Information processing method, device, product, storage medium and equipment for delivery waybill

Abstract

The embodiments of this specification provide a method, apparatus, product, storage medium and equipment for processing information of a delivery waybill, the method comprising: obtaining the waybill information of a delivery waybill that is currently in a delivery state at the current moment; inputting the waybill information at the current moment into a machine learning model, and obtaining the mean value of the remaining delivery time and the variance of the remaining delivery time at the current moment predicted by the machine learning model; wherein the machine learning model is trained based on the normal distribution of the remaining delivery time at the current moment; determining the normal distribution of the remaining delivery time at the current moment based on the predicted mean value of the remaining delivery time and the variance of the remaining time; obtaining the remaining committed user time at the current moment, and determining the timeout probability of the delivery waybill at the current moment using the normal distribution.

Description

Technical Field

[0001] This specification relates to the field of Internet technology, and in particular to methods, devices, products, storage media and equipment for processing information on delivery orders. Background Art

[0002] Delivery services are widely used in online shopping, food delivery, errand shopping and other scenarios. For example, in food delivery and other scenarios, after the user places an order using the client, the server will quickly allocate delivery capacity for the delivery waybill. The delivery capacity will arrive at the store's location, pick up the items to be delivered and deliver them to the delivery location specified by the user. In this scenario, the delivery status of the delivery waybill is of great significance to the business. For example, when the user places an order, the server will estimate the estimated delivery time of the delivery waybill and provide it to the user as the promised user time promised to the user; during the delivery process, the actual delivery time will change dynamically. At the current moment, the server can estimate the remaining delivery time between the actual final delivery time and the current moment, or estimate whether the waybill has timed out. Based on this, accurately estimating whether the delivery waybill has timed out during the delivery process has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] To overcome the problems existing in the related art, the embodiments of this specification provide a method, apparatus, product, storage medium and device for processing information of a delivery waybill.

[0004] According to a first aspect of an embodiment of this specification, a method for processing information of a delivery waybill is provided, the method comprising:

[0005] Get the waybill information of the delivery waybill that is currently in delivery status at the current moment;

[0006] Input the waybill information at the current moment into the machine learning model, and obtain the mean value and variance of the remaining delivery time at the current moment predicted by the machine learning model; wherein the machine learning model is trained based on the normal distribution of the remaining delivery time at the current moment;

[0007] Determine the normal distribution of the remaining delivery time at the current moment according to the predicted mean value and variance of the remaining delivery time at the current moment;

[0008] The remaining committed user time at the current moment is obtained, and the timeout probability of the delivery waybill at the current moment is determined using the normal distribution.

[0009] According to a second aspect of an embodiment of this specification, there is provided an information processing device for a delivery waybill, including:

[0010] The waybill information acquisition module is used to: obtain the waybill information of the delivery waybill currently in the delivery state at the current moment;

[0011] A model prediction module, used to: input the waybill information at the current moment into a machine learning model, and obtain the mean value and variance of the remaining delivery time at the current moment predicted by the machine learning model; wherein the machine learning model is trained based on the normal distribution of the remaining delivery time at the current moment;

[0012] A normal distribution determination module, used to: determine the normal distribution of the remaining delivery time at the current moment according to the predicted mean value and variance of the remaining delivery time at the current moment;

[0013] The timeout determination module is used to obtain the remaining committed user time at the current moment, and determine the timeout probability of the delivery waybill at the current moment using the normal distribution.

[0014] According to a third aspect of the embodiments of this specification, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method embodiment described in the first aspect are implemented.

[0015] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method embodiment described in the first aspect are implemented.

[0016] According to a fifth aspect of the embodiments of this specification, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method embodiment described in the first aspect are implemented.

[0017] The technical solutions provided by the embodiments of this specification may have the following beneficial effects:

[0018] In the embodiments of this specification, it is designed to train the machine learning model with the remaining delivery time at the current moment obeying the normal distribution, and the output of the machine learning model is also designed to be the mean and variance of the remaining delivery time at the current moment. Based on this, the information estimated by these two models can be used to determine the normal distribution curve of the remaining delivery time at the current moment, and the normal distribution is used to determine the timeout probability of the delivery waybill at the current moment in combination with the remaining committed user time at the current moment. In this embodiment, by designing that the remaining delivery time at the current moment obeys the normal distribution, the two pieces of information are associated, and a model is used to simultaneously estimate the two related pieces of information, so the difficulty of training and maintaining the model is significantly reduced, and the prediction accuracy of the model is also guaranteed. In addition, based on the two pieces of information output by the model, it is designed to use the normal distribution to determine the timeout probability. Due to the accurate output of the model, an accurate timeout probability can be obtained.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the specification and, together with the description, serve to explain the principles of the specification.

[0021] Figure 1A It is a schematic diagram of a delivery scenario shown in this specification according to an exemplary embodiment.

[0022] Figure 1B It is a schematic diagram of a time axis according to an exemplary embodiment of the present specification.

[0023] Figure 2A It is a flowchart of a method for processing information of a delivery waybill according to an exemplary embodiment of the present specification.

[0024] Figure 2B It is a schematic diagram of a normal distribution of the remaining delivery time at the current moment according to an exemplary embodiment of the present specification.

[0025] Figure 3 It is a hardware structure diagram of a computer device where an information processing device for a delivery waybill is located according to an exemplary embodiment of this specification.

[0026] Figure 4 It is a block diagram of an information processing device for a delivery waybill according to an exemplary embodiment of the present specification. DETAILED DESCRIPTION

[0027] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this specification. Instead, they are merely examples of devices and methods consistent with some aspects of this specification as detailed in the appended claims.

[0028] The terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit this specification. The singular forms "a", "the" and "the" used in this specification and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0029] It should be understood that although the terms first, second, third, etc. may be used in this specification to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this specification, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0030] Delivery services are widely used in online shopping, food delivery, errand shopping, etc. Figure 1A Shown is a schematic diagram of a delivery scenario shown in this specification according to an exemplary embodiment, including a business party, delivery capacity and users. Among them, the service party has built a service end and provides a client to the user, and the user can use the services provided by the service party through the client; different from the client for users, the service party also provides a client for the store party to use, and the store party can use the services provided by the service party through the client. Among them, each user can trade with the store party through the client, and each transaction order corresponds to a delivery waybill. The service party can allocate delivery capacity for the delivery waybill. The delivery capacity of this embodiment refers to a party with delivery capabilities, including but not limited to delivery personnel, such as so-called riders. The delivery capacity can communicate with the service end through the client used for the delivery capacity; in other examples, the delivery capacity can also include unmanned delivery equipment, such as drones, unmanned vehicles, etc.

[0031] In some scenarios, users are more concerned about the status of delivery services. For example, in the delivery services of food delivery, users usually use the client to initiate a transaction order, and the server generates a corresponding delivery waybill for the transaction order. The delivery waybill involves the delivery starting point and delivery destination specified by the user. The server dispatches the delivery capacity according to the waybill information, and the delivery capacity delivers the items required by the user from the delivery starting point to the delivery destination. In some scenarios with high requirements for delivery timeliness, such as instant delivery services, users tend to pay more attention to the delivery status of the waybill after initiating the waybill.

[0032] like Figure 1B FIG. 1 is a schematic diagram of a time axis according to an exemplary embodiment of the present specification, combined with Figure 1B The delivery status of a delivery order involves at least the following four concepts related to time:

[0033] (1) The promised delivery time for the user, that is, the time promised to the user when the user places the order; for example, when a delivery waybill is generated, the server will estimate the delivery time for the delivery waybill to reach the delivery destination specified by the user based on multiple factors such as the delivery starting point and delivery destination of the delivery waybill, the location of the delivery capacity, and the number of waybills delivered by the delivery capacity. For example, Figure 1B In the example, a delivery order is generated at time A, and the server estimates that the promised delivery time of the delivery order is N; time N minus time A is the promised user duration.

[0034] (2) Actual delivery time, that is, the time from the time the user places an order to the time the food is actually delivered to the designated delivery destination. The delivery of the food to the designated delivery destination indicates that the delivery order has been completed. In actual applications, a delivery order may face many variable factors during the delivery process, such as weather, road congestion, extended meal delivery time or the impact of other delivery orders, etc., which may cause the actual delivery time to be different from the promised user time, which may be longer or shorter than the promised user time. When the actual delivery time is longer than the promised user time, it is considered a timeout. For example, Figure 1B Taking the example of the actual delivery time being greater than the promised delivery time, the actual delivery time is M, and time M minus time A is the actual delivery time.

[0035] (3) The remaining delivery time at the current time, that is, the time between the actual final delivery time and the current time from the generation of a delivery order to the final completion of delivery. The current time is any time between time A and time M, for example, Figure 1B In the equation, time O is the current time, the current time is time O, and the remaining delivery time at time O is M minus A.

[0036] (4) The remaining committed user time at the current moment, that is, the time between the promised delivery time and the current moment from the time a delivery order is generated to the time it is finally delivered. For example, Figure 1B In the equation, time O is the current time, the current time is time O, and the remaining delivery time at time O is N minus A.

[0037] During the process of unfinished delivery of a waybill, the actual delivery time is unknown, so the remaining delivery time at the current moment is also unknown. However, in actual business, there is a need to accurately estimate the remaining delivery time at the current moment during the process of unfinished delivery of a waybill. For example, in an instant delivery scenario, some users will pay more attention to the delivery status of the delivery waybill. When the user checks the client, the delivery status page provided by the client displays the real-time location information of the delivery capacity at the current moment, and can also estimate and display the remaining delivery time at the current moment based on a variety of actual factors. Optionally, the timeout probability of the waybill can also be estimated based on the remaining delivery time at the current moment. When the user checks the client at different times, the remaining delivery time and timeout probability displayed by the client may be different.

[0038] On the other hand, the server can also continuously estimate the remaining delivery time and timeout probability at multiple times during the delivery process of the delivery waybill, and then determine whether to execute other business processing.

[0039] For example, if the remaining delivery time of the delivery order is short and the timeout probability is low, it means that the delivery order can be completed ahead of schedule within the promised user time, and other delivery orders can be added to the delivery capacity based on this; if the remaining delivery time of the delivery order is long and the timeout probability is high, it means that the delivery order cannot be completed within the promised user time, and other delivery orders of the delivery capacity can be adjusted based on this. Alternatively, the status of the delivery capacity can be determined in a timely manner, and the reason for the timeout of the delivery capacity can be determined to ensure the delivery safety of the delivery capacity, etc.; or, reminders can be given to the users of the delivery order to relieve their anxiety and improve the user experience, etc.

[0040] Therefore, during the delivery process of the delivery waybill, whether the delivery waybill is timed out involves two important information: the remaining delivery time at the current moment and the timeout probability. It is of great significance to accurately estimate the remaining delivery time and the timeout probability at the current moment.

[0041] Since it involves two pieces of information: the remaining delivery time at the current moment and the timeout probability, some processing ideas are to use machine learning models to estimate the remaining delivery time at the current moment, and then divide the remaining delivery time at the current moment by the remaining committed user time, and use the value as the timeout probability; however, in this solution, the estimation of the timeout probability only considers two pieces of information: the remaining delivery time at the current moment and the remaining committed user time, and the timeout probability is affected by many factors faced in the delivery process, so the estimated timeout probability is not accurate enough.

[0042] Another processing idea is to use two machine learning models to estimate the remaining delivery time and the timeout probability at the current moment respectively; however, this solution requires maintaining two models, and the remaining delivery time at the current moment and the timeout probability at the current moment are related to each other, which requires that the training data of the two models be completely consistent. If the training data of the two models are inconsistent, the output results of the two models may be inconsistent.

[0043] From the above analysis, it can be seen that the timeout prediction of delivery waybills involves multiple pieces of information, including the actual delivery time, the remaining delivery time at the current moment, and the remaining promised user time at the current moment. It is a technical problem that needs to be solved urgently, such as accurately making timeout predictions and facilitating model maintenance.

[0044] Based on this, the embodiment of this specification provides a method for estimating the timeout of a delivery order, such as Figure 2A As shown, Figure 2A This is a flowchart of a method for processing information of a delivery waybill according to an exemplary embodiment of the present specification, comprising the following steps:

[0045] In step 202, the waybill information of the delivery waybill currently in the delivery state at the current moment is obtained;

[0046] In step 204, the waybill information at the current moment is input into the machine learning model to obtain the mean value and variance of the remaining delivery time at the current moment predicted by the machine learning model;

[0047] Among them, the machine learning model is trained based on the normal distribution of the remaining delivery time at the current moment.

[0048] In step 206, the normal distribution of the remaining delivery time at the current moment is determined according to the predicted mean value and variance of the remaining delivery time at the current moment;

[0049] In step 208, the remaining committed user time at the current moment is obtained, and the normal distribution is used to determine the timeout probability of the delivery waybill at the current moment.

[0050] This embodiment involves a machine learning model. The training process of the machine learning model can be: first, a model is represented by modeling, then the model is evaluated by constructing an evaluation function, and finally, the evaluation function is optimized according to sample data and optimization methods to adjust the model to meet the set conditions.

[0051] As described above, in the field of machine learning, there are many steps involved from the modeling to the training stage, such as the selection and processing of sample data, the design of data features, the design of models, the design of loss functions or optimization methods, etc. Subtle differences in any step are factors that lead to subtle defects in prediction accuracy.

[0052] In this embodiment, the designed machine learning model is trained based on the remaining delivery time at the current moment following a normal distribution.

[0053] If the random variable x follows a normal distribution with a mathematical expectation of μ and a variance of σ 2 is denoted as N(μ, σ 2 ). That is to say, the probability density function y is a normal distribution. Its expected value μ determines the position of the distribution, and its variance σ determines the amplitude of the distribution. When μ = 0 and σ = 1, the normal distribution is the standard normal distribution.

[0054] The expected value μ is also the mean value, which is the center of the normal distribution curve and determines the position of the curve peak. The width of the normal distribution curve is determined by the variance σ. The variance σ represents the variability of the normal distribution. Changes in the variance cause the curve to become narrower or wider and have an inverse effect on the height of the curve. Therefore, once the mean and variance are determined, the corresponding normal distribution curve can be determined.

[0055] The probability of the random variable x in a certain interval, such as the interval (a, b), that is, the probability of a < X < b, is the area under the probability density curve in this interval. Mathematically, it is the integral of the probability density function over the interval (a, b). Therefore, the size of the probability is the size of "the area under the probability density function curve".

[0056] Exemplarily, the normal distribution can be expressed by the formula:

[0057] In this embodiment, it is designed to assume that the remaining delivery time at the current moment follows a normal distribution, that is, the remaining delivery time at the current moment is used as the random variable value x of the normal distribution curve. If the mean and variance of the remaining delivery time at the current moment can be estimated, then the probability of x in different intervals can be calculated according to the normal distribution formula and integration. Based on this, the output of the machine learning model designed in this embodiment is two pieces of information: the mean of the remaining delivery time at the current moment and the variance of the remaining delivery time at the current moment. Furthermore, the normal distribution of the overtime probability of the remaining delivery time can be determined, and then the probability of the remaining delivery time in different intervals can be calculated according to the normal distribution formula and integration.

[0058] Among them, the training data set of the machine learning model can be flexibly determined as needed; illustratively, the training data can be obtained using the data of historical delivery waybills of multiple delivery transport capacities. For example, with the user's authorization and consent, the server can obtain the data sent by the client used by the delivery transport capacity, and the client can continuously collect the delivery waybill information of the delivery waybill according to the set period. The delivery waybill information of the delivery waybill may include the geographical location information of the delivery transport capacity at the time of collection, traffic information at the location, the number of delivery waybills undertaken by the delivery transport capacity, the collection status information of the delivery items of each delivery waybill undertaken by the delivery transport capacity (for example, whether the items have been collected, etc.), the attribute information of the collected items of each delivery waybill undertaken by the delivery transport capacity (for example, the size, weight or type of the items, etc.), or the historical delivery information of the delivery transport capacity (for example, the historical delivery efficiency of the delivery transport capacity, the number of timeout information, the timeout ratio information or the delivery speed information, etc.).

[0059] For example, for the data of a historical delivery waybill, a time interval can be set to determine multiple samples. The time interval can be flexibly determined according to needs, such as a custom time of one minute, two minutes, or five minutes. Therefore, for each historical delivery waybill data, multiple samples can be obtained. Each sample corresponds to a historical moment and the remaining delivery time at the historical moment. The remaining delivery time at the historical moment can be used as the label value of the sample.

[0060] Another important aspect of the training process is to design an appropriate evaluation function based on business needs. In the scenario of supervised model training, the sample data is labeled with true values, and the evaluation function is used to estimate the error between the model's predicted value and the true value. The evaluation function is crucial to the recognition accuracy of the model, and it is difficult to design an evaluation function based on the existing sample data and model requirements.

[0061] When solving the model parameters, the maximum likelihood estimation method is often used to solve, that is, to find a set of parameters under which the likelihood (probability) of the data is maximized. This embodiment designs that the remaining delivery time at the current moment obeys the normal distribution. Based on the normal distribution, it can be determined that its maximum likelihood estimation function is obtained, as an example:

[0062]

[0063] Therefore, this embodiment designs the evaluation function of the model based on this, as an example:

[0064]

[0065] Among them, pred_y is the mean of the remaining delivery time at the current moment estimated by the model, and pred_std is the variance of the remaining delivery time at the current moment estimated by the model; that is, the evaluation function simultaneously estimates the mean and variance of the normal distribution, and label is the label value, whose true value is the remaining delivery time at the current moment in the sample.

[0066] That is, the optimization target of the evaluation function of the model in this embodiment is two parameters of the normal distribution, namely, the mean of the remaining delivery time at the current moment and the variance of the remaining delivery time at the current moment.

[0067] The specific formula of the above evaluation function is only for illustration. In practical applications, the specific mathematical description of the function can be flexibly configured as needed, and whether to add a regularization term can be determined as needed. This embodiment does not limit this.

[0068] like Figure 2B , which is a schematic diagram of a normal distribution of the remaining delivery time at the current moment according to an exemplary embodiment of the present specification. Figure 2B In the example, the mean of the remaining delivery time at the current moment is 10 minutes (minutes), and the variance of the remaining delivery time at the current moment is 2 minutes. The remaining user commitment time of the delivery order at the current moment is 16 minutes. In the estimated normal distribution, the true value label appears in different intervals with different probabilities in the normal distribution. When the label is larger than the remaining committed user time, the delivery order has timed out. For example, Figure 2B Taking the current remaining committed user time of 16 minutes as an example, the time after x=16 represents the delivery order timeout time. Therefore, the shaded area after the dotted line where x=16 is located can be integrated to obtain the timeout probability of the delivery order under the currently estimated normal distribution; for example, the integral formula of the normal distribution is as follows:

[0069]

[0070] Where P represents the timeout probability of x.

[0071] Exemplarily, the machine learning model of this embodiment may be a neural network model, such as a deep learning model DeepFM, etc. It may be flexibly determined according to needs in actual applications, and this embodiment does not limit this.

[0072] Through the above embodiments, a machine learning model can be trained. In practical applications, the trained machine learning model can be set in the client or in the server. When used online, through the machine learning model, each unfinished delivery waybill of the delivery capacity can call the model to predict the remaining delivery time and timeout probability at each moment.

[0073] Exemplarily, the waybill information of the delivery waybill that is currently in the delivery status at the current moment is obtained; wherein the waybill information, as in the aforementioned embodiment, may include the geographical location information of the delivery capacity, the traffic information of the location of the delivery capacity, the number of delivery waybills undertaken by the delivery capacity, the collection status information of the delivery items of each delivery waybill undertaken by the delivery capacity, the attribute information of the delivery items of each delivery waybill undertaken by the delivery capacity, or the historical delivery information of the delivery capacity.

[0074] The input of the machine learning model is the waybill information at the current moment. Based on the waybill information at the current moment, the model predicts the mean and variance of the remaining delivery time at the current moment. The normal distribution of the remaining delivery time at the current moment can be determined by the mean and variance. According to the remaining committed time at the current moment and the normal distribution, the timeout probability of the delivery waybill at the current moment is determined. For example, as in the aforementioned embodiment, the integral formula of the normal distribution is used to calculate the size of the area after the remaining committed time at the current moment, that is, to calculate the integral of the x value greater than the remaining committed time at the current moment, and the obtained value is the timeout probability at the current moment.

[0075] In actual applications, during the delivery process of a delivery order, the solution of this embodiment can be used multiple times as needed, for example, at multiple different times, to predict the remaining delivery time and timeout probability at each time.

[0076] It can be seen that this embodiment is designed to train the machine learning model based on the normal distribution of the remaining delivery time at the current moment, and the output of the machine learning model is designed to be the mean of the remaining delivery time at the current moment and the variance of the remaining delivery time. Based on this, the information estimated by these two models can be used to determine the normal distribution curve of the remaining delivery time at the current moment, and combined with the remaining committed user time at the current moment, the normal distribution is used to determine the timeout probability of the delivery waybill at the current moment. In this embodiment, by designing that the remaining delivery time at the current moment follows the normal distribution, the two pieces of information are associated, and one model simultaneously estimates the two related pieces of information, so the difficulty of training and maintaining the model is significantly reduced, and the prediction accuracy of the model is also guaranteed.

[0077] Based on the two pieces of information output by the model, a normal distribution is designed to determine the timeout probability. Due to the accurate output of the model, an accurate timeout probability can be obtained. Moreover, this embodiment uses the model to estimate the distribution of the data itself, which can not only obtain an estimated value of the true value, but also calculate the timeout probability according to the distribution, thereby improving versatility.

[0078] This embodiment estimates the distribution of the waybills in delivery, obtains the estimated value of the remaining delivery time, and also obtains the overtime rate of the current waybills. The solution is simple and easy to implement, which greatly simplifies and improves the effect. In addition, the estimated distribution method of this embodiment can maintain the unity of the mean and variance fields at the same time, and can also cope with the need for other estimated fields in the future; by obtaining a normal distribution through the mean and variance, other various parameters can be calculated according to the mathematical formula of the distribution, which improves the versatility of the solution of this embodiment and facilitates maintenance and iteration.

[0079] By using the information processing method of the delivery waybill of this embodiment, after obtaining the remaining delivery time at the current moment estimated by the model and the estimated timeout probability, the remaining waiting time can be displayed to the user through the client, and whether to trigger the timeout warning process can be determined according to the timeout probability. For example, if the timeout probability is high, the reason for the current timeout can be pushed to the user, or rights and interests can be issued to the user, etc., to improve the user experience. Alternatively, the estimated information can also be provided to the waybill dispatching system, so that the waybill dispatching can refer to the estimated information and allocate delivery capacity for the delivery waybill to be allocated.

[0080] Taking the takeaway scenario as an example, the server can be configured with a waybill dispatching system. During the dispatching and assigning of waybill, the waybill dispatching system will make judgments on the waybill that the rider is responsible for delivering (i.e., the back order) and the waybill that will be dispatched and assigned to the rider. For example: the current rider has a higher risk of timeout for the back order; when the waybill that is currently in the delivery stage also has a higher risk of timeout, the algorithm for dispatching and assigning waybill will make targeted adjustments to the dispatching and assignment of waybill, thereby reducing the overall proportion of timeout waybill. However, when the dispatching system makes an inaccurate estimate of the timeout risk for the waybill, it will cause instability in the dispatching and assignment of the entire waybill dispatching chain, resulting in reduced efficiency of the dispatching system and inaccurate waybill assignment. By using the solution of this embodiment, since the accurate timeout probability can be estimated, the dispatching accuracy of the waybill dispatching system can be improved, and the stable operation of the entire waybill dispatching chain can be ensured.

[0081] Corresponding to the above-mentioned embodiment of the method for processing information of delivery bills, this specification also provides an embodiment of an apparatus for processing information of delivery bills and a computer device used therein.

[0082] The embodiments of the information processing device for delivery waybills in this specification can be applied to computer devices, such as servers or terminal devices. The device embodiments can be implemented by software, hardware, or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of the file processing in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory and running them. From the hardware level, if Figure 3 The figure is a hardware structure diagram of the computer equipment where the information processing device for the delivery order of this manual is located, except Figure 3 In addition to the processor 310, memory 330, network interface 320, and non-volatile memory 340 shown, the computer device where the information processing device 331 for the delivery waybill is located in the embodiment may also include other hardware according to the actual function of the computer device, which will not be described in detail.

[0083] like Figure 4 As shown, Figure 4 : is a block diagram of a delivery waybill information processing device according to an exemplary embodiment of the present specification, the device comprising:

[0084] The waybill information acquisition module 41 is used to: acquire the waybill information of the delivery waybill currently in the delivery state at the current moment;

[0085] The model prediction module 42 is used to: input the waybill information at the current moment into the machine learning model to obtain the mean value and variance of the remaining delivery time at the current moment predicted by the machine learning model; wherein the machine learning model is trained based on the normal distribution of the remaining delivery time at the current moment;

[0086] The normal distribution determination module 43 is used to determine the normal distribution of the remaining delivery time at the current moment according to the predicted mean value and variance of the remaining delivery time at the current moment;

[0087] The timeout determination module 44 is used to obtain the remaining committed user time at the current moment, and determine the timeout probability of the delivery waybill at the current moment using the normal distribution.

[0088] In some examples, when the machine learning model is trained, the optimization objectives of the evaluation function include: the difference between the mean remaining delivery time and the variance of the remaining delivery time at the current moment estimated by the machine learning model for the delivery waybill samples, and the remaining delivery time as the true value in the delivery waybill samples.

[0089] In some examples, the determining the timeout probability of the delivery waybill at the current moment using the normal distribution includes:

[0090] The area of ​​the horizontal axis interval of the normal distribution curve after the remaining committed time at the current moment is calculated to obtain the timeout probability.

[0091] In some examples, the training data of the machine learning model includes: historical delivery waybill data, the historical delivery waybill data includes the promised user time, actual delivery time and waybill information of the historical delivery waybill; wherein the historical delivery waybill data includes multiple samples determined according to set time intervals.

[0092] In some examples, the waybill information includes one or more of the following information:

[0093] The geographical location information of the delivery capacity, the traffic information of the location of the delivery capacity, the number of delivery waybills undertaken by the delivery capacity, the collection status information of the delivery items of each delivery waybill undertaken by the delivery capacity, the attribute information of the delivery items of each delivery waybill undertaken by the delivery capacity, or the historical delivery information of the delivery capacity.

[0094] In some examples, the method further comprises one or more of the following steps:

[0095] The average remaining delivery time at the current moment predicted by the machine learning model and the timeout probability of the delivery waybill are displayed to the user;

[0096] Determine whether to trigger the execution of a preset timeout warning process according to the timeout probability; or,

[0097] According to the normal distribution of the remaining delivery time at the current moment, other information of the delivery waybill at the current moment is determined.

[0098] The implementation process of the functions and effects of each module in the above-mentioned delivery waybill information processing device is specifically described in the implementation process of the corresponding steps in the above-mentioned delivery waybill information processing method, which will not be repeated here.

[0099] Accordingly, an embodiment of the present specification also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the aforementioned delivery waybill information processing method embodiment.

[0100] Correspondingly, an embodiment of the present specification also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the embodiment of the information processing method for delivery waybills are implemented.

[0101] Accordingly, the embodiments of the present specification also provide a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the embodiment of the information processing method for delivery waybills are implemented.

[0102] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this specification. A person of ordinary skill in the art can understand and implement it without paying creative labor.

[0103] Exemplarily, the method of this embodiment can be applied to one or more electronic devices, where the electronic device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and the hardware of the electronic device includes but is not limited to a microprocessor, an application specific integrated circuit (ASIC), a programmable gate array (FPGA), a digital processor (DSP), an embedded device, etc.

[0104] The electronic device may be any electronic product that can perform human-computer interaction with a user, such as a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), a game console, an interactive network television (IPTV), a smart wearable device, etc.

[0105] The electronic device may also include a network device and / or a user device, wherein the network device includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud consisting of a large number of hosts or network servers based on cloud computing.

[0106] The network where the electronic device is located includes but is not limited to the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.

[0107] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] The step division of the above methods is only for clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the protection scope of this patent; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the protection scope of this application.

[0109] The description of "specific examples" or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of this specification. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0110] Those skilled in the art will readily appreciate other embodiments of the specification after considering the specification and practicing the invention claimed herein. The specification is intended to cover any variations, uses or adaptations of the specification that follow the general principles of the specification and include common knowledge or customary techniques in the art that are not claimed in the specification. The specification and examples are to be considered exemplary only, and the true scope and spirit of the specification are indicated by the following claims.

[0111] It should be understood that the present description is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present description is limited only by the appended claims.

[0112] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

Claims

1. A method for processing information of a delivery waybill, comprising: Get the waybill information of the delivery waybill that is currently in delivery status at the current moment; Input the waybill information at the current moment into the machine learning model, and obtain the mean value and variance of the remaining delivery time at the current moment predicted by the machine learning model; wherein the machine learning model is trained based on the normal distribution of the remaining delivery time at the current moment; Determine the normal distribution of the remaining delivery time at the current moment according to the predicted mean value and variance of the remaining delivery time at the current moment; Obtain the remaining committed user time at the current moment, and use the normal distribution to determine the timeout probability of the delivery waybill at the current moment; wherein the timeout probability is obtained based on the area of ​​the horizontal axis interval of the normal distribution curve after the remaining committed time at the current moment.

2. According to the method of claim 1, when the machine learning model is trained, the optimization objectives of the evaluation function include: The difference between the mean and variance of the remaining delivery time at the current moment estimated by the machine learning model for the delivery waybill sample and the remaining delivery time as the true value in the delivery waybill sample.

3. According to the method of claim 1, the training data of the machine learning model includes: Historical delivery waybill data, the historical delivery waybill data includes the promised user duration, actual delivery duration and waybill information of the historical delivery waybill; wherein the historical delivery waybill data includes multiple samples determined according to set time intervals.

4. According to the method of claim 1 or 3, the waybill information includes one or more of the following information: The geographical location information of the delivery capacity, the traffic information of the location of the delivery capacity, the number of delivery waybills undertaken by the delivery capacity, the collection status information of the delivery items of each delivery waybill undertaken by the delivery capacity, the attribute information of the delivery items of each delivery waybill undertaken by the delivery capacity, or the historical delivery information of the delivery capacity.

5. The method according to claim 1, further comprising one or more of the following steps: The average remaining delivery time at the current moment predicted by the machine learning model and the timeout probability of the delivery waybill are displayed to the user; Determine whether to trigger the execution of a preset timeout warning process according to the timeout probability; or, According to the normal distribution of the remaining delivery time at the current moment, other information of the delivery waybill at the current moment is determined.

6. A delivery waybill information processing device, comprising: The waybill information acquisition module is used to: obtain the waybill information of the delivery waybill currently in the delivery state at the current moment; A model prediction module, used to: input the waybill information at the current moment into a machine learning model, and obtain the mean value and variance of the remaining delivery time at the current moment predicted by the machine learning model; wherein the machine learning model is trained based on the normal distribution of the remaining delivery time at the current moment; A normal distribution determination module, used to: determine the normal distribution of the remaining delivery time at the current moment according to the predicted mean value and variance of the remaining delivery time at the current moment; The timeout determination module is used to obtain the remaining committed user time at the current moment, and determine the timeout probability of the delivery waybill at the current moment using the normal distribution.

7. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 5.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

Images