Order payment management method and device, equipment and storage medium

Abstract

The application relates to the field of logistics, and discloses an order payment management method, device and equipment and a storage medium, which are used for managing unpaid orders and generating settlement report forms. The method comprises the following steps: acquiring corresponding order information based on the order number of an unpaid order, extracting the logistics state of the unpaid order, screening out the unpaid order in the logistics state of delivery, and marking the unpaid order as a to-be-handled order; acquiring the payment state and the express courier state information of the to-be-handled order at regular time intervals, inputting the order information and the express courier state information of the to-be-handled order into a trained multilayer perception machine model if the payment state of the to-be-handled order is unpaid, acquiring the reminding priority level and the reminding mode corresponding to the to-be-handled order output by the trained multilayer perception machine model, generating a reminding message, and pushing the reminding message to the express courier; and marking the to-be-handled order with the payment state of having been paid as a completed order to generate a settlement report form.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of logistics, and in particular to an order payment management method and device, equipment and a storage medium. BACKGROUND

[0002] With the development of e-commerce business, express service is closely combined with payment business such as collection of freight and cash on delivery. However, the express delivery personnel face many problems when processing orders containing payment information. For example, it is inconvenient to query the payment state. When an express delivery personnel is responsible for a batch of orders, he or she cannot quickly know the payment state, and finds that some cash on delivery orders have not been paid when delivering the goods, which affects the delivery efficiency. It is easy to miss or lose the payment. When collecting the freight, the express delivery personnel forgets to collect the freight of a customer due to no effective reminder. The settlement process after collecting the freight is complicated and prone to errors. The express delivery personnel needs to manually arrange a large amount of order payment information, which is prone to data errors. It is difficult to real-time master the overall payment situation of the responsible orders, and it is impossible to plan an efficient delivery route according to the payment situation. At the same time, the existing logistics system focuses on the control of the package transportation process, and lacks support for order payment management.

[0003] Therefore, the prior art still needs to be improved and developed. SUMMARY

[0004] The present application provides an order payment management method, device, equipment and storage medium for managing unpaid orders and settlement report generation.

[0005] The first aspect of the present application provides an order payment management method, which comprises: obtaining the order number of an unpaid order, and obtaining the order information corresponding to the unpaid order based on the order number of the unpaid order; extracting the logistics state of the unpaid order based on the order information, screening out the unpaid order in the logistics state of delivery, and marking it as a to-be-processed order; obtaining the payment state and the express delivery personnel state information of the to-be-processed order at regular intervals, if the payment state of the to-be-processed order is unpaid, inputting the order information and the express delivery personnel state information of the to-be-processed order into a trained multi-layer perception machine model, obtaining the reminding priority level and the reminding mode corresponding to the to-be-processed order output by the trained multi-layer perception machine model, generating a reminding message based on the reminding priority level and the reminding mode corresponding to the to-be-processed order, and pushing the reminding message to the express delivery personnel; when the payment state of the to-be-processed order changes from unpaid to paid, marking the to-be-processed order with the payment state of paid as a completed order, and regularly collecting the order number and the corresponding payment information of the completed order to generate a settlement report.

[0006] Optionally, in the first implementation manner of the first aspect, the order number of the uncompleted payment order is acquired, the order information corresponding to the uncompleted payment order is acquired based on the order number of the uncompleted payment order, and the method comprises: acquiring the order numbers of the uncompleted payment orders in batches, constructing an SQL query statement based on the order numbers of the uncompleted payment orders; executing the SQL query statement to acquire the order information corresponding to the uncompleted payment orders; and converting the order information corresponding to the uncompleted payment orders into a dictionary form and storing the order information in a pre-constructed order information list.

[0007] Optionally, in the second implementation manner of the first aspect, the logistics state of the uncompleted payment order is acquired, the uncompleted payment order in the delivery state is screened out, and the uncompleted payment order is marked as a to-be-handled order, and the method comprises: traversing the order information of each uncompleted payment order in the order information list to determine whether there is a logistics state field; if yes, adding the value of the logistics state field and the corresponding uncompleted payment order to a logistics state information list; and screening out the uncompleted payment order corresponding to the value of the logistics state field from the logistics state information list, and marking the screened uncompleted payment order as a to-be-handled order.

[0008] Optionally, in the third implementation manner of the first aspect, the payment state and the courier state information of the to-be-handled order are acquired at a timing, if the payment state of the to-be-handled order is unpayment, the order information of the to-be-handled order and the courier state information are input into a trained multi-layer perception model to acquire the reminding priority level and the reminding manner corresponding to the to-be-handled order output by the trained multi-layer perception model, and the method comprises: constructing and training the multi-layer perception model to obtain the trained multi-layer perception model; acquiring the payment state and the courier state information of the to-be-handled order at a timing, the payment state of the to-be-handled order comprises unpayment and payment, and the courier state information comprises the busy degree of the courier and the distance between the current position of the courier and the position of the consignee; and if the payment state of the to-be-handled order is unpayment, the order information of the to-be-handled order and the courier state information are input into the trained multi-layer perception model to acquire the reminding priority level and the reminding manner corresponding to the to-be-handled order output by the trained multi-layer perception model.

[0009] Optionally, in the fourth implementation form of the first aspect of the present application, the constructing and training the multilayer perceptron model to obtain the trained multilayer perceptron model comprises: constructing the multilayer perceptron model using a deep learning framework, the multilayer perceptron model comprising an input layer, a hidden layer and an output layer, the input layer being configured to receive order information of a to-be-processed order and courier state information, the hidden layer being configured to determine a reminding priority level and a reminding manner corresponding to the to-be-processed order according to information output by the input layer, and output a determination result through the output layer; collecting historical order data, the historical order data comprising order types, payment amounts, busy degrees of couriers, distances between current positions of couriers and positions of consignees, and reminding priority levels and reminding manners of historical orders; training the multilayer perceptron model using the historical order data, and optimizing parameters of the multilayer perceptron model using forward propagation and back propagation during the training to obtain the trained multilayer perceptron model.

[0010] Optionally, in the fifth implementation form of the first aspect of the present application, the generating a reminding message based on the reminding priority level and the reminding manner corresponding to the to-be-processed order and pushing the reminding message to the courier comprises: pre-constructing reminding message generation templates corresponding to the reminding priority level and the reminding manner, and saving the reminding message generation templates in a template library in a classified manner; calling a corresponding reminding message generation template from the template library based on the reminding priority level and the reminding manner corresponding to the to-be-processed order; filling order information of the to-be-processed order into the called reminding message generation template to generate the reminding message, and pushing the reminding message to the courier.

[0011] Optionally, in the sixth implementation form of the first aspect of the present application, the marking the to-be-processed order with the payment state of paid as a completed order when the payment state of the to-be-processed order changes from unpaid to paid, and periodically collecting order numbers and corresponding payment information of the completed orders to generate a settlement report comprises: marking the to-be-processed order with the payment state of paid as a completed order when the payment state of the to-be-processed order changes from unpaid to paid; periodically collecting order numbers and corresponding payment information of the completed orders, and dividing the completed orders into normal orders and abnormal orders using a pre-trained convolutional neural network model; respectively summarizing the normal orders and the abnormal orders, and generating a settlement report.

[0012] The second aspect of the present application provides an order payment management device, comprising: an acquisition module configured to acquire an order number of an uncompleted payment order, and acquire order information corresponding to the uncompleted payment order based on the order number of the uncompleted payment order; a marking module configured to extract a logistics state of the uncompleted payment order based on the order information, screen out the uncompleted payment order in a delivery state, and mark the uncompleted payment order as a to-be-handled order; a judgment module configured to acquire a payment state and a courier state information of the to-be-handled order at a regular time, input the order information and the courier state information of the to-be-handled order into a trained multi-layer perception machine model if the payment state of the to-be-handled order is uncompleted, acquire a reminding priority level and a reminding mode corresponding to the to-be-handled order output by the trained multi-layer perception machine model, and generate a reminding message based on the reminding priority level and the reminding mode corresponding to the to-be-handled order; a pushing module configured to push the reminding message to a courier; and a settlement module configured to mark the to-be-handled order with a payment state of completed as a completed order when the payment state of the to-be-handled order changes from uncompleted to completed, and regularly collect order numbers and corresponding payment information of the completed orders to generate a settlement report.

[0013] Optionally, in the first implementation manner of the second aspect of the present application, the acquisition module comprises: a first acquisition unit configured to batch acquire order numbers of uncompleted payment orders, and construct a SQL query statement based on the order numbers of the uncompleted payment orders; a second acquisition unit configured to execute the SQL query statement, and acquire order information corresponding to the uncompleted payment orders; and a conversion unit configured to convert the order information corresponding to the uncompleted payment orders into a dictionary form, and store the order information in a pre-constructed order information list.

[0014] Optionally, in the second implementation manner of the second aspect of the present application, the marking module comprises: a first judgment unit configured to traverse order information of each uncompleted payment order in the order information list, and judge whether there is a logistics state field; an adding unit configured to add a value of the logistics state field and a corresponding uncompleted payment order to a logistics state information list when there is the logistics state field; and a first marking unit configured to screen out the uncompleted payment order corresponding to the value of the logistics state field from the logistics state information list, and mark the screened uncompleted payment order as a to-be-handled order.

[0015] Optionally, in a third implementation form of the second aspect of the present application, the judging module comprises: a training unit configured to construct and train a multi-layer perception model to obtain a trained multi-layer perception model; a third obtaining unit configured to obtain payment state and courier state information of the to-be-processed order at a regular time, the payment state of the to-be-processed order comprising non-payment and payment, and the courier state information comprising a busy degree of the courier and a distance between a current position of the courier and a position of the consignee; and a second judging unit configured to input order information of the to-be-processed order and the courier state information into the trained multi-layer perception model when the payment state of the to-be-processed order is non-payment, and obtain a reminding priority level and a reminding manner corresponding to the to-be-processed order output by the trained multi-layer perception model.

[0016] Optionally, in a fourth implementation form of the second aspect of the present application, the pushing module comprises: a constructing unit configured to pre-construct a reminding message generation template corresponding to the reminding priority level and the reminding manner, and save the reminding message generation template in a template library; a generating unit configured to call a corresponding reminding message generation template from the template library based on the reminding priority level and the reminding manner corresponding to the to-be-processed order; and a pushing unit configured to fill order information of the to-be-processed order into the called reminding message generation template to generate a reminding message, and push the reminding message to the courier.

[0017] Optionally, in a fifth implementation form of the second aspect of the present application, the settlement module comprises: a second marking unit configured to mark the to-be-processed order with the payment state of payment as a completed order when the payment state of the to-be-processed order changes from non-payment to payment; a dividing unit configured to periodically collect order numbers and corresponding payment information of the completed orders, and divide the completed orders into normal orders and abnormal orders by using a pre-trained convolutional neural network model; and a settlement unit configured to respectively aggregate the normal orders and the abnormal orders, and generate a settlement report.

[0018] The third aspect of the present application provides an order payment management device, comprising: a memory and at least one processor, the memory storing computer readable instructions, and the memory and the at least one processor being interconnected by a circuit; the at least one processor invoking the computer readable instructions in the memory to make the order payment management device perform each step of the order payment management method.

[0019] The fourth aspect of the present application provides a computer readable storage medium, the computer readable storage medium storing computer readable instructions, when the computer readable instructions are run on a computer, making the computer perform each step of the order payment management method.

[0020] The technical scheme provided by the application can effectively identify and process orders in a delivery state but not yet paid, intelligently judge and remind priority levels and modes by using a multi-layer perception machine model, and timely notify the express courier to pay attention, thereby improving the automation and intelligent level of order processing, ensuring the timeliness and accuracy of commodity distribution, regularly collecting completed order information to generate settlement reports, facilitating financial management and data analysis, and improving the overall operation efficiency and optimizing the customer experience. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The first flowchart of the order payment management method provided by the embodiment of the application;

[0022] Figure 2 The second flowchart of the order payment management method provided by the embodiment of the application;

[0023] Figure 3 The third flowchart of the order payment management method provided by the embodiment of the application;

[0024] Figure 4 The fourth flowchart of the order payment management method provided by the embodiment of the application;

[0025] Figure 5 The fifth flowchart of the order payment management method provided by the embodiment of the application;

[0026] Figure 6 The sixth flowchart of the order payment management method provided by the embodiment of the application;

[0027] Figure 7 The structural schematic diagram of the order payment management device provided by the embodiment of the application;

[0028] Figure 8 The structural schematic diagram of the order payment management device provided by the embodiment of the application; DETAILED DESCRIPTION

[0029] The embodiment of the present application provides an order payment management method, device and equipment and a storage medium, and the method is used for managing an unpaid order and generating a settlement report. The method comprises the following steps: obtaining an order number of an unpaid order, obtaining order information corresponding to the unpaid order based on the order number of the unpaid order; extracting a logistics state of the unpaid order based on the order information, screening out the unpaid order in a delivery state, and marking the unpaid order as a to-be-handled order; obtaining a payment state and a courier state information of the to-be-handled order at a fixed time, if the payment state of the to-be-handled order is unpaid, inputting the order information and the courier state information of the to-be-handled order into a trained multilayer perception machine model, obtaining a reminding priority level and a reminding mode corresponding to the to-be-handled order output by the trained multilayer perception machine model, generating a reminding message based on the reminding priority level and the reminding mode corresponding to the to-be-handled order, and pushing the reminding message to a courier; and when the payment state of the to-be-handled order is changed from unpaid to paid, marking the to-be-handled order with the payment state of paid as a completed order, and regularly collecting order numbers and corresponding payment information of the completed order to generate a settlement report.

[0030] The terms "first", "second", "third", "fourth" and the like in the description and claims of the present application, and above figures, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of the terms so termed herein is only for distinguishing between the similar objects and not necessarily for describing a particular sequential or chronologic order. It is to be understood that the data so used in this detailed description are interchangeable under appropriate circumstances so that the embodiments of the application described herein are capable of operating in other sequences than the one illustrated or other than the one explicitly described herein. Moreover, the terms "comprising", "having", "including", and the like, as can be used in the detailed description herein are specifically intended to be interpreted as an open-ended transitional phrase, for example, comprising, comprising a series of steps or elements and / or process, method, system, product or apparatus, not by way of restriction, but to indicate the inclusion of the listed steps or elements or process, method, system, product or apparatus, and to further allow for combinations of elements, steps or process, method, system, product or apparatus not expressly listed, other steps or elements known to those of ordinary skill in the art to be inherent in the process, method, system, product or apparatus, and allowing for insubstantial or insubstantial changes therein, or additions thereto, including progressions in the art.

[0031] For the convenience of understanding, the specific flow of the embodiment of the present application is described below. Please refer to Figure 1 The first embodiment of the order payment management method in the embodiment of the present application comprises the following steps:

[0032] S101, obtaining an order number of an unpaid order, and obtaining order information corresponding to the unpaid order based on the order number of the unpaid order.

[0033] It can be understood that the execution subject of the present application can be an order payment management device, and can also be a terminal or a server, and the specific execution subject is not limited herein. The embodiment of the present application takes the server as the execution subject for example.

[0034] In this embodiment, through the order management system, all orders with uncompleted payment are screened out, and the order numbers of these orders are obtained. The order numbers are used as indexes to query and obtain the detailed order information corresponding to these uncompleted payment orders from the database, including but not limited to commodity information, buyer information, logistics information, etc.

[0035] S102, based on the order information, the logistics state of the uncompleted payment order is extracted, the uncompleted payment order with the logistics state of delivery in progress is screened out, and is marked as a to-be-handled order.

[0036] In this embodiment, the order information covers a plurality of dimensions of data, and the logistics state is one of the key information. The logistics state value of each uncompleted payment order is extracted and stored in a temporary data structure (such as a list, a dictionary or a database table) together with the order information thereof. The order information in the temporary data structure is traversed, and it is checked whether the logistics state of each order is "in delivery".

[0037] For the uncompleted payment order with the logistics state of "in delivery", it is marked as a to-be-handled order.

[0038] S103, the payment state and the courier state information of the to-be-handled order are obtained at a time, if the payment state of the to-be-handled order is unpayment, the order information and the courier state information of the to-be-handled order are input into the trained multilayer perceptron model, and the reminding priority level and the reminding mode corresponding to the to-be-handled order output by the trained multilayer perceptron model are obtained.

[0039] In this embodiment, a timer is set to check the payment state of the to-be-handled order and the state information (such as position, delivery progress, etc.) of the courier responsible for the delivery at regular intervals (such as every 30 minutes or every hour).

[0040] In this embodiment, when the payment state of the to-be-handled order is unpayment, the order information (such as order amount, order type, etc.) and the courier state information are input into the trained multilayer perceptron model. The model outputs the reminding priority level (such as high, medium, low) and the reminding mode (such as sound reminding, pop-up window reminding, etc.) corresponding to the to-be-handled order according to the patterns learned during training.

[0041] S104, a reminding message is generated based on the reminding priority level and the reminding mode corresponding to the to-be-handled order, and the reminding message is pushed to the courier.

[0042] In this embodiment, different reminder priority levels and reminder methods are predefined corresponding to the reminder message templates. For example, for a high-priority voice reminder, the appropriate push channel is selected to push the reminder message to the courier according to the reminder method. If it is a voice reminder, a voice message can be sent through the mobile application used by the courier; if it is a pop-up reminder, a reminder window is popped up on the application interface.

[0043] S105, when the payment state of the to-be-processed order changes from unpaid to paid, the to-be-processed order with the payment state of paid is marked as a completed order, and the order number and corresponding payment information of the completed order are collected periodically to generate a settlement report.

[0044] In this embodiment, once it is monitored that the payment state of the to-be-processed order changes from “unpaid” to “paid”, the payment state field of the to-be-processed order in the database is updated immediately, and the to-be-processed order is marked as a completed order.

[0045] In this embodiment, the number, payment amount, payment time and other information of the completed order are collected from the database at a fixed period, for example, every day. The collected information is summarized and counted to generate a settlement report, which facilitates financial management and data analysis of the enterprise and helps understand the fund inflow situation.

[0046] The present embodiment provides an order payment management method, which can effectively identify and process orders in the delivery state but not yet paid, intelligently judge the reminder priority level and method by using a multi-layer perception machine model, and timely notify the courier to pay attention, which not only improves the automation and intelligent level of order processing, but also ensures the timeliness and accuracy of commodity distribution. At the same time, the completed order information is summarized periodically to generate a settlement report, which facilitates financial management and data analysis, and overall improves the operation efficiency and optimizes the customer experience.

[0047] Referring to Figure 2 , the second embodiment of the order payment management method in the present embodiment includes:

[0048] S201, batch obtaining order numbers of orders with incomplete payment, and constructing an SQL query statement based on the order numbers of the orders with incomplete payment.

[0049] In this embodiment, generally, there is a special field in the order table to record the payment state of the order, such as the “payment state” field, which may have different values such as “paid”, “unpaid” and “payment in progress”. The order numbers of the orders with incomplete payment can be batch obtained by using a database query language (such as SQL).

[0050] S202, executing the SQL query statement to obtain order information corresponding to the orders with incomplete payment.

[0051] In this embodiment, the query statement is executed using database linking.

[0052] S203, convert the order information corresponding to the uncompleted payment order into a dictionary form and store it in the pre-constructed order information list.

[0053] In this embodiment, after obtaining the order information corresponding to the uncompleted payment order from the database, the information is usually presented in the form of a table, with each row representing an order and each column representing an order attribute. In order to facilitate subsequent processing and use, it is necessary to convert these table data into a dictionary form.

[0054] Dictionary is a very flexible data structure, which can store order information in the form of key-value pairs by taking order attribute names as keys and attribute values as values. For example, for an order, the attribute names such as "order number", "order amount", "order type" can be taken as keys, and the corresponding specific values as values to construct a dictionary object.

[0055] In this embodiment, each row of order information returned from the database is traversed, and each row of data is converted into a dictionary.

[0056] In this embodiment, the order numbers of uncompleted payment orders are obtained in batches, and SQL query statements are constructed to directly obtain the detailed information of these orders, effectively improving the data retrieval efficiency. At the same time, the order information is converted into a dictionary form and stored in a list, which facilitates subsequent data processing and analysis, and overall improves the automation and intelligence level of uncompleted payment order management.

[0057] Please refer to Figure 3 , a third embodiment of the order payment management method in the embodiment of the application includes:

[0058] S301, traverse the order information of each uncompleted payment order in the order information list, and judge whether there is a logistics state field.

[0059] In this embodiment, the order information list is accessed, and the list is traversed to check the payment status of each order. If the payment status of the order is displayed as "uncompleted payment", the next step is entered. For each uncompleted payment order, further check whether the order information contains a logistics state field. The logistics state field is usually an identifier representing the current logistics stage of the order (such as shipped, in transit, signed, etc.).

[0060] If the field exists, the next step is entered; if it does not exist, the order is skipped and the next one is processed.

[0061] S302, if yes, add the value of the logistics state field and the corresponding uncompleted payment order to the logistics state information list.

[0062] In this embodiment, when it is confirmed that the logistics state field exists in the information of a certain uncompleted payment order, the value of the field and the corresponding order information are extracted and added to a special logistics state information list. This list plays the role of data integration and screening preparation, and all uncompleted payment orders containing logistics state information are concentrated together. For example, the value of the logistics state field in the order information may be "collected", "in transit", "in delivery", "signed for", etc. After adding these values and the corresponding order information to the logistics state information list, it is convenient to screen according to the specific logistics state subsequently.

[0063] S303, from the logistics state information list, the value of the logistics state field and the uncompleted payment order corresponding to the delivery are screened out, and the screened uncompleted payment order is marked as a to-be-processed order.

[0064] In this embodiment, for each order screened out, a marking operation is performed, and its state is updated to "to be processed", which involves updating the order state field in the database or moving the order information to another data structure or queue specially used for processing such orders.

[0065] In this embodiment, by systematically traversing the order information list, checking the logistics state of each uncompleted payment order, and screening out those orders with a logistics state of "in delivery" for special processing (marked as to-be-processed orders), accurate management and timely response to the order state are realized.

[0066] Referring to Figure 4 , a fourth embodiment of the order payment management method in the embodiment of the application includes:

[0067] S401, a multilayer perception machine model is constructed and trained to obtain a trained multilayer perception machine model.

[0068] In this embodiment, a multilayer perception machine model is constructed and trained to obtain a trained multilayer perception machine model, which specifically includes: using a deep learning framework to construct a multilayer perception machine model, the multilayer perception machine model including an input layer, a hidden layer and an output layer, the input layer being used to receive order information of a to-be-processed order and courier state information, the hidden layer being used to judge a reminding priority level and a reminding manner corresponding to the to-be-processed order according to information output by the input layer, and output the judgment result through the output layer; collecting historical order data, the historical order data including order types, payment amounts, busy degrees of couriers, distances between current positions of couriers and positions of consignees, and reminding priority levels and reminding manners of historical orders; using the historical order data to train the multilayer perception machine model, and using forward propagation and backward propagation to optimize parameters of the multilayer perception machine model in the training process to obtain the trained multilayer perception machine model.

[0069] In this embodiment, a multi-layer perceptron model is constructed using TensorFlow or PyTorch. The hidden layer utilizes multiple neurons and complex connection structures to perform nonlinear transformation on the input information, thereby intelligently determining the reminder priority level and reminder method corresponding to the order to be processed. The output layer is responsible for presenting the judgment results of the hidden layer in a clear and explicit manner.

[0070] In this embodiment, the training process is divided into two stages: forward propagation and backward propagation. In the forward propagation stage, the model calculates the predicted results of the output layer through the nonlinear transformation of the hidden layer based on the input historical order data. Then, the predicted results are compared with the actual reminder priority level and reminder method to calculate the loss function value. In the backward propagation stage, the model parameters are updated according to the loss function value using optimization algorithms such as gradient descent, in order to reduce the difference between the predicted results and the actual results. This process is iterated until the performance of the model reaches the preset standard or converges.

[0071] S402, the payment status and courier status information of the order to be processed are acquired in a timely manner.

[0072] In this embodiment, the payment status of the order to be processed includes unpaid and paid, and the courier status information includes the busy degree of the courier and the distance between the current position of the courier and the location of the consignee. The busy degree of the courier includes light busy, medium busy and high busy.

[0073] S403, if the payment status of the order to be processed is unpaid, the order information and courier status information of the order to be processed are input into the trained multi-layer perceptron model to obtain the reminder priority level and reminder method corresponding to the order to be processed output by the trained multi-layer perceptron model.

[0074] In this embodiment, it is assumed that there is an order to be processed, and the order information includes:

[0075] Order type: fresh food

[0076] Payment amount: 200 yuan

[0077] Expected delivery time: before 4 pm;

[0078] The courier status information includes:

[0079] Courier busy degree: medium (currently has 5 orders, all of which can be delivered within the specified time)

[0080] Distance between the current position of the courier and the location of the consignee: 10 kilometers (estimated driving time: 20 minutes)

[0081] The order information and the courier state information are input into the trained multi-layer perceptron model. The model first preprocesses the information, such as converting the text description into numerical features, converting the distance and time into relative values or standardized values, and then performing nonlinear transformation on the input information according to the learned weights and biases to extract features related to the reminder priority level and the reminder method. After a series of calculations, the model finally outputs one or more reminder priority levels and reminder methods corresponding to the to-be-processed order. For example, in this example, the model outputs: the reminder priority level is high, and the reminder method is SMS reminder and APP push reminder at the same time. The reminder priority level is high because the order type is fresh food, which needs to be delivered as soon as possible to maintain freshness. The reminder method is SMS reminder and APP push reminder at the same time, because the courier is currently busy at a medium level, and the distance from the recipient location is far, so multiple reminder methods are needed to ensure that the courier will not forget or delay.

[0082] In this embodiment, a multi-layer perceptron model is constructed and trained using a deep learning framework. The model can receive order information and courier state information of a to-be-processed order as input, process it through hidden layers, intelligently determine and output the reminder priority level and the reminder method corresponding to the to-be-processed order, ensure the timeliness and effectiveness of the reminder, and greatly optimize the work flow of the courier.

[0083] Please refer to Figure 5 The fifth embodiment of the order payment management method in the embodiment of the application includes:

[0084] S501, pre-construct a reminder message generation template corresponding to the reminder priority level and the reminder method, and save the reminder message generation template in the template library.

[0085] In this embodiment, each template should contain necessary placeholders for subsequent insertion of specific order information. For example, for an SMS reminder template, the following design can be used: “

Order reminder

[0086] In this embodiment, the templates are classified and stored according to the reminder priority level (such as high, medium, and low) and the reminder method (such as SMS, APP push, phone, etc.). In this way, the corresponding template can be quickly found according to the specific reminder priority level and reminder method when calling later.

[0087] S502, based on the reminder priority level and the reminder method corresponding to the to-be-processed order, call the corresponding reminder message generation template from the template library.

[0088] In this embodiment, the reminder priority level and the reminder method are extracted from the output of the model. Then, according to these information, the corresponding reminder message generation template is found in the template library.

[0089] S503, fill the order information of the to-be-processed order into the called reminder message generation template to generate a reminder message, and push the reminder message to the courier.

[0090] In this embodiment, after the information filling is completed, the template will be converted into a complete reminder message. This message contains all the necessary order information and reminder information to ensure that the courier can clearly understand the situation of the to-be-processed order. Finally, according to the reminder method (such as SMS, APP push, etc.), the generated reminder message is pushed to the courier.

[0091] In this embodiment, by intelligently judging the reminder priority level and the reminder method of the to-be-processed order, and pre-building the corresponding reminder message generation template, the order information is accurately filled into the template and pushed to the courier, which effectively improves the timeliness and accuracy of order processing, ensures that the courier can respond in time and process important orders in priority, and thus improves customer satisfaction and logistics operation efficiency.

[0092] Please refer to Figure 6 , the sixth embodiment of the order payment management method in the embodiment of the application includes:

[0093] S601, when the payment state of the to-be-processed order changes from unpaid to paid, mark the to-be-processed order with the payment state of paid as a completed order.

[0094] In this embodiment, the payment state of the to-be-processed order is continuously monitored. When it is detected that the payment state of a to-be-processed order changes from “unpaid” to “paid”, it means that the payment process of the order has been completed, and then the to-be-processed order with the payment state of paid is marked as a completed order.

[0095] S602, periodically collect the order number and the corresponding collection information of the completed order, and use a pre-trained convolutional neural network model to divide the completed order into a normal order and an abnormal order.

[0096] In this embodiment, data of completed orders is collected according to a set time period, such as every day. The collected content mainly includes the order number of the completed order and the corresponding collection information, which covers the key data such as the collection amount and the collection time. After the collection is completed, the pre-trained convolutional neural network model is used to further analyze these completed orders. The convolutional neural network divides the completed orders into normal orders and abnormal orders according to the characteristics of the orders, such as the fluctuation of the order amount, the regularity of the payment time, the historical payment behavior of the customer, etc. The normal order represents the transaction that conforms to the regular business process and the expectation, while the abnormal order may contain some situations that need special attention, such as the abnormal payment time of a large order, frequent small payments in a short period of time, etc.

[0097] Suppose there is a customer who usually makes one or two small payments every month to buy some daily necessities. However, on a certain day, he suddenly makes a dozen small payments, and each payment amount is slightly different. This behavior pattern is significantly different from his historical payment behavior.

[0098] When the convolutional neural network is used to analyze the orders of this customer, the model will identify this abnormal payment behavior and divide the orders of this customer into abnormal orders, thereby triggering the risk warning mechanism.

[0099] S603, respectively, summarize the normal orders and the abnormal orders, and generate a settlement report.

[0100] In this embodiment, after the classification of the completed orders is completed, the normal orders and the abnormal orders are summarized respectively. For the normal orders, the order quantity, the total collection amount, the average order amount, etc. are counted; for the abnormal orders, in addition to the basic statistics, the abnormal characteristics and related information are also recorded in detail. Then, the settlement report is generated according to the summarized data.

[0101] In this embodiment, through the timely marking of the payment status change orders, the regular data collection and the scientific order classification, and then the final generation of the settlement report, the completeness and systematicness of the order payment management process are realized. On the one hand, timely marking of the completed orders ensures the accuracy of the order status management, providing a reliable foundation for subsequent data processing; on the other hand, using the convolutional neural network model to divide the order type can quickly identify abnormal orders, helping enterprises to discover potential risks such as fraudulent transactions, abnormal payment patterns, etc., to protect the safety of enterprise funds. At the same time, the settlement report is generated respectively, which provides clear data support for the financial analysis of enterprises.

[0102] The order payment management method in the embodiment of the application is described above, and the device in the embodiment of the application is described below. Please refer to Figure 7The embodiment of the order payment management device in the application comprises:

[0103] The acquisition module 701 is configured to acquire an order number of an uncompleted payment order, and acquire order information corresponding to the uncompleted payment order based on the order number of the uncompleted payment order.

[0104] The marking module 702 is configured to extract a logistics state of the uncompleted payment order based on the order information, filter out the uncompleted payment order in the logistics state of delivery, and mark the uncompleted payment order as a to-be-handled order.

[0105] The judgment module 703 is configured to acquire a payment state and a courier state information of the to-be-handled order at a regular time, input the order information of the to-be-handled order and the courier state information into a trained multi-layer perception machine model if the payment state of the to-be-handled order is unpayment, and acquire a reminding priority level and a reminding mode corresponding to the to-be-handled order output by the trained multi-layer perception machine model.

[0106] The pushing module 704 is configured to generate a reminding message based on the reminding priority level and the reminding mode corresponding to the to-be-handled order, and push the reminding message to a courier.

[0107] The settlement module 705 is configured to mark the to-be-handled order with the payment state of payment as a completed order when the payment state of the to-be-handled order changes from unpayment to payment, and regularly collect order numbers and corresponding collection information of the completed orders to generate a settlement report.

[0108] In the embodiment, the acquisition module 701 comprises: a first acquisition unit 7011 configured to acquire order numbers of uncompleted payment orders in batches, and construct a SQL query statement based on the order numbers of the uncompleted payment orders; a second acquisition unit 7012 configured to execute the SQL query statement, and acquire order information corresponding to the uncompleted payment orders; and a conversion unit 7013 configured to convert the order information corresponding to the uncompleted payment orders into a dictionary form, and store the order information in a pre-constructed order information list.

[0109] In the embodiment, the marking module 702 comprises: a first judgment unit 7021 configured to traverse order information of each uncompleted payment order in the order information list, and judge whether there is a logistics state field; an adding unit 7022 configured to add a value of the logistics state field and a corresponding uncompleted payment order to a logistics state information list when there is the logistics state field; and a first marking unit 7023 configured to filter out the uncompleted payment order corresponding to the value of the logistics state field from the logistics state information list, and mark the filtered uncompleted payment order as a to-be-handled order.

[0110] In this embodiment, the judging module 703 comprises: a training unit 7031 configured to construct and train a multi-layer perception model to obtain a trained multi-layer perception model; a third acquisition unit 7032 configured to acquire payment state and courier state information of the to-be-processed order at regular time intervals, wherein the payment state of the to-be-processed order comprises non-payment and payment, and the courier state information comprises a busy degree of the courier and a distance between a current position of the courier and a position of the consignee; and a second judging unit 7033 configured to input order information of the to-be-processed order and the courier state information into the trained multi-layer perception model when the payment state of the to-be-processed order is non-payment, and acquire a reminding priority level and a reminding manner corresponding to the to-be-processed order output by the trained multi-layer perception model.

[0111] In this embodiment, the pushing module 704 comprises: a construction unit 7041 configured to pre-construct a reminding message generation template corresponding to the reminding priority level and the reminding manner, and save the reminding message generation template in a template library; a generation unit 7042 configured to call a corresponding reminding message generation template from the template library based on the reminding priority level and the reminding manner corresponding to the to-be-processed order; and a pushing unit 7043 configured to fill order information of the to-be-processed order into the called reminding message generation template to generate a reminding message, and push the reminding message to the courier.

[0112] In this embodiment, the settlement module 705 comprises: a second marking unit 7051 configured to mark the to-be-processed order with the payment state of payment as a completed order when the payment state of the to-be-processed order changes from non-payment to payment; a division unit 7052 configured to periodically collect order numbers and corresponding collection information of the completed orders, and divide the completed orders into normal orders and abnormal orders by using a pre-trained convolutional neural network model; and a settlement unit 7053 configured to respectively aggregate the normal orders and the abnormal orders, and generate a settlement report.

[0113] In this embodiment, the order payment management device can effectively identify and process the order with the logistics state of delivery in progress but not yet paid, intelligently judge the reminding priority level and the reminding manner by using the multi-layer perception model, and timely notify the courier to pay attention, thereby improving the automation and intelligent level of order processing, ensuring the timeliness and accuracy of commodity distribution, periodically aggregating the completed order information to generate a settlement report, facilitating financial management and data analysis, and improving the overall operation efficiency and optimizing the customer experience.

[0114] Figure 7 The structure of the order payment management device shown does not constitute a limitation on the order payment management device, and can implement the steps of the order payment management method provided by each method embodiment.

[0115] The above Figure 7The order payment management apparatus in the embodiment of the present application is described in detail from the perspective of the modular functional entity, and the order payment management apparatus in the embodiment of the present application is described in detail from the perspective of hardware processing.

[0116] Figure 8 Fig. 8 is a structural schematic diagram of an order payment management apparatus provided by the embodiment of the present application. The apparatus 800 can be quite different in configuration or performance, and can include one or more processors (central processing units, CPU) 810 (for example, one or more processors) and a memory 820, one or more storage media 830 (for example, one or more mass storage devices) for storing application programs 833 or data 832. The memory 820 and the storage media 830 can be temporary storage or persistent storage. The programs stored in the storage media 830 can include one or more modules (not shown in the figure), and each module can include a series of instruction operations in the apparatus 800. Furthermore, the processor 810 can be configured to communicate with the storage media 830, and execute the series of instruction operations in the storage media on the apparatus 800.

[0117] The apparatus 800 can also include one or more power supplies 840, one or more wired or wireless network interfaces 850, one or more input / output interfaces 860, and / or one or more operating systems 831, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc.

[0118] The embodiment of the present application also provides a computer readable storage medium, which can be a non-volatile computer readable storage medium, and can also be a volatile computer readable storage medium. The computer readable storage medium stores instructions, and when the instructions run on a computer, the computer executes the steps of the order payment management method.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system or apparatus, unit can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0120] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the entire or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0121] The above-described embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the same; even though the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that modifications can still be made to the technical solutions recorded in the foregoing embodiments, or equivalent replacements can be made to some of the technical features; and these modifications or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An order payment management method, characterized in that, The order payment management method includes: Obtain the order number of the incomplete payment order, and obtain the order information corresponding to the incomplete payment order based on the order number of the incomplete payment order; Based on the order information, extract the logistics status of incomplete payment orders, filter out incomplete payment orders with the logistics status of being delivered, and mark them as pending orders; The payment status and courier status information of the pending orders are periodically obtained. If the payment status of the pending orders is unpaid, the order information and courier status information of the pending orders are input into the trained multilayer perceptron model to obtain the reminder priority level and reminder method corresponding to the pending orders output by the trained multilayer perceptron model. A reminder message is generated based on the reminder priority level and reminder method corresponding to the pending order, and the reminder message is pushed to the courier; When the payment status of the pending order changes from unpaid to paid, the pending order with the payment status of paid is marked as a completed order, and the order number and corresponding payment information of the completed order are collected periodically to generate a settlement report.

2. The order payment management method according to claim 1, characterized in that, The step of obtaining the order number of an incomplete payment order and obtaining the order information corresponding to the incomplete payment order based on the order number includes: Retrieve the order numbers of incomplete payment orders in batches, and construct an SQL query statement based on the order numbers of the incomplete payment orders; Execute an SQL query to retrieve the order information corresponding to the incomplete payment order; The order information corresponding to the incomplete payment orders is converted into dictionary form and stored in a pre-built list of order information.

3. The order payment management method according to claim 2, characterized in that, The process of obtaining the logistics status of incomplete payment orders, filtering out incomplete payment orders with a logistics status of "in delivery", and marking them as pending orders includes: Iterate through the order information of each incomplete payment order in the order information list and determine whether a logistics status field exists; If so, add the value of the logistics status field and the corresponding unpaid order to the logistics status information list; Filter the logistics status information list to find the incomplete payment orders that correspond to the values ​​of the logistics status field in the delivery process, and mark the filtered incomplete payment orders as pending orders.

4. The order payment management method according to claim 1, characterized in that, The system periodically acquires the payment status and courier status information of the pending orders. If the payment status of the pending order is unpaid, the system inputs the order information and courier status information of the pending order into a trained multilayer perceptron model to obtain the reminder priority level and reminder method corresponding to the pending order output by the trained multilayer perceptron model, including: Construct and train a multilayer perceptron model to obtain a trained multilayer perceptron model; The payment status and courier status information of the pending orders are obtained periodically. The payment status of the pending orders includes unpaid and paid, and the courier status information includes the courier's busyness level and the distance between the courier's current location and the recipient's location. If the payment status of the pending order is unpaid, the order information and courier status information of the pending order are input into the trained multilayer perceptron model to obtain the reminder priority level and reminder method corresponding to the pending order output by the trained multilayer perceptron model.

5. The order payment management method according to claim 4, characterized in that, The construction and training of the multilayer perceptron model to obtain the trained multilayer perceptron model includes: A multilayer perceptron model is constructed using a deep learning framework. The multilayer perceptron model includes an input layer, a hidden layer, and an output layer. The input layer is used to receive order information and courier status information of the order to be processed. The hidden layer is used to determine the reminder priority level and reminder method corresponding to the order to be processed based on the information output by the input layer, and the judgment result is output through the output layer. Collect historical order data, which includes the order type, payment amount, courier's busyness level, distance between the courier's current location and the recipient's location, and reminder priority and reminder method; The multilayer perceptron model is trained using the historical order data, and the parameters of the multilayer perceptron model are optimized using forward propagation and backpropagation during the training process to obtain a trained multilayer perceptron model.

6. The order payment management method according to claim 1, characterized in that, The step of generating a reminder message based on the reminder priority level and reminder method corresponding to the pending order, and pushing the reminder message to the courier, includes: Pre-build reminder message generation templates corresponding to reminder priority levels and reminder methods, and categorize and save the reminder message generation templates in the template library; Based on the reminder priority and reminder method corresponding to the order to be processed, the corresponding reminder message is retrieved from the template library to generate a template; Enter the order information of the pending orders into the invoked reminder message generation template to generate a reminder message, and then push the reminder message to the courier.

7. The order payment management method according to claim 1, characterized in that, When the payment status of the pending order changes from unpaid to paid, the pending order with the payment status of paid is marked as a completed order, and the order number and corresponding payment information of the completed orders are collected periodically to generate a settlement report, including: When the payment status of the pending order changes from unpaid to paid, the pending order with the payment status of paid is marked as a completed order; The order numbers and corresponding payment information of the completed orders are collected periodically, and the completed orders are divided into normal orders and abnormal orders using a pre-trained convolutional neural network model. The normal orders and the abnormal orders are summarized separately to generate a settlement report.

8. An order payment management device, characterized in that, include: The acquisition module is used to acquire the order number of an incomplete payment order and acquire the order information corresponding to the incomplete payment order based on the order number of the incomplete payment order. The tagging module is used to extract the logistics status of incomplete payment orders based on order information, filter out incomplete payment orders with a logistics status of "delivery in progress", and mark them as pending orders; The judgment module is used to periodically obtain the payment status and courier status information of the pending order. If the payment status of the pending order is unpaid, the order information and courier status information of the pending order are input into the trained multilayer perceptron model to obtain the reminder priority level and reminder method corresponding to the pending order output by the trained multilayer perceptron model. The push module is used to generate a reminder message based on the reminder priority and reminder method corresponding to the order to be processed, and push the reminder message to the courier; The settlement module is used to mark the pending orders with a payment status of "paid" as completed orders when the payment status of the pending orders changes from "unpaid" to "paid", and to periodically collect the order numbers and corresponding payment information of the completed orders in order to generate a settlement report.

9. An order payment management device, characterized in that, It includes a memory and at least one processor, wherein the memory stores computer-readable instructions; The at least one processor invokes the computer-readable instructions in the memory to perform the steps of the order payment management method as described in any one of claims 1-7.

10. A computer-readable storage medium storing computer-readable instructions thereon, characterized in that, When the computer-readable instructions are executed by a processor, they implement the steps of the order payment management method as described in any one of claims 1-7.

Images