Supply chain order management method and device, and medium

Abstract

The embodiment of the invention discloses a supply chain order management method and device and a medium, and relates to the technical field of supply chains, and the method comprises the steps: receiving the return request information of a user, recognizing the returned goods according to the return request information, determining the saleable identification attribute of the returned goods, and sending the saleable identification attribute to the user; the return request information comprises user description information, returned cargo image information and returned cargo information, and the returned cargo information comprises cargo identification; based on the cargo identifier, determining a supply order cooperation mode corresponding to the returned cargo, so as to determine target return information of the returned cargo based on the supply order cooperation mode and the saleable identifier attribute; generating goods return intention order information through the target goods return information of the returned goods and a pre-obtained goods return rule, and sending the goods return intention order information to the user side; and under the confirmation trigger of the user side, based on the return intention order information, generating a return and exchange tracking order of the returned goods so as to perform return and exchange order management on the returned goods.

Description

Technical Field

[0001] The present invention relates to the field of supply chain technology, and in particular to a supply chain order management method, device and medium. Background Art

[0002] The supply chain is a functional network chain structure model that controls the flow of information, logistics, and funds, starting from the purchase of raw materials, making intermediate products and final products, and finally delivering products to consumers through the sales network. Accurate order management can avoid overproduction or underproduction, reduce losses caused by inventory backlogs or out-of-stocks, and optimize logistics distribution to reduce transportation costs and improve resource utilization efficiency. In the supply chain role, there are different cooperation models between distributors and suppliers. In the buyout sales cooperation model, distributors purchase products from suppliers for sale, and distributors bear the inventory risk and market sales risk of the products; in the agency sales cooperation model, distributors act as agents of suppliers and sell products according to the prices and sales policies specified by suppliers.

[0003] With the increasingly fierce market competition, consumers have higher and higher requirements for shopping experience, and return service has become one of the important factors affecting consumers' purchasing decisions. In the traditional return management method, users usually choose to send back the goods, and after the users send back the goods, the merchants determine whether they can be put into the next sales process. However, in the application scenario of the supply chain, the number of ordered goods is large, and the impact of return orders is different for different cooperation models. For example, in the buyout sales cooperation model, the number of returns will lead to the risk of backlog of distributors' inventory; in the agency sales cooperation model, returns will affect the supplier's inventory and increase transportation costs. In other words, whether for sellers or consumers, the impact of user returns on suppliers and distributors under different cooperation models is not considered, and the return logistics process is put in advance, and the subsequent process is executed after the return logistics arrives, which not only increases the risk of inventory backlog of returned goods for sellers, but also increases the return cost of consumers. Consumers lack the choice of return warehouses in the return process, and the return costs are not transparent, which affects the purchasing experience.

[0004] Therefore, in the current traditional supply chain order management process, the return logistics process is put in place in advance, without fully considering the impact of user returns on suppliers and distributors under different cooperation models, resulting in the need to improve the order management effect for return orders. Summary of the invention

[0005] One or more embodiments of the present specification provide a supply chain order management method, device and medium for solving the following technical problems: In the current traditional supply chain order management process, the return logistics process is put in advance, and the impact of user returns on suppliers and distributors under different cooperation models is not fully considered, resulting in the need to improve the order management effect for return orders.

[0006] One or more embodiments of this specification adopt the following technical solutions: One or more embodiments of the present specification provide a supply chain order management method, the method comprising: receiving a user's return request information, and under the triggering of the return request information, identifying the returned goods according to the return request information, and determining the saleable identification attribute of the returned goods, wherein the return request information includes user description information, returned goods image information and returned goods information, and the returned goods information includes a goods identification; based on the goods identification, determining a supply order cooperation mode corresponding to the returned goods, so as to determine the target return information of the returned goods based on the supply order cooperation mode and the saleable identification attribute; generating return intention order information through the target return information of the returned goods and the pre-acquired return rules, and sending the return intention order information to the user end, wherein the return intention order information includes user return cost information; under the confirmation triggering of the user end, generating a return tracking order for the returned goods based on the return intention order information, so as to perform return order management for the returned goods.

[0007] One or more embodiments of this specification provide a supply chain order management device, including: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above method.

[0008] One or more embodiments of the present specification provide a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to execute the above method.

[0009] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects: through the technical solutions in the embodiments of this specification, under the traditional method, consumers lack the right to choose the return warehouse and can only passively accept the return address specified by the merchant. They cannot choose the best solution according to their own situation. The embodiments of this specification take into account various factors and provide consumers with multiple return receiving warehouses and related information. Consumers can choose a return warehouse based on their own geographical location, requirements for return processing speed, and cost considerations. In the supply chain, different cooperation models have very different requirements for return processing. The embodiment of this specification determines the supply order cooperation model based on the goods identification, and then determines the target return information in combination with the saleable identification attribute of the returned goods. Traditional return management only performs subsequent processing after receiving the goods. It lacks preliminary analysis of the returned goods and the cooperation model, which easily leads to disorderly processing and backlog of returned goods. The embodiment of this specification identifies the saleable identification attribute of the returned goods in advance and formulates a return processing strategy in a targeted manner, thereby avoiding the long-term stay of returned goods in the warehouse and unnecessary inventory backlogs. Based on real-time tracking and analysis of returned goods, enterprises can dynamically adjust inventory according to the return situation. The dynamic inventory adjustment mechanism helps enterprises maintain reasonable inventory levels under different cooperation models and reduce the risk of inventory backlogs. Risk and inventory cost; in traditional return management, due to the lack of unified management and planning, duplication of work between departments is prone to occur. For example, when determining whether the returned goods are saleable, multiple departments may need to repeat inspection and confirmation. The embodiments of this specification avoid duplication of work between departments by establishing a unified return management process and information sharing mechanism, improve work efficiency, and reduce labor and time costs; the return logistics process is postponed. Before the return logistics, the return request is analyzed first, and the impact of user returns on suppliers and distributors under different cooperation models is fully considered. Optional warehouses under the corresponding cooperation model are provided to users. When selecting optional warehouses, the actual situation of suppliers or distributors in terms of inventory, cost, etc. is considered to ensure that the selected optional warehouses are the best return warehouses for suppliers or distributors. It also provides users with multiple warehouse options and allows them to freely choose according to the cost of the return. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. In the drawings: Figure 1 A flowchart of a supply chain order management method provided in an embodiment of this specification; Figure 2 A schematic diagram of the structure of a supply chain order management device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0011] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0012] The embodiments of this specification provide a supply chain order management method. It should be noted that the execution subject in the embodiments of this specification can be a server or any device with data processing capabilities. Figure 1 A flow chart of a supply chain order management method provided in an embodiment of this specification, such as Figure 1 As shown, it mainly includes the following steps: Step S101, receiving the user's return request information, and under the triggering of the return request information, identifying the returned goods according to the return request information, and determining the saleable identification attribute of the returned goods.

[0013] In one embodiment of the present specification, a user-friendly front-end interface is pre-set to facilitate users to submit return requests. The front-end interface includes multiple input areas, allowing users to enter user description information, such as a text input box, where users can describe the reason for the return (such as "scratches on the product" and "malfunction", etc.), and a file upload component can also be provided to facilitate users to upload image information of returned goods; in addition, an input box or drop-down menu is provided to facilitate users to enter or select a goods identification, which can be a unique identifier such as a product serial number, barcode, order number, etc., which helps to accurately identify the returned goods. The user's return request information is received through the back-end API interface, wherein the return request information includes user description information, returned goods image information and returned goods information, and the returned goods information includes a goods identification. The return and exchange image data can be provided by the user taking a photo or uploading a taken picture or video through a mobile terminal, which can intuitively display the appearance of the returned and exchanged goods; the goods identification can be a product serial number, barcode number, etc., which is used to accurately locate the goods involved.

[0014] According to the return request information, the returned goods are identified and the saleable identification attributes of the returned goods are determined, specifically including: using text extraction technology to determine the description problem type of the returned goods through the user description information in the return request information, wherein the description problem type includes appearance problems, functional problems and description discrepancies; according to the description problem type of the returned goods, in a pre-constructed model library, a corresponding goods quality recognition model is matched to identify the image information of the returned goods through the goods quality recognition model to determine the identification problem characteristics of the returned goods; the goods type of the returned goods is determined, and based on the goods type and the identification problem characteristics, similar cases are retrieved in a pre-constructed return case library to determine the saleable identification attributes of the returned goods, wherein the saleable identification attributes include any one of saleable, sale after repair, and non-saleable.

[0015] In one embodiment of the present specification, the text extraction technology herein may use a keyword extraction algorithm in natural language processing, such as a TF-IDF (term frequency-inverse document frequency) algorithm, to process the description information of returned goods. For example, for the user input "the product shell has scratches, and there is an abnormal sound after turning on the machine", the TF-IDF algorithm can identify keywords such as "scratches" and "abnormal sound" to facilitate a preliminary judgment of the problems existing in the product and determine the problem types corresponding to the keywords. The description problem types include appearance problems, functional problems, and description discrepancies. It should be noted that problems with the visible parts of the product, such as scratches, wear, deformation, stains, fading, etc., are appearance problems. Functional problems are problems involving whether the various functions of the product are operating normally, such as the screen of an electronic product is not bright, the buttons are malfunctioning, the software is faulty, the mechanical equipment is operating abnormally, and the power is insufficient. The description discrepancy problem refers to the inconsistency between the actual situation of the product and the description at the time of sale, such as color deviation, size discrepancy, material discrepancy, etc.

[0016] In one embodiment of the present specification, in order to accurately identify different types of problems, a model library needs to be built in advance. For appearance problems, an image recognition model based on a convolutional neural network (CNN) is constructed, and a large amount of product image data with appearance problem annotations (such as scratches, stains, etc.) is used for training, so that the model can learn the characteristic patterns of various appearance problems. Functional problems require the construction of an action recognition model based on deep learning, such as a 3D convolutional neural network (3D-CNN). It is trained through a large amount of video data with functional operation labels (such as normal startup, normal function switching, fault startup, fault function switching, etc.). These video data should contain the functional operation processes of various products in normal and faulty states, and the model learns the spatiotemporal characteristics of these operations, such as the movement trajectory of product parts during operation, the change law of indicator lights, etc. For the problem of description discrepancy, a model based on text and image matching is constructed, and the feature extraction and matching of the product description text and the actual product image is used to determine whether there is a description discrepancy. It should be noted that with the increase in product types, the diversification of problem types, and the changes in market feedback, the model library needs to be continuously maintained and updated, and new problem case data should be collected regularly to retrain the model to improve the accuracy and adaptability of the model. For example, when a new type of product appearance problem emerges, the relevant image data is added to the training set and the appearance problem recognition model is retrained so that it can recognize the new problem.

[0017] In one embodiment of the present specification, after determining the description problem type of the returned goods, the corresponding goods quality recognition model is searched in the model library. For example, if the description problem type is an appearance problem, a pre-trained appearance problem recognition model will be automatically selected; if it is a functional problem, the functional problem recognition model will be matched. The selected goods quality recognition model recognizes the returned goods image information. Taking the appearance problem recognition model as an example, the model will analyze the product in the image pixel by pixel, extract the texture, shape, color and other features of the image, and compare them with various appearance problem features stored in the model. For example, when identifying scratches, the model will look for line shapes, color changes, etc. in the image that meet the scratch characteristics. For functional problems, the returned goods image information can be a functional operation video composed of multiple goods images, and key frames are extracted from the functional operation video. A method based on inter-frame differences can be used to calculate the pixel differences between adjacent frames, and when the difference exceeds a certain threshold, the frame is marked as a key frame. In addition, it can also be combined with video content analysis, for example, according to the key steps of product function operation such as the start-up, key operation, function switching and other moments of electronic products to extract key frames. The extracted key frames and the time series information between key frames are input into the trained action recognition model. The model recognizes the functional operations in the video based on the learned normal and faulty operation modes. For example, for a power-on function problem of an electronic product, the model can determine whether the product is powered on normally by analyzing a series of key frames after the power-on button is pressed; if the model finds that the indicator light in the key frame does not light up according to the normal pattern, or the screen does not display the power-on screen normally, it can preliminarily determine that there is a power-on function problem.

[0018] Through the recognition process of the model, the identification problem features of the returned goods are determined. The identification problem features here include the specific type of problem (such as the length and location of the scratch, the specific manifestation of the functional failure, etc.), severity (such as minor, moderate, severe) and other information. For example, the appearance problem recognition model determines that there is a scratch on the surface of the product located in the upper left corner, with a length of 2 cm, and the scratch is shallow, which is a minor appearance problem; the functional problem recognition model determines that a chip of the electronic product is faulty, causing the product to be unable to start normally, which is a serious functional problem. Based on the type of goods and the identification problem features, similar cases are retrieved in the pre-built return case library to determine the saleable identification attributes of the returned goods. The saleable identification attributes include any one of saleable, sale after repair, and non-saleable.

[0019] Through the above technical solution, the text extraction technology is used to determine the type of description problem from the user description information, and then combined with the image information of the returned goods for analysis, avoiding the limitation of relying on a single information source for judgment. For example, it is difficult to determine the nature of the problem only from the user description "the product has a problem", but combined with the image information, if the image shows that the product shell has scratches, it can be accurately judged as an appearance problem. This multi-dimensional information fusion method makes the judgment of the problem of returned goods more comprehensive and accurate, thereby improving the accuracy of determining the attributes of the saleable identification.

[0020] According to different description problem types, the corresponding goods quality identification model in the pre-built model library is matched. Through this targeted model matching, the problem features can be extracted more effectively, which greatly improves the accuracy of identification compared with the general model. After determining the type of goods and identifying the problem features, the saleable identification attributes are determined by retrieving similar cases in the pre-built return case library, which provides a rich reference basis for the current processing of returned goods and makes the decision more scientific and reasonable. By combining the type of goods and the problem features to determine the saleable identification attributes, it can better adapt to the characteristics of different goods. Through text extraction and model matching, the problem type and characteristics of the returned goods can be quickly determined, and then similar cases can be quickly retrieved in the case library, and the saleable identification attributes and corresponding processing methods can be quickly determined. Compared with manual analysis and judgment one by one, the processing time is greatly shortened and the efficiency of return processing is improved. From text extraction, model matching to case retrieval, each link has clear operating specifications and standards, which improves processing efficiency, ensures the consistency and stability of processing results, and avoids processing differences caused by human factors.

[0021] Based on the type of goods and the identification problem characteristics, before searching for similar cases in a pre-constructed return case library to determine the saleable identification attribute of the returned goods, the method also includes: obtaining historical return records, wherein the historical return records include historical return cases of multiple types of goods, each of which includes a historical return reason, a historical return disposal method, and corresponding user evaluation information after the return disposal; based on the user evaluation information, screening the historical return records to determine a non-negative review return case set and a negative review case set; optimizing the historical goods handling method in the negative review case set to determine a negative review optimization case set, and constructing a return case library through the non-negative review return case set and the negative review optimization case set.

[0022] In one embodiment of the present specification, a historical return record is obtained, and the historical return record includes historical return cases of multiple types of goods. Each of the historical return cases includes historical return reasons, historical return disposal methods, and corresponding user evaluation information after return disposal. It should be noted that the user evaluation information here is not the user evaluation before the return, but the user evaluation after the return disposal. If the returned goods are of an unsalable type, the corresponding user evaluation is empty. The collected historical return records are stored according to a specific structure. A database table is created, and the fields in the table include the type of goods, historical return reasons, historical return disposal methods, and user evaluation information after return disposal. For example, for a batch of clothing returns, the "goods type" field is recorded as "clothing-tops", the "historical return reason" field may be "inappropriate size", the "historical return disposal method" field is recorded as "sale after size modification", and the "user evaluation information after disposal" field is "good quality, satisfied". Through this structured storage method, it is convenient to query, filter and analyze data efficiently.

[0023] Clearly define the criteria for distinguishing good and bad reviews, and set them from two aspects: keywords and ratings. For keywords, such as "satisfied" and "quickly resolved" and other related words that represent positive reviews, they can be used as the basis for judging non-bad reviews; while negative words such as "terrible", "unresolved problems", and "delay" tend to be bad reviews. In terms of rating, assuming a 5-point system, 2 points or more can be defined as non-bad reviews, and 2 points and below can be defined as bad reviews. For some semantically ambiguous reviews, natural language processing (NLP) technology is used for sentiment analysis. The evaluation text is processed using a sentiment analysis algorithm, and its sentiment tendency is determined by calculating the ratio of positive and negative sentiment words in the text, so as to accurately classify it. Use database query statements or data analysis tools to filter out non-bad and bad return cases respectively, and store the results to obtain a non-bad return case set and a bad case set.

[0024] For the negative review case set, the historical goods processing method in the negative review case set is optimized to determine the negative review optimization case set. It should be noted that the negative reviews in the negative review case set are used to represent the problems caused by the historical goods processing method in the return case after resale. For example, if consumers give negative reviews during the second sale of goods that have been processed and resold, it means that the quality of the resold goods cannot meet user requirements. Therefore, the corresponding historical goods processing method is optimized to be unsalable to ensure subsequent consumer feedback. Through the above non-negative review return and exchange case set and the negative review optimization case set, a return case library is constructed. The cases in the return case library obtained by the above method do not have the problem of users giving negative reviews due to the processing method.

[0025] Based on the goods type and the identification problem feature, similar case retrieval is performed in a pre-built return case library to determine the saleable identification attribute of the returned goods, specifically including: taking the goods type as the retrieval target, performing fine matching in the return case to determine the return case subset corresponding to the goods type; performing similarity calculation on the historical return reasons in the return case subset and the identification problem feature to determine multiple similarity results, and based on the similarity results, screening multiple similar return cases that meet preset requirements; determining the return processing method with the largest number through the corresponding return processing methods in the multiple similar return cases to determine the saleable identification attribute of the returned goods.

[0026] In one embodiment of the present specification, in the return case library, each case records the corresponding goods type information. After obtaining the type of the current returned goods, all cases matching the goods type are screened out through a database query statement or algorithm to form a return case subset corresponding to the goods type. Fine matching with the goods type as the search target can quickly narrow the search scope, determine the historical cases closely related to the current returned goods, and improve the search efficiency and accuracy. For example, the return reasons and handling methods of electronic products and clothing are quite different. Separately searching and analyzing the cases of the two can avoid interference from irrelevant cases and more accurately determine the handling methods. Even for the same type of goods, different return reasons may lead to different handling methods. By calculating the similarity between the historical return reasons and the current identification problem features, the historical cases closest to the current return situation are matched. A variety of text similarity calculation methods can be used, such as cosine similarity, edit distance, etc. The identification problem features of the current returned goods and the historical return reasons of each case in the return case subset are converted into a vector form (such as a word vector) that can be processed by a computer, and then the selected similarity calculation method is used to obtain the similarity result of each case with the current return situation. For example, when using cosine similarity calculation, the similarity is measured by calculating the cosine value of the angle between two vectors. The closer the value is to 1, the higher the similarity.

[0027] Based on the calculated similarity results, a preset requirement is set (such as a similarity threshold of 0.8), and multiple similar return cases with a similarity higher than the threshold are screened out. The screened cases are highly similar to the current returned goods in terms of return reasons and problem characteristics. In multiple similar return cases, different cases may correspond to different return processing methods. By counting the number of various return processing methods in similar return cases, the principle of majority voting is adopted to determine the saleable identification attribute corresponding to the return processing method with the largest number of appearances as the saleable identification attribute of the current returned goods. For example, among the 10 similar return cases screened out, 6 cases are handled as "sale after repair", 3 are "saleable", and 1 is "unsaleable", then the saleable identification attribute of the current returned goods is determined to be "sale after repair".

[0028] Through the above technical solution, the type of goods is used as the search target for fine matching, and a subset of cases directly related to it can be quickly screened out from the huge return case library. Due to the large differences in the return processing of different types of goods, classified retrieval avoids the interference of irrelevant cases, greatly reduces the amount of data to be processed, and significantly improves the retrieval efficiency; focusing on cases of the same type of goods can more accurately find historical cases closely related to the current return goods; even for the same type of goods, different return reasons will lead to different processing methods. Through similarity calculation, the historical cases that are closest to the current return situation in terms of problem nature can be found, and the return reasons and problem characteristics are converted into vector form for quantitative analysis, making the measurement of similarity more scientific and objective; based on the similarity calculation results, reasonable preset requirements are set (such as a similarity threshold of 0.8) to screen similar return cases, ensuring that the screened cases are highly similar to the current return goods in terms of return reasons and problem characteristics; in multiple similar return cases, the majority voting principle is used to determine the saleable identification attributes of the returned goods, which can effectively avoid the particularity of a single case from having too much impact on the decision. By counting the number of various return processing methods, the saleable identification attribute corresponding to the most frequently occurring processing method is used as the result, which is more stable and universal.

[0029] Step S102, based on the goods identification, determining the supply order cooperation model corresponding to the returned goods, and determining the target return information of the returned goods based on the supply order cooperation model and the saleable identification attribute.

[0030] In one embodiment of the present specification, in the supply chain management database, a cooperation mode mark is associated with each supply order, and a special field (such as the "cooperation mode" field) can be used here to record whether it is "buyout sales" or "agent sales". At the same time, the database table should also contain other order-related information, such as order number, supplier information, distributor / agent information, etc. According to the supply order cooperation mode and saleable identification attribute corresponding to the returned goods, the target return information of the returned goods is determined, wherein the target return information includes at least one return receiving warehouse.

[0031] Based on the supply order cooperation model and the saleable identification attribute, the target return information of the returned goods is determined, specifically including: according to the supply order cooperation model, a rule search is performed in a preset return processing rule library to determine the return goods processing rule corresponding to the saleable identification attribute, wherein the return goods processing rule includes a goods processing subject, and the goods processing subject includes any one of a supplier and a distributor; at least one return receiving warehouse is determined through the return goods processing rule and the real-time goods data of the goods processing subject acquired in advance, so as to generate the target return information of the returned goods based on the at least one return receiving warehouse.

[0032] In one embodiment of the present specification, a rule engine is constructed to automatically infer target return information based on the supply order cooperation mode and the saleable identification attribute. The rule engine can use an existing business rule management system (BRMS) or develop a rule-based algorithm on its own. A cooperation mode rule base is established to store general rules for return processing under different cooperation modes. For example, in a buyout sales mode, the rule may include "the distributor bears the inventory risk, and the distributor's internal allocation is given priority for saleable goods"; in an agent sales mode, the rule is "the supplier is responsible for the final processing of the goods, and the agent is mainly responsible for assisting the return process." The rules in the cooperation mode rule base and the saleable identification attribute processing rule base are loaded into the rule engine. For example, the rule engine loads the rule "when the buyout sale is saleable (no processing required) , the distributor's internal inventory allocation is given priority." When a return request is received, the rule engine first obtains the supply order cooperation mode and the saleable identification attribute from the database. For example, the cooperation mode is obtained as "buyout sale" and the saleable identification attribute is "saleable (simple processing required)". The rule engine makes inferences based on the information obtained and the loaded rules. In the above example, it searches for the rules for "saleable (simple processing required)" products under buyout sales. If the rule is "If the distributor has the processing capacity, the product will be returned to the distributor's warehouse for processing; otherwise, negotiate a repurchase with the supplier", the rule engine will then query the distributor's processing capacity data (stored in the distributor information table). If the distributor has the processing capacity, the rule engine automatically determines that the target return information is returned to the warehouse with processing capacity specified by the distributor, that is, the goods processing subject in the return goods processing rules is the distributor; if it does not have the processing capacity, the rule engine may generate a prompt message to negotiate a repurchase with the supplier, and preliminarily determine that the target return information is returned to the repurchase receiving warehouse specified by the supplier (the warehouse information can be obtained from the supplier information table), that is, the goods processing subject in the return goods processing rules is the supplier.

[0033] At least one return receiving warehouse is determined through the return goods processing rules and the real-time goods data of the goods processing entity obtained in advance, specifically including: obtaining the real-time goods data, the real-time goods data including the real-time inventory data, sales outbound speed and warehouse transportation cost data of each warehouse; quantifying the return receiving capacity of each warehouse according to the real-time goods data and the saleable identification attribute, and determining the return receiving parameters of each warehouse; selecting a specified number of warehouses in order of the return receiving parameters from low to high to determine at least one return receiving warehouse.

[0034] In one embodiment of the present specification, the real-time cargo data is obtained, and the real-time cargo data includes the real-time inventory data, sales outbound speed and warehouse transportation cost data of each warehouse. The real-time inventory data of each warehouse is usually stored in the inventory management system of the enterprise, and the inventory quantity is updated in real time through the warehouse entry and exit records, such as using barcode scanning or radio frequency identification (RFID) technology to track the entry and exit of goods. Whenever there is goods entering or leaving the warehouse, the inventory management system automatically updates the inventory records of the corresponding warehouse and products. The sales outbound speed data can be obtained from the sales record system, which records the sales time, quantity and other information of each warehouse. By analyzing the sales data within a period of time (such as the past week or month), the average sales outbound speed of each warehouse can be calculated. For example, warehouse A sold a total of 100 pieces of a certain product in the past month, with an average daily sales outbound of about 3.3 pieces, which is the sales outbound speed of warehouse A for the product. The warehouse transportation cost data comes from the logistics management system, which records the transportation costs from different shipping places (including distributor warehouses, supplier warehouses, etc.) to various destinations (such as other warehouses, customer locations, etc.). The transportation cost may be affected by factors such as distance, mode of transportation (such as road transportation, rail transportation, etc.), weight and volume of goods, etc. For example, the transportation cost from the supplier warehouse to the distributor warehouse B is 5 yuan per kilogram, which is stored in the logistics management system and updated regularly according to the actual situation.

[0035] After the cargo handling entity is determined, there are multiple warehouses for both suppliers and distributors. The warehouse here refers to the warehouse of the cargo handling entity. According to the real-time cargo data of the warehouse and the saleable identification attribute, the return receiving capacity of each warehouse is quantified, and the return receiving parameters of each warehouse are determined. According to the order of the return receiving parameters from low to high, a specified number of warehouses are selected in turn to determine at least one return receiving warehouse. The specified number here can be set to 3. The higher the receiving parameter, the weaker the corresponding receiving capacity. Therefore, screening out a specified number of warehouses from low to high can ensure that the warehouse with the strongest receiving capacity is selected for the return user.

[0036] According to the real-time cargo data and the saleable identification attribute, the return receiving capacity of each warehouse is quantified, and the return receiving parameters of each warehouse are determined, specifically including: determining the inventory margin of each warehouse according to the real-time inventory data, and determining the inventory level index of the warehouse based on the ratio of the inventory margin and the safety stock of the warehouse; determining the average outbound speed of multiple warehouses through the sales outbound speed of each warehouse, and determining the outbound level index of the warehouse based on the sales outbound speed and the average outbound speed; determining the average transportation cost of multiple warehouses based on the warehouse transportation cost data of each warehouse, and determining the transportation level index of the warehouse based on the ratio of the warehouse transportation cost data and the average transportation cost; matching the weight combination through the saleable identification attribute, and quantifying the return receiving capacity of each warehouse based on the weight combination and the inventory level index, the outbound level index and the transportation level index, and determining the return receiving parameters of each warehouse.

[0037] In one embodiment of the present specification, the inventory margin of each warehouse is determined according to the difference between the safety stock corresponding to the safety stock of the warehouse and the real-time inventory data, and the inventory level index of the warehouse is determined by the ratio of the inventory margin to the safety stock of the warehouse. For example, the ratio of the inventory margin to the safety stock of warehouse A is 30%, while the ratio of the inventory margin to the safety stock of warehouse B is 80%. In this case, warehouse A may have more advantages in receiving returned goods in terms of inventory replenishment, that is, the smaller the inventory level index, the stronger the corresponding receiving capacity.

[0038] Through the sales outbound speed of each warehouse, the average outbound speed of multiple warehouses corresponding to the cargo handling entity is determined. First, the ratio of the sales outbound speed to the average outbound speed is calculated, and the outbound level index of the warehouse is determined in the form of 1-ratio. It should be noted that the greater the sales outbound speed, the greater the ratio, and the smaller the corresponding outbound level index, indicating that the receiving capacity is stronger. Based on the warehouse transportation cost data of each warehouse, the average transportation cost of multiple warehouses is determined, and the transportation level index of the warehouse is determined based on the ratio of the warehouse transportation cost data and the average transportation cost. The smaller the warehouse transportation cost, the smaller the corresponding transportation level index, and the stronger the corresponding receiving capacity.

[0039] In one embodiment of the present specification, the saleable identification attribute is used to match the weight combination, and the sale identification attribute reflects the different states of the returned goods, such as "saleable", "sale after repair", and "unsaleable". Returned goods in different states have different requirements for warehouses, so it is necessary to match them with corresponding weight combinations to reasonably evaluate the warehouse's ability to receive different types of returned goods. For "saleable" returned goods, inventory turnover speed and sales and outbound capabilities are crucial because they need to be quickly put back on the shelves for sale. At the same time, transportation costs will also affect the overall benefits. The weight combination is determined as follows: the inventory level indicator weight (W1) is set to 0.4, the outbound level indicator weight (W2) is set to 0.4, and the transportation level indicator weight (W3) is set to 0.2. Selling goods after repair requires the warehouse to have a certain amount of storage and repair processing space, and transportation costs must also be considered. Therefore, the weight combination is set as follows: the weight of the inventory level indicator (W1) is 0.3, the weight of the outbound level indicator (W2) is 0.3, and the weight of the transportation level indicator (W3) is 0.4. The goods need to be stored for a period of time after repair, and the importance of the inventory level is relatively reduced, while the transportation cost has a greater impact on the transportation from the repair site to the warehouse. For "unsaleable" goods, the warehouse is mainly used for temporary storage and subsequent processing, and more attention is paid to the storage cost and the convenience of the processing flow. The weight combination is: the weight of the inventory level indicator (W1) is 0.2, the weight of the outbound level indicator (W2) is 0.2, and the weight of the transportation level indicator (W3) is 0.6. Such goods will no longer be sold, and the outbound speed is relatively unimportant, while the impact of transportation cost on the processing cost is more prominent. Based on the weight combination and the inventory level indicator, the outbound level indicator and the transportation level indicator, the return receiving capacity of each warehouse is weighted and summed to determine the return receiving parameters of each warehouse. According to the order of the return receiving parameters from low to high, a specified number of warehouses are selected in turn to determine at least one return receiving warehouse. The specified quantity here can be set to 3. The higher the receiving parameter, the weaker the corresponding receiving capacity. Therefore, filtering out a specified number of warehouses from low to high can ensure that the warehouse with the strongest receiving capacity is available for selection by the return user.

[0040] Through the above technical solution, by calculating the inventory level index, outbound level index and transportation level index respectively, the warehouse situation in terms of inventory, sales and transportation costs is comprehensively considered, which can accurately reflect the receiving capacity of each warehouse for returned goods; according to the different saleable identification attributes of returned goods, different weight combinations are matched, and the targeted weight setting makes the evaluation of warehouse receiving capacity more reasonable and accurate; a specified number of warehouses are selected in order from low to high in return receiving parameters. Since the higher the receiving parameter, the weaker the receiving capacity, this can ensure that the warehouse with the strongest receiving capacity is selected for return users to choose; help enterprises to reasonably allocate return goods to the most suitable warehouse according to the characteristics of return goods and the actual situation of the warehouse, optimize inventory management and logistics resource allocation in the supply chain; provide return users with the choice of warehouses with the strongest receiving capacity, and the return process is smoother and more efficient.

[0041] Step S103, generating return intention order information through the target return information of the returned goods and the pre-acquired return rules, and sending the return intention order information to the user end.

[0042] The return intention order information includes user return cost information; Generate return intention order information through the target return information of the return goods and the pre-acquired return rules, specifically including: according to the target return information of the return goods, obtain the warehouse receiving information of each return receiving warehouse, wherein the warehouse receiving information includes warehouse receiving logistics and warehouse location information; generate the return logistics cost information corresponding to each return receiving warehouse through the warehouse receiving information of each return receiving warehouse and the pre-acquired user location information; obtain the return fee association rule in the return rule, and determine the user return cost information corresponding to each return receiving warehouse based on the return fee association rule and the return logistics cost information; generate the return intention order information corresponding to the return goods according to the user return cost information corresponding to each return receiving warehouse.

[0043] In one embodiment of the present specification, the target return information includes the warehouse receiving information of the return receiving warehouse. From the determined return receiving warehouse, further obtain detailed receiving information of each warehouse, including warehouse receiving logistics and warehouse location information. Warehouse receiving logistics information may cover the logistics method, logistics partner, logistics receiving time window, etc. adopted by the warehouse to accept the returned goods, which determines how the returned goods are received by the warehouse. The warehouse location information clarifies the geographical location of the warehouse and is used to calculate the logistics cost. The user location information is determined by the delivery address filled in by the user when placing an order, or the return address provided in the return application. The receiving information of each return receiving warehouse is combined with the user location information. Using the logistics cost calculation method, the warehouse location, user location, warehouse receiving logistics method (such as express delivery, logistics dedicated line, etc.), cargo weight, volume and other factors are comprehensively considered to generate the return logistics cost information corresponding to each return receiving warehouse. For example, if the user is located in city A and warehouse B is located in city B, the logistics cost calculation system takes into account the distance between the two places, the unit price of the selected logistics method, and the weight of the returned goods, and the logistics cost of returning the goods from the user to warehouse B is X yuan. The return rules are a series of rules for return processing formulated in advance by the enterprise. Among them, the return cost association rules clarify how to calculate and collect other return-related costs in addition to logistics costs, such as return handling fees, goods inspection fees, depreciation fees (if applicable) and other costs related to return behavior, as well as the calculation method of logistics costs such as the proportion of return logistics costs. According to the return cost association rules, the return logistics cost information is comprehensively calculated with other related costs to determine the user return cost information corresponding to each return receiving warehouse. For example, the return cost association rules stipulate that for the return of a certain type of goods, in addition to the logistics cost, a return handling fee of 5% of the value of the goods is required. Assuming that the value of the returned goods is 1,000 yuan and the logistics cost is 50 yuan, then the user return cost = 50 + 1,000 × 5% = 100 yuan. The user return cost information corresponding to each return receiving warehouse is summarized. Based on this sorted information, the return intention order information corresponding to the returned goods is generated. The return intention order information includes the user return cost corresponding to each return receiving warehouse, as well as the return goods information (such as name, specification, quantity, etc.), return reason, expected return processing time, etc., providing users with a complete return intention. Users can clearly understand the cost situation under different return receiving warehouse selections based on the return intention order information. For example, the return intention order information shows that the returned goods are a mobile phone of a certain brand, the reason for the return is a functional failure, warehouse A can be selected, the user return cost is 150 yuan, and the expected processing time is 3 working days; warehouse B can be selected, the user return cost is 130 yuan, and the expected processing time is 5 working days, etc.

[0044] In traditional return management, consumers often lack a clear understanding of return costs, which increases the uncertainty of shopping decisions. When generating return intention order information, the embodiments of this specification list the user's return cost information in detail. When initiating a return, consumers can clearly know the logistics costs, possible handling fees, testing fees and other expenses required for the return. The return costs under different return receiving warehouse selections are clearly presented to users to avoid dissatisfaction with orders caused by unclear costs. Multiple return receiving warehouses and their corresponding information are provided, allowing users to choose according to their own needs and preferences. For example, cost-sensitive users can choose warehouses with low return costs; time-sensitive users can choose warehouses with short expected processing time, fully respecting the user's wishes. According to different warehouse receiving information and user choices, return resources are reasonably allocated to improve resource utilization efficiency. When generating return logistics cost information, factors such as warehouse receiving logistics methods, locations and user locations are considered, which helps to plan logistics resources in advance. Enterprises can reasonably arrange warehouse resources, such as storage space and manpower, according to the user's choice of different warehouses.

[0045] Step S104, under the confirmation trigger of the user end, based on the return intention order information, a return tracking order for the returned goods is generated to manage the return order for the returned goods.

[0046] In one embodiment of the present specification, after the return intention order information is determined, the return intention order information is displayed in a clear and intuitive manner on the user-side interface, including the return goods information (such as name, specification, quantity, etc.), the user return and exchange cost corresponding to each return receiving warehouse, the reason for return, and the estimated return processing time, etc. A "Confirm Return" button is set on the interface, and when the user clicks the button, a confirmation event is triggered. This trigger mechanism ensures that the return process will only be further promoted after the user fully understands the content of the return intention order and actively confirms it, fully respecting the user's right to choose independently. Use JavaScript to add a click event listener to the button, and in the event processing function, send key information such as the return receiving warehouse selected by the user to the back end.

[0047] After receiving the user's confirmation request, the backend server first extracts key data from the return intention order information, including the return receiving warehouse information selected by the user, as well as the detailed information of the returned goods, the reason for the return, etc. At the same time, the server will also obtain additional necessary information from other related systems (such as the order management system, user information system, etc.), such as the user's full address, contact information, etc., to ensure that the generated return tracking order information is complete and accurate. It should be noted that compared with the return intention order, the warehouse in the return tracking order is the return receiving warehouse selected by the user.

[0048] Generate a unique order ID for each return tracking order, which is used to track and identify the order throughout the return process. It can be generated based on timestamps, random numbers, or other algorithms to ensure its uniqueness and traceability in the system. Build the details of the return tracking order based on the extracted and integrated data. The order content usually includes the order number, return goods information, return receiving warehouse information, user information, return reason, estimated processing time, and initial order status (such as "confirmed, waiting for shipment", "in transit", "received", "processing", "processing completed", etc.).

[0049] In one embodiment of the present specification, a complete return order status management system is established to track and record the entire life cycle of the order. When the order status changes, the record in the database is updated in a timely manner, and the latest status is fed back to the user end and relevant departments within the enterprise. For example, when the logistics system updates the cargo transportation status to "shipped", the status of the return tracking order will also be updated from "confirmed, waiting to be shipped" to "in transit". In addition, users can enter the order number or use other related information (such as return cargo identification, user account, etc.) through the user interface to query the order status. Internal personnel of the enterprise (such as customer service personnel, warehouse management personnel, logistics dispatch personnel, etc.) can also query and statistically analyze the return tracking order according to various conditions (such as order status, time range, return receiving warehouse, etc.) through a special management system. In addition, the system can also set up an automatic feedback mechanism. For example, when there is a major change in the order status (such as the warehouse has received the goods, the processing is completed, etc.), the user is notified through SMS, email or user-side push messages to improve the user experience. Return order management is not only about tracking the order status, but also involves process collaboration between multiple departments within the enterprise and external partners (such as logistics suppliers). For example, the warehouse management department prepares to receive the return goods according to the return tracking order information, the logistics department arranges transportation and distribution according to the order, and the quality inspection department conducts quality inspection after the goods arrive at the warehouse.

[0050] Through the above technical solution, after confirming the return intention order information, the user triggers the generation of the return tracking order, and can clearly know that the return process has been officially started. By tracking the order, the user can understand the status of the returned goods in real time; if there is a problem in the return process, the user can trace and protect his rights based on the tracking order information. At the same time, a good return experience can enhance the user's favorability and loyalty to the company and increase the possibility of re-purchase; through the effective management of the return tracking order, the company can grasp the return dynamics in a timely manner and respond and solve the problems that arise quickly; the return tracking order information is shared in all links of the supply chain, strengthening the coordination of suppliers, logistics companies, warehouses and other parties. Logistics companies arrange transportation plans based on orders to ensure that the returned goods are delivered to the designated warehouse in time; warehouses make preparations for receiving according to orders to improve the efficiency of goods circulation; suppliers can understand the return situation through tracking orders, provide a basis for replenishment or product improvement, and enable the entire supply chain to collaborate efficiently in return processing.

[0051] Through the technical solutions in the embodiments of this specification, in the traditional way, consumers lack the right to choose the return warehouse and can only passively accept the return address specified by the merchant, and cannot choose the best solution according to their own situation. The embodiments of this specification provide consumers with multiple return receiving warehouses and related information by comprehensively considering multiple factors. Consumers can choose the return warehouse independently according to their own geographical location, requirements for return processing speed, and cost considerations. In the supply chain, different cooperation models have very different requirements for return processing. The embodiments of this specification determine the supply order cooperation model based on the goods identification, and then determine the target return information in combination with the saleable identification attribute of the returned goods. Traditional return management only performs subsequent processing after receiving the goods, lacks preliminary analysis of the returned goods and the cooperation model, and easily leads to disorderly processing and backlog of returned goods. The embodiments of this specification identify the saleable identification attribute of the returned goods in advance, and formulate return processing strategies in a targeted manner, avoiding the long-term stay of returned goods in the warehouse and unnecessary inventory backlogs. Based on real-time tracking and analysis of returned goods, enterprises can dynamically adjust inventory according to the return situation. The dynamic inventory adjustment mechanism helps enterprises maintain reasonable inventory levels under different cooperation models and reduce the risk of inventory backlogs. Risk and inventory cost; in traditional return management, due to the lack of unified management and planning, duplication of work between departments is prone to occur. For example, when determining whether the returned goods are saleable, multiple departments may need to repeat inspection and confirmation. The embodiments of this specification avoid duplication of work between departments by establishing a unified return management process and information sharing mechanism, improve work efficiency, and reduce labor and time costs; the return logistics process is postponed. Before the return logistics, the return request is analyzed first, and the impact of user returns on suppliers and distributors under different cooperation models is fully considered. Optional warehouses under the corresponding cooperation model are provided to users. When selecting optional warehouses, the actual situation of suppliers or distributors in terms of inventory, cost, etc. is considered to ensure that the selected optional warehouses are the best return warehouses for suppliers or distributors. It also provides users with multiple warehouse options and allows them to freely choose according to the cost of the return.

[0052] The embodiment of this specification also provides a supply chain order management device, such as Figure 2 As shown, the device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above method.

[0053] The embodiments of the present specification also provide a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to execute the above method.

[0054] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0055] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0056] The devices and media provided in the embodiments of this specification correspond one-to-one to the methods. Therefore, the devices and media also have similar beneficial technical effects as the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0057] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0058] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0059] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0060] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0061] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0062] Memory may include non-permanent storage in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0063] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0064] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0065] The above description is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of this specification.

Claims

1. A supply chain order management method, characterized in that: The method comprises: Receive a user's return request information, and under the trigger of the return request information, identify the returned goods according to the return request information, and determine the saleable identification attribute of the returned goods, wherein the return request information includes user description information, returned goods image information and returned goods information, and the returned goods information includes goods identification; Based on the goods identification, determining the supply order cooperation mode corresponding to the returned goods, so as to determine the target return information of the returned goods based on the supply order cooperation mode and the saleable identification attribute; Generate return intention order information through the target return information of the returned goods and the pre-acquired return rules, and send the return intention order information to the user end, wherein the return intention order information includes user return cost information; Under the confirmation trigger of the user terminal, a return tracking order for the returned goods is generated based on the return intention order information to perform return order management on the returned goods.

2. A supply chain order management method according to claim 1, characterized in that: According to the return request information, the returned goods are identified and the saleable identification attributes of the returned goods are determined, specifically including: Using text extraction technology, the description problem type of the returned goods is determined based on the user description information in the return request information, wherein the description problem type includes appearance problems, functional problems, and description discrepancies; According to the description problem type of the returned goods, a corresponding goods quality recognition model is matched in a pre-built model library, so as to recognize the image information of the returned goods through the goods quality recognition model and determine the recognition problem characteristics of the returned goods; Determine the type of the returned goods, and based on the type of goods and the identification problem features, perform a similar case search in a pre-built return case library to determine a saleable identification attribute of the returned goods, wherein the saleable identification attribute includes any one of saleable, saleable after repair, and non-saleable.

3. A supply chain order management method according to claim 2, characterized in that: Based on the type of goods and the identification problem characteristics, similar cases are searched in a pre-built return case library to determine the saleable identification attributes of the returned goods, specifically including: Taking the goods type as the search target, performing fine matching on the return cases to determine a subset of return cases corresponding to the goods type; Calculate the similarity between the historical return reasons and the identified problem features in the return case subset to determine a plurality of similarity results, and screen a plurality of similar return cases that meet preset requirements based on the similarity results; The return processing method with the largest number is determined through the corresponding return processing methods in the multiple similar return and exchange cases to determine the saleable identification attribute of the returned goods.

4. A supply chain order management method according to claim 2, characterized in that: Based on the type of goods and the identification problem characteristics, before searching for similar cases in a pre-built return case library to determine the saleable identification attribute of the returned goods, the method further includes: Obtaining historical return records, wherein the historical return records include historical return cases of multiple types of goods, and each of the historical return cases includes historical return reasons, historical return disposal methods, and corresponding user evaluation information after the return disposal; According to the user evaluation information, screening is performed in the historical return records to determine a non-negative review return case set and a negative review case set; The historical goods handling methods in the negative review case set are optimized to determine the negative review optimization case set, and a return case library is constructed through the non-negative review return and exchange case set and the negative review optimization case set.

5. A supply chain order management method according to claim 1, characterized in that: Determining target return information of the returned goods based on the supply order cooperation model and the saleable identification attribute specifically includes: According to the supply order cooperation mode, a rule search is performed in a preset return processing rule library to determine a return goods processing rule corresponding to the saleable identification attribute, wherein the return goods processing rule includes a goods processing subject, and the goods processing subject includes any one of a supplier and a distributor; At least one return receiving warehouse is determined through the return goods processing rules and the real-time goods data of the goods processing entity acquired in advance, so as to generate the target return information of the return goods based on the at least one return receiving warehouse.

6. A supply chain order management method according to claim 5, characterized in that: Determining at least one return receiving warehouse by using the return goods processing rules and the pre-acquired real-time goods data of the goods processing subject, specifically includes: Acquire the real-time cargo data, which includes real-time inventory data, sales delivery speed and warehouse transportation cost data of each warehouse; quantifying the return receiving capacity of each warehouse according to the real-time cargo data and the saleable identification attribute, and determining the return receiving parameters of each warehouse; A specified number of warehouses are selected in order from low to high order of the return receiving parameters to determine at least one return receiving warehouse.

7. A supply chain order management method according to claim 6, characterized in that: According to the real-time cargo data and the saleable identification attribute, the return receiving capacity of each warehouse is quantified, and the return receiving parameters of each warehouse are determined, specifically including: Determine the inventory margin of each warehouse according to the real-time inventory data, so as to determine the inventory level index of the warehouse based on the ratio of the inventory margin to the safety stock of the warehouse; Determine the average outbound speed of the plurality of warehouses through the sales outbound speed of each warehouse, and determine the outbound level index of the warehouse based on the sales outbound speed and the average outbound speed; Based on the warehouse transportation cost data of each of the warehouses, determining an average transportation cost of a plurality of the warehouses, and determining a transportation level indicator of the warehouse based on a ratio of the warehouse transportation cost data and the average transportation cost; The salable identification attribute is matched with a weight combination to quantify the return receiving capacity of each warehouse based on the weight combination and the inventory level index, the outbound level index and the transportation level index, so as to determine the return receiving parameters of each warehouse.

8. A supply chain order management method according to claim 1, characterized in that: Generate return intention order information based on the target return information of the returned goods and the pre-acquired return rules, specifically including: According to the target return information of the returned goods, the warehouse receiving information of each return receiving warehouse is obtained, wherein the warehouse receiving information includes warehouse receiving logistics and warehouse location information; Generate return logistics cost information corresponding to each return receiving warehouse through warehouse receiving information of each return receiving warehouse and pre-acquired user location information; Acquire the return cost association rule in the return rule, so as to determine the user return cost information corresponding to each return receiving warehouse based on the return cost association rule and the return logistics cost information; According to the user return cost information corresponding to each of the return receiving warehouses, the return intention order information corresponding to the returned goods is generated.

9. A supply chain order management device, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the method according to any one of claims 1 to 8.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured to execute the method according to any one of claims 1 to 8.

Images