Abnormal order processing method and device, electronic equipment and storage medium

CN113408783BActive Publication Date: 2026-11-03BEIJING SANKUAI ONLINE TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202110519682.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-05-12
Publication Date
2026-11-03
Estimated Expiration
2041-05-12

AI Technical Summary

Technical Problem

In existing technologies, abnormal disturbances during the food preparation process by merchants can lead to inaccurate estimates of order preparation time, affecting the delivery efficiency of riders, especially for large orders or orders with multiple dishes, which can cause riders to wait for too long.

Method used

By acquiring the identification features of orders, a classification model is used to determine the probability that an order is an abnormal order, and the target extension time for order preparation time is determined based on the probability, so as to carry out delivery scheduling to reduce rider waiting time.

Benefits of technology

It enables accurate prediction and reasonable delivery scheduling of abnormal orders, reduces rider waiting time, and improves delivery efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113408783B_ABST
    Figure CN113408783B_ABST
Patent Text Reader

Abstract

Embodiments of the application disclose an abnormal order processing method and device, electronic equipment and a storage medium. The method comprises: obtaining an identification feature of a to-be-identified order, the identification feature being a feature that affects order preparation time; determining, according to the identification feature, a probability that the to-be-identified order is an abnormal order by using a classification model; determining, according to the probability, a target extension time of order preparation time of the to-be-identified order and orders within a first preset time after the to-be-identified order; and performing delivery scheduling on the to-be-identified order and the orders within the first preset time according to the target extension time. The embodiments of the application can accurately estimate the target extension time of abnormal orders and the corresponding target extension time, thereby performing reasonable delivery scheduling, reducing the waiting time of a rider, and improving delivery efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to an abnormal order processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] In the food delivery industry, it is common to encounter situations where restaurants are slow to prepare food while riders arrive at the store early. This situation affects the riders' delivery efficiency. Existing technologies can provide some guidance for riders to arrive at the store by estimating the order preparation time.

[0003] However, during the actual food preparation process, there may be some abnormal disturbances, such as a sudden influx of large orders or a large number of orders for a variety of dishes. These disturbances can make the food preparation time estimate inaccurate. Moreover, the food preparation time estimate can only respond to the current order and does not take into account the impact on subsequent orders, which will cause riders to wait for a long time and affect delivery efficiency. Summary of the Invention

[0004] This application provides an abnormal order processing method, apparatus, electronic device, and storage medium, which helps reduce rider waiting time and improve delivery efficiency.

[0005] To address the aforementioned problems, in a first aspect, embodiments of this application provide a method for handling abnormal orders, including:

[0006] Obtain the identification features of the order to be identified, wherein the identification features are those that affect the order preparation time;

[0007] Based on the identified features, the probability that the order to be identified is an abnormal order is determined by a classification model;

[0008] Based on the probability, determine the target extension time for the order preparation time of the order to be identified and the order within a first preset time after the order to be identified;

[0009] Based on the target extension time, delivery scheduling is performed on the orders to be identified and the orders within the first preset time.

[0010] Secondly, embodiments of this application provide an abnormal order processing device, including:

[0011] The identification feature acquisition module is used to acquire the identification features of the order to be identified, wherein the identification features are those that affect the order preparation time;

[0012] The probability prediction module is used to determine the probability that the order to be identified is an abnormal order based on the identification features and a classification model.

[0013] The extended time determination module is used to determine, based on the probability, the target extended time for the order preparation time of the order to be identified and the orders within a first preset time after the order to be identified;

[0014] The delivery scheduling module is used to schedule the delivery of the orders to be identified and the orders within the first preset time period according to the target extension time.

[0015] Thirdly, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the abnormal order processing method described in embodiments of this application.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the abnormal order processing method disclosed in embodiments of this application.

[0017] The abnormal order processing method, apparatus, electronic device, and storage medium provided in this application obtain the identification features of the order to be identified, determine the probability that the order to be identified is an abnormal order based on the identification features and a classification model, determine the target extension time for the order preparation time of the order to be identified and the orders within a first preset time after the order to be identified based on the probability, and schedule the delivery of the order to be identified and the orders within the first preset time based on the target extension time. This achieves accurate estimation of abnormal orders and the target extension time corresponding to abnormal orders, thereby enabling reasonable delivery scheduling, reducing rider waiting time, and improving delivery efficiency. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of the abnormal order processing method according to Embodiment 1 of this application;

[0020] Figure 2 This is a schematic diagram of the abnormal order processing device according to Embodiment 2 of this application;

[0021] Figure 3 This is a schematic diagram of the structure of the electronic device according to Embodiment 3 of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] Example 1

[0024] This embodiment provides a method for handling abnormal orders, such as... Figure 1 As shown, the method includes steps 110 to 140.

[0025] Step 110: Obtain the identification features of the order to be identified, wherein the identification features are features that affect the order preparation time.

[0026] The abnormal orders identified in this application are those with excessively long actual order preparation times. In the food delivery industry, this can include large orders and / or orders with a large number of dishes. When such orders are received, the merchant's production capacity cannot meet the instantaneous high demand, resulting in order delays. Therefore, in the food delivery industry, such orders are also called "canceled orders." Typical characteristics of these abnormal orders are that they have large order amounts or contain a large number of dishes. The current order experiences an extended preparation time, which also affects the preparation time of subsequent orders, causing them to also be extended. The order preparation time refers to the merchant's promised preparation time for the items in the order. For instant delivery, the order preparation time is crucial information for guiding the rider's arrival time at the store. In the food delivery industry, the order preparation time is also the promised food preparation time, which is the food preparation time promised by the merchant who has signed a service agreement with the food delivery platform. An extended order preparation time means a delayed food preparation, i.e., the actual food preparation time is longer than the promised food preparation time.

[0027] When it is necessary to identify whether an order is abnormal, it is necessary to obtain the identification features of the order. These features can be obtained based on pre-set factors that affect order preparation time. The order to be identified can be an abnormal order reported by the merchant, or it can be any order.

[0028] In one embodiment of this application, obtaining the identification features of the order to be identified includes: obtaining merchant attribute information, merchant capacity information, and order attribute information corresponding to the order to be identified; and encoding the merchant attribute information, merchant capacity information, and order attribute information into the identification features.

[0029] The merchant attribute information includes at least one of the following: merchant size information, merchant business characteristics, city level of the merchant, merchant level, and merchant type;

[0030] The merchant's production capacity information includes at least one of the following: order information during the merchant's historical peak hours, order information during the merchant's historical off-peak hours, order information within the historical time frame of the order to be identified, order information within a second preset time period prior to the order to be identified, and the ratio of the merchant's offline and online transaction information.

[0031] The order attribute information includes at least one of the following: order dish tags, price normalization information, dish normalization information, and order dish price segmentation information.

[0032] Factors leading to extended order preparation time primarily include: merchant attribute information, merchant capacity information, and order attribute information. Merchant attribute information includes at least one of the following: merchant size, merchant operating characteristics, city tier, merchant tier, and merchant type. Merchant size information can include the merchant's order volume ranking within the same city and category, and the normalized value of the merchant's order volume. Merchant operating characteristics are related to the merchant's location. For example, if a merchant is located in an office building or shopping mall, its operating characteristics will vary significantly between weekends and weekdays. In this case, the merchant's operating characteristics can be characterized by the ratio of the merchant's average daily order volume on weekdays to the average daily order volume on weekends. City tier refers to the tier of the city where the merchant is located. For example, if the merchant is located in Beijing, its city tier is Tier 1. Merchant tier refers to the tier of the merchant, such as whether it is a large, medium, or small merchant. Merchant type refers to the type of merchant, which can include chain stores or city-based stores, etc.

[0033] The merchant's capacity information includes at least one of the following: historical peak-hour order information, historical off-peak-hour order information, order information within the historical time frame of the order to be identified, order information within a second preset time period prior to the order to be identified, and the ratio of the merchant's offline to online transaction information. Historical peak-hour order information refers to order information from historical peak-hour periods (such as lunch and / or dinner hours), including: the number of orders received during peak hours within a certain period (such as within a month), the number of orders picked up by riders within a certain period, rider waiting time, actual order preparation time (such as food preparation time), and timeout rate. The timeout rate is the probability that an order's actual order preparation time exceeds the promised order preparation time within a certain period. Historical off-peak-hour order information refers to order information from historical off-peak-hour periods (such as time periods other than lunch and dinner hours), including: the number of orders received during off-peak periods within a certain period, rider waiting time, actual order preparation time (such as food preparation time), and timeout rate. A time slice is created by dividing a 24-hour day into multiple time slices, each with a specific moment and duration. A time slice includes a pre-set length for a given time interval, such as 10 minutes per slice. The order information within the historical time slice of the order to be identified includes the order volume, transaction amount, average order price, number of orders accepted, number of orders picked up by riders, and actual order preparation time within that time slice. The second preset time is typically a preset number of minutes, such as 10 minutes or 5 minutes. The order information within the second preset time period preceding the order to be identified includes the order volume, transaction amount, average order price, number of orders accepted, number of orders picked up by riders, and actual order preparation time within the N minutes prior to the order to be identified. The ratio of offline to online transaction information includes the ratio of offline transaction amount to online transaction amount over a historical period, and / or the ratio of offline order volume to online order volume over a historical period.

[0034] The order attribute information includes at least one of the following: order dish tags, price normalization information, dish normalization information, and order dish price segmentation information. Order dish tags primarily reflect whether the dishes in the order can be pre-made and whether the serving efficiency is affected by quantity. They can include cooking methods and product categories. Cooking methods can include braising, boiling, stir-frying, etc., and product categories can include steamed buns, hot pot, rice balls, etc. Price normalization information is obtained by normalizing the price of the order to be identified based on order prices over a historical period. Dish normalization information is obtained by normalizing the dishes in the order to be identified based on the dish information within orders over a historical period. Order dish price segmentation information refers to the price segment of the dishes in the order to be identified. This price segmentation is a pre-set segmentation of dish prices, such as between 10 and 20 yuan, between 20 and 30 yuan, and above 30 yuan, etc.

[0035] After obtaining the merchant attribute information, merchant capacity information, and order attribute information corresponding to the order to be identified, the merchant attribute information, merchant capacity information, and order attribute information are encoded according to the corresponding encoding methods to obtain the identification features of the order to be identified. The obtained identification features can well reflect the impact on order preparation time, so as to classify the order to be identified and obtain the accurate probability that the order to be identified is an abnormal order.

[0036] Step 120: Based on the identification features, determine the probability that the order to be identified is an abnormal order using a classification model.

[0037] After obtaining the identification features of the order to be identified, the identification features are input into a classification model. The classification model processes the identification features to obtain the probability that the order to be identified is an abnormal order. The classification model can be a neural network model.

[0038] Step 130: Based on the probability, determine the target extension time for the order preparation time of the order to be identified and the order within a first preset time after the order to be identified.

[0039] A pre-defined correlation can be established between probability and the extension of order preparation time. After determining the probability that an order to be identified as an abnormal order through a classification model, the correlation is queried to obtain the target extension time for the order preparation time of the order to be identified and all orders received within a first preset time after the order to be identified. After determining the target extension time, the target extension time can be extended based on the order preparation time promised by the merchant, serving as the order preparation time for the order to be identified and all orders received within the first preset time after the order to be identified. The first preset time is in minutes, such as 10 minutes, 15 minutes, etc.

[0040] In one embodiment of this application, determining the target extension time for the order preparation time of the order to be identified and the order within a first preset time period after the order to be identified, based on the probability, includes: determining the target extension time for the order preparation time of the order to be identified and the order within a first preset time period after the order to be identified, based on the probability and the correspondence between the preset probability range and the extension time level.

[0041] Multiple extension time levels can be preset, and a probability range corresponding to each extension time level can be set. After identifying the probability that an order to be identified is an abnormal order, the probability range within which the probability falls is determined. The extension time level corresponding to the determined probability range is obtained from the correspondence between the probability range and the extension time level. This extension time level is used as the target extension time for the order preparation time of the order to be identified and orders within a first preset time period after the order to be identified. The probabilities within the probability ranges are all greater than or equal to a probability threshold. For example, if the probability threshold is 70%, it can include three probability ranges: 70% to 80%, 80% to 90%, and 90% to 100%. Each probability range corresponds to an extension time level. For example, the extension time levels can be 5 minutes, 8 minutes, and 10 minutes respectively. The extension time level corresponding to the probability range "70% to 80%" is 5 minutes, the extension time level corresponding to the probability range "80% to 90%" is 8 minutes, and the extension time level corresponding to the probability range "90% to 100%" is 10 minutes. The extended time tiers can be obtained by statistically analyzing the actual extended time of abnormal orders in a specific region based on the promised order preparation time. This results in more reasonable extended time tiers.

[0042] By determining the target extension time for order preparation time of the order to be identified and the order within the first preset time after the order to be identified, a more reasonable target extension time can be determined based on the correspondence between the pre-set probability range and the extension time level. This can be effectively used to guide delivery scheduling and improve delivery efficiency.

[0043] Step 140: Based on the target extension time, schedule the delivery of the order to be identified and the orders within the first preset time.

[0044] After determining the target extension time for the order preparation time of the order to be identified and the orders received within a first preset time after the order to be identified, the target extension time is extended based on the order preparation time to serve as the order preparation time promised by the merchant within the first preset time. Based on this extended order preparation time, delivery scheduling is performed for the order to be identified and the orders received within the first preset time after the order to be identified. When scheduling delivery based on the extended order preparation time, the delivery time of the order to be identified and the orders received within the first preset time after the order to be identified can be estimated. Based on the estimated delivery time, orders already assigned to riders can be reassigned to influence rider arrival times and avoid excessive waiting times for riders at the merchant.

[0045] The abnormal order processing method provided in this application obtains the identification features of the order to be identified, determines the probability that the order to be identified is an abnormal order based on the identification features through a classification model, determines the target extension time for the order preparation time of the order to be identified and the orders within a first preset time after the order to be identified based on the probability, and schedules the delivery of the order to be identified and the orders within the first preset time according to the target extension time. This achieves accurate estimation of abnormal orders and the target extension time corresponding to abnormal orders, thereby enabling reasonable delivery scheduling, reducing rider waiting time and improving delivery efficiency.

[0046] Based on the above technical solution, after determining the target extension time of the order preparation time of the order to be identified and the order within a first preset time after the order to be identified, the method further includes: using the target extension time as the default selected time level among the extension time levels, and sending the extension time level to the merchant end of the order to be identified, so that the merchant end displays the extension time level and obtains the merchant's selected time level; receiving the selected time level sent by the merchant end;

[0047] Based on the target extended time, the delivery scheduling of the orders to be identified and the orders within the first preset time period includes: based on the selected time range, the delivery scheduling of the orders to be identified and the orders within the first preset time period.

[0048] Since the target extension time is one of several pre-set extension time ranges, it is used as the default selected time range. Multiple extension time ranges, including the default selected time range, are sent to the merchant's end. This allows the merchant to be reminded that the order preparation time for orders awaiting identification can be extended. Upon receiving these multiple extension time ranges, the merchant displays them and defaults to selecting the target extension time range. The selected time range is then retrieved. The merchant can directly select the default selected time range or choose another extension time range. After obtaining the selected time range, the merchant sends it to the backend server. The backend server receives the selected time range from the merchant and uses it to schedule delivery for orders awaiting identification and orders within a first preset time period. By pushing multiple extension time ranges to the merchant, they can select the appropriate extension time range based on their own circumstances, thus providing more reasonable guidance for subsequent delivery scheduling and further improving delivery efficiency.

[0049] Based on the above technical solution, it also includes: determining the abnormal scenario corresponding to the order to be identified based on the order amount and the number of dishes in the order to be identified;

[0050] Sending the extended time range to the merchant's end of the order to be identified includes: generating a prompt message including the extended time range based on the abnormal scenario, and sending the prompt message to the merchant's end of the order to be identified.

[0051] The abnormal scenarios may include scenarios involving large sums of money or scenarios involving multiple dishes.

[0052] Obtain the average order price of the merchant for the order to be identified within a historical statistical period (e.g., within one month). (If the average order price of the merchant is unavailable, use the average order price of the same category in the same city.) Determine the ratio of the order amount to the average order price, and use this ratio as the order amount ratio. If the order amount ratio is greater than or equal to a preset ratio threshold, the abnormal scenario of the order to be identified is determined to be a large-amount scenario. Remove duplicate dishes from the dishes in the order to be identified, and remove packaging and gifts, etc., to obtain the number of dishes. If the number of dishes is greater than or equal to a preset quantity threshold, the abnormal scenario of the order to be identified is determined to be a multi-dish scenario.

[0053] The text of the prompt message is different for different abnormal scenarios. After obtaining the abnormal scenario of the order to be identified, the prompt message text corresponding to the abnormal scenario can be obtained, and the prompt message including the extended time level can be generated. The prompt message is sent to the merchant terminal of the order to be identified. When the merchant terminal detects the operation instruction of the prompt message, it displays multiple extended time levels and selects the extended time level corresponding to the target extended time by default.

[0054] For example, if the probability of an order being identified as an abnormal order is greater than or equal to a first probability threshold and the order amount ratio is greater than or equal to a preset ratio threshold, then the order to be identified is classified as a high-value order and pushed to the platform. If the probability of an order being identified as an abnormal order is greater than or equal to a first probability threshold and the number of dishes in the order is greater than or equal to a preset quantity threshold, then the order to be identified is classified as a multi-dish order and pushed to the platform. Based on the predicted probability, the platform recommends that the order preparation time be extended by X minutes. For example, the recommendation to extend the order preparation time can be divided into three levels: 5 minutes, 8 minutes, and 10 minutes. The 5-minute level is selected by default if the predicted probability is greater than or equal to the first probability threshold and less than the second probability threshold; the 8-minute level is selected by default if the predicted probability is greater than or equal to the second probability threshold and less than the third probability threshold; and the 10-minute level is selected by default if the predicted probability is greater than or equal to the third probability threshold.

[0055] By generating corresponding prompts based on the abnormal scenarios of the orders to be identified, merchants can clearly understand the abnormal situation of abnormal orders.

[0056] Based on the above technical solution, before determining the probability that the order to be identified is an abnormal order through a classification model according to the identification features, the method further includes: determining historical orders that meet preset conditions as positive samples of abnormal orders, and determining historical orders that do not meet the preset conditions as negative samples of abnormal orders; training the classification model based on the positive samples and the negative samples to obtain a trained classification model.

[0057] The preset condition is that the average delay time of order preparation time corresponding to the current historical order and at least one historical order within a third preset time period is greater than a fourth preset time. The third preset time can be greater than the fourth preset time, and both are minute-level data. For example, the third preset time is 10 minutes, the fourth preset time is 5 minutes, and the preset condition is that the average delay time of the current historical order and historical orders within the next 10 minutes is greater than 5 minutes.

[0058] When labeling samples, identifying abnormal orders differs from estimating food preparation time. While food preparation time or rider pickup time can be directly used as training labels, the labels for abnormal order identification need to be adjusted based on data analysis and business characteristics. Analysis of online time reveals that abnormal orders generally impact orders within the first preset timeframe (e.g., 10 minutes), resulting in higher rider waiting time delays and even higher rates of severe delays. Therefore, historical orders containing promised order preparation times and using platform-provided delivery can be selected from historical data. From these historical orders, orders meeting and not meeting preset conditions are identified. Orders meeting the preset conditions are designated as positive samples, and those not meeting the conditions as negative samples. When the merchant's actual order preparation time is unavailable, the rider's pickup time can be used as a substitute.

[0059] After obtaining the labeled positive and negative samples, the identification features of each sample are obtained and input into the classification model. The network parameters of the classification model are adjusted based on the labeled values ​​of the samples until the classification model is trained.

[0060] By identifying historical orders that meet preset conditions as positive samples of abnormal orders and other historical orders as negative samples of abnormal orders, automatic labeling of sample data is achieved.

[0061] Example 2

[0062] This embodiment provides an abnormal order processing device, such as... Figure 2 As shown, the abnormal order processing device 200 includes:

[0063] The identification feature acquisition module 210 is used to acquire the identification features of the order to be identified, wherein the identification features are features that affect the order preparation time;

[0064] The probability prediction module 220 is used to determine the probability that the order to be identified is an abnormal order based on the identification features and through a classification model.

[0065] The extended time determination module 230 is used to determine, based on the probability, the target extended time for the order preparation time of the order to be identified and the order within a first preset time after the order to be identified;

[0066] The delivery scheduling module 240 is used to schedule the delivery of the order to be identified and the order within the first preset time according to the target extension time.

[0067] Optionally, the extended time determination module is specifically used for:

[0068] Based on the correspondence between the probability and the pre-set probability range and the extension time level, the target extension time for the order preparation time of the order to be identified and the order within the first preset time after the order to be identified is determined.

[0069] Optionally, the device further includes:

[0070] The time-range sending module is used to use the target extended time as the default selected time range among the extended time ranges, and send the extended time range to the merchant's end of the order to be identified, so that the merchant's end can display the extended time range and obtain the merchant's selected time range;

[0071] The selected time slot receiving module is used to receive the selected time slot sent by the merchant.

[0072] The delivery scheduling module is specifically used for:

[0073] Based on the selected time range, the delivery scheduling is performed on the order to be identified and the order within the first preset time period.

[0074] Optionally, the device further includes:

[0075] An abnormal scenario determination module is used to determine the abnormal scenario corresponding to the order to be identified based on the order amount and the quantity of dishes in the order to be identified.

[0076] The time-level transmission module includes:

[0077] The prompt information sending unit is used to generate prompt information including the extended time level according to the abnormal scenario, and send the prompt information to the merchant terminal of the order to be identified.

[0078] Optionally, the feature acquisition module includes:

[0079] The information acquisition module is used to acquire the merchant attribute information, merchant capacity information, and order attribute information corresponding to the order to be identified;

[0080] The encoding module is used to encode the merchant attribute information, merchant capacity information, and order attribute information into the identification features.

[0081] Optionally, the merchant attribute information includes at least one of the following: merchant size information, merchant operating characteristics, city classification of the merchant, merchant classification, and merchant type;

[0082] The merchant's production capacity information includes at least one of the following: order information during the merchant's historical peak hours, order information during the merchant's historical off-peak hours, order information within the historical time frame of the order to be identified, order information within a second preset time period prior to the order to be identified, and the ratio of the merchant's offline and online transaction information.

[0083] The order attribute information includes at least one of the following: order dish tags, price normalization information, dish normalization information, and order dish price segmentation information.

[0084] Optionally, the device further includes:

[0085] The sample labeling module is used to identify historical orders that meet preset conditions as positive samples of abnormal orders, and to identify historical orders that do not meet the preset conditions as negative samples of abnormal orders.

[0086] The classification model training module is used to train the classification model based on the positive samples and the negative samples to obtain the trained classification model.

[0087] Optionally, the preset condition is that the average delay time of the order preparation time corresponding to the current historical order and at least one historical order within a third preset time period is greater than a fourth preset time.

[0088] The abnormal order processing device provided in this application embodiment is used to implement the steps of the abnormal order processing method described in Embodiment 1 of this application. The specific implementation of each module of the device is described in the corresponding steps, and will not be repeated here.

[0089] The abnormal order processing device provided in this application embodiment obtains the identification features of the order to be identified through an identification feature acquisition module, determines the probability that the order to be identified is an abnormal order based on the identification features and a classification model, determines the target extension time for the order preparation time of the order to be identified and the orders within a first preset time after the order to be identified based on the probability, and performs delivery scheduling based on the target extension time for the order to be identified and the orders within the first preset time. This achieves accurate estimation of abnormal orders and the target extension time corresponding to abnormal orders, thereby enabling reasonable delivery scheduling, reducing rider waiting time, and improving delivery efficiency.

[0090] Example 3

[0091] This application also provides an electronic device, such as... Figure 3As shown, the electronic device 300 may include one or more processors 310 and one or more memories 320 connected to the processors 310. The electronic device 300 may also include an input interface 330 and an output interface 340 for communicating with another device or system. Program code executed by the processor 310 may be stored in the memory 320.

[0092] The processor 310 in the electronic device 300 calls the program code stored in the memory 320 to execute the abnormal order processing method in the above embodiment.

[0093] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the abnormal order processing method as described in Embodiment 1 of this application.

[0094] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus embodiments, since they are fundamentally similar to the method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0095] The above provides a detailed description of an abnormal order processing method, apparatus, electronic device, and storage medium provided by the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

[0096] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

Claims

1. A method for handling abnormal orders, characterized in that, include: Historical orders that meet preset conditions are identified as positive samples of abnormal orders, and historical orders that do not meet the preset conditions are identified as negative samples of abnormal orders. The classification model is trained based on the positive and negative samples to obtain the trained classification model. Obtain the identification features of the order to be identified, wherein the identification features are those that affect the order preparation time; Based on the identified features, the probability that the order to be identified is an abnormal order is determined by the classification model; Based on the correspondence between the probability and the preset probability range and the extension time level, the target extension time for the order preparation time of the order to be identified and the order within the first preset time after the order to be identified is determined; Based on the target extension time, the delivery scheduling is performed on the orders to be identified and the orders within the first preset time. The preset condition is that the average delay time of the order preparation time corresponding to the current historical order and at least one historical order within the third preset time period is greater than the fourth preset time.

2. The method according to claim 1, characterized in that, After determining the order to be identified and the target extension time for the order preparation time within a first preset time after the order to be identified, the method further includes: The target extended time is used as the default selected time range among the extended time ranges, and the extended time range is sent to the merchant's end of the order to be identified, so that the merchant's end displays the extended time range and obtains the merchant's selected time range; Receive the selected time range sent by the merchant; Based on the target extension time, delivery scheduling is performed on the orders to be identified and the orders within the first preset time period, including: Based on the selected time range, the delivery scheduling is performed on the order to be identified and the order within the first preset time period.

3. The method according to claim 2, characterized in that, Also includes: Based on the order amount and the number of dishes in the order to be identified, determine the abnormal scenario corresponding to the order to be identified; Sending the extended time range to the merchant's end of the order to be identified includes: Based on the abnormal scenario, a prompt message including the extended time level is generated, and the prompt message is sent to the merchant's end of the order to be identified.

4. The method according to claim 1, characterized in that, The process of obtaining the identification features of the order to be identified includes: Obtain the merchant attribute information, merchant capacity information, and order attribute information corresponding to the order to be identified; The merchant attribute information, merchant capacity information, and order attribute information are encoded into the identification features.

5. The method according to claim 4, characterized in that, The merchant attribute information includes at least one of the following: merchant size information, merchant business characteristics, city level where the merchant is located, merchant level, and merchant type. The merchant's production capacity information includes at least one of the following: order information during the merchant's historical peak hours, order information during the merchant's historical off-peak hours, order information within the historical time frame of the order to be identified, order information within a second preset time period prior to the order to be identified, and the ratio of the merchant's offline and online transaction information. The order attribute information includes at least one of the following: order dish tags, price normalization information, dish normalization information, and order dish price segmentation information.

6. An abnormal order processing device, characterized in that, include: The sample labeling module is used to identify historical orders that meet preset conditions as positive samples of abnormal orders, and to identify historical orders that do not meet the preset conditions as negative samples of abnormal orders. The classification model training module is used to train the classification model based on the positive samples and the negative samples to obtain the trained classification model. The identification feature acquisition module is used to acquire the identification features of the order to be identified, wherein the identification features are those that affect the order preparation time; The probability prediction module is used to determine the probability that the order to be identified is an abnormal order based on the identification features and the classification model. The extended time determination module is used to determine the target extended time for the order preparation time of the order to be identified and the order within a first preset time after the order to be identified, based on the probability and the correspondence between the preset probability range and the extended time level. The delivery scheduling module is used to schedule the delivery of the order to be identified and the order within the first preset time according to the target extended time. The preset condition is that the average delay time of the order preparation time corresponding to the current historical order and at least one historical order within the third preset time period is greater than the fourth preset time.

7. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the abnormal order processing method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the abnormal order processing method according to any one of claims 1 to 5.

Citation Information

Patent Citations

  • Order meal supply time prediction model construction and prediction method, model and device

    CN108345958A

  • Data processing method and device, model training method and device, electronic equipment and storage medium

    CN111598487A