| US9552270B2 (en) | System and method for receiving analysis requests and configuring analytics systems | |
| CN110334839A (en) | Flight is delayed prediction technique, device, equipment and storage medium | |
| CN110287332B (en) | Method and device for selecting simulation model in cloud environment | |
| US7716535B2 (en) | Kalman filtering for grid computing telemetry and workload management | |
| JP6740860B2 (en) | Safety stock determination device, method and program | |
| CN109409780B (en) | Change processing method, device, computer equipment and storage medium | |
| CN116415923B (en) | Intelligent gas work order implementation management method and Internet of things system and device | |
| CN117350640A (en) | Project progress management method and system | |
| CN110070392B (en) | User loss early warning method and device | |
| CN109658153B (en) | Spare part demand prediction method and device and electronic equipment | |
| CN112308636B (en) | Market demand value calculation method and device based on market demand change | |
| CN110717620A (en) | Order meal delivery time prediction method and device, electronic equipment and storage medium | |
| CN112560939B (en) | Model verification method and device and computer equipment | |
| CN109219002A (en) | A kind of reminding short message sending method, device, equipment and readable storage medium storing program for executing | |
| CN116911735A (en) | Nuclear power spare part safety stock quantity determination method, device, equipment and medium | |
| CN110413482B (en) | Detection method and device | |
| CN114066079B (en) | Multi-tenant-based network appointment vehicle supply and demand difference prediction method and device | |
| CN110210645A (en) | A kind of development trend data capture method, device and readable storage medium storing program for executing | |
| JP2010020573A (en) | System for managing life cycle of equipment and method therefor |
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 |