Supply chain logistics information updating and pushing method, system and platform

Abstract

The invention relates to a supply chain logistics information updating and pushing method, system and platform, and relates to the technical field of supply chains, and the method comprises the steps: obtaining the accumulated transportation time and the latest logistics information of a target batch of goods in transportation in supply chain logistics, and extracting and obtaining a delay node and delay time when logistics information pushing delay occurs; obtaining the demand updating frequency and the logistics query frequency of a plurality of users for the same kind of goods; according to the plurality of demand update frequencies, the plurality of logistics query frequencies and the delay time, configuring logistics prediction parameters, performing arrival time prediction of the delay node, and obtaining predicted logistics node information; according to the multiple demand updating frequencies, the multiple logistics query frequencies and the delay time, error identification is conducted on the predicted logistics node information, multiple pieces of logistics information are generated, and updating pushing is conducted. The technical problems that in the prior art, supply chain logistics information pushing is poor in timeliness, and negative influences are generated on the overall efficiency of a supply chain are solved.

Description

Technical Field

[0001] The present invention relates to the field of supply chain technology, and in particular to a supply chain logistics information update push method, system and platform. Background Art

[0002] The product logistics supply of multiple levels of suppliers and users is designed within the supply chain. The timely push of logistics information can enable multiple levels of users to understand the product distribution situation and then formulate corresponding sales and usage strategies. The timeliness of logistics information is crucial to corporate operations and customer experience.

[0003] In the existing technology, the push of logistics information is delayed due to factors such as data collection delays and network transmission problems in the supply chain, which makes it impossible for users to obtain the logistics information of products in a timely manner, affecting inventory management, transportation scheduling and customer satisfaction. Therefore, the push of supply chain logistics information in the existing technology has the technical problem of poor timeliness and negative impact on the overall efficiency of the supply chain. Summary of the invention

[0004] The present invention aims to solve the technical problems in the prior art that the push of supply chain logistics information is ineffective and has a negative impact on the overall efficiency of the supply chain, and proposes a supply chain logistics information update push method, system and platform.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: In a first aspect, the present invention provides a supply chain logistics information update and push method, comprising: in supply chain logistics, obtaining the cumulative transportation time and the latest logistics information of a target batch of goods in transportation, and when a logistics information push delay occurs, extracting the delay node and delay time; obtaining the demand update frequency and logistics query frequency of multiple users associated with the target batch of goods for the same type of goods; configuring logistics prediction parameters according to multiple demand update frequencies, multiple logistics query frequencies and delay times, predicting the arrival time of the delay node, and obtaining predicted logistics node information; respectively performing error identification on the predicted logistics node information according to multiple demand update frequencies, multiple logistics query frequencies and delay times, generating multiple logistics information, and performing update and push.

[0006] In a second aspect, the present invention provides a supply chain logistics information update and push system, including: a delay information acquisition module, used to obtain the cumulative transportation time and the latest logistics information of a target batch of goods in transportation in the supply chain logistics, and extract the delay node and delay time when a logistics information push delay occurs; a user information acquisition module, used to obtain the demand update frequency and logistics query frequency of multiple users associated with the target batch of goods for the same type of goods; a logistics information prediction module, used to configure logistics prediction parameters according to multiple demand update frequencies, multiple logistics query frequencies and delay times, predict the arrival time of the delay node, and obtain predicted logistics node information; a logistics information push module, used to respectively identify errors in the predicted logistics node information according to multiple demand update frequencies, multiple logistics query frequencies and delay times, generate multiple logistics information, and push updates.

[0007] In a third aspect, the present invention provides a supply chain logistics information update push platform, including a supply chain logistics information update push system in the second aspect.

[0008] The beneficial effects of the present invention are as follows: the present invention can monitor the transportation status of goods in real time by obtaining the cumulative transportation time and the latest logistics information of the target batch of goods in transportation, and provide accurate basic data for logistics information push. When there is a delay in the push of logistics information, the delay node and delay time can be automatically extracted, the abnormal logistics status can be accurately identified, and the impact of information lag on user decision-making can be avoided. Further combined with the demand update frequency and logistics query frequency of multiple users for similar goods, the logistics information can be predicted and pushed to meet the needs of different users. The present invention also configures logistics prediction parameters to predict logistics nodes. When the logistics data is lagging, the historical data and machine learning algorithm can be used to infer the time when the goods arrive at the delay node, so that the logistics information push no longer depends on a single logistics data source, and the uncertainty caused by information delay is effectively reduced. Combined with the error identification mechanism, it is ensured that the pushed information is more valuable for reference, and the user experience and supply chain management efficiency are improved. In summary, the present invention can effectively solve the problem of delayed logistics information push in the prior art and improve the timeliness of logistics information. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic diagram of a flow chart of a supply chain logistics information update and push method provided by the present invention.

[0010] Figure 2 A schematic diagram of the structure of a supply chain logistics information update push system provided by the present invention.

[0011] Figure numerals: delay information acquisition module 11, user information acquisition module 12, logistics information prediction module 13, logistics information push module 14. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0013] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0014] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0015] Embodiment 1, as Figure 1 As shown, an embodiment of the present invention provides a supply chain logistics information update push method, the method specifically includes the following steps: S10: In supply chain logistics, the cumulative transportation time and the latest logistics information of the target batch of goods in transportation are obtained, and when there is a delay in pushing logistics information, the delay node and delay time are extracted.

[0016] In the embodiment of the present application, in supply chain logistics management, in order to ensure the accuracy and real-time nature of logistics information, it is first necessary to obtain the cumulative transportation time and the latest logistics information of the target batch of goods in transportation.

[0017] Based on the accumulated transportation time and the latest logistics information, it is determined whether the logistics information of the target batch of goods is pushed in time. When there is a delay in pushing the logistics information, the delay node and delay time are extracted.

[0018] Step S30 in the method provided in the embodiment of the present application includes: In supply chain logistics, obtain the transportation route of the target batch of goods in transit, and monitor and obtain the cumulative transportation time and the latest logistics information of the target batch of goods according to the preset monitoring frequency; According to the transportation route and accumulated transportation time of the target batch of goods, the expected logistics information is retrieved and obtained, the latest logistics information and the expected logistics information are distinguished and verified, and when the latest logistics information is delayed, the delay time is calculated and the expected logistics information is used as the delay node; When there is no delay in the latest logistics information, no subsequent processing will be performed.

[0019] In the embodiment of the present application, in supply chain logistics management, in order to monitor the transportation status of the target batch of goods in real time, it is first necessary to obtain the transportation route, that is, all the logistics nodes that the goods pass through from the starting point (such as the supplier's warehouse or production plant) to the destination (such as the customer's warehouse, retail terminal). The transportation route is usually composed of multiple key nodes, such as "supplier warehouse-A transfer center-B distribution station-wholesaler warehouse-C transfer center-D distribution station-retailer warehouse-E transfer center-F distribution station-end user". By pushing the overall logistics information of the supply chain, the overall supply chain efficiency and information transparency can be improved.

[0020] Furthermore, according to the preset monitoring frequency, the cumulative transportation time and the latest logistics information of the target batch of goods are monitored and obtained. The preset monitoring frequency refers to the regular data collection interval preset by the logistics system, such as updating the logistics status every 30 minutes or every hour, so as to regularly monitor whether there are abnormal delays in the update push of the logistics information.

[0021] During the monitoring process, the accumulated transportation time and the latest logistics information are obtained according to the preset monitoring frequency. The accumulated transportation time refers to the total transportation time of the goods from the time of shipment to the current time. For example, a batch of goods was shipped from warehouse A at 10:00 on February 25, and the current time is 22:00 on February 25, then the accumulated transportation time is 12 hours. The latest logistics information reflects the real-time status of the goods, such as "at 22:00 on February 25, the goods arrived at distribution station B".

[0022] Furthermore, in order to determine whether the goods are transported as planned, the expected logistics information is retrieved and matched according to the transportation route and cumulative transportation time of the target batch of goods. The expected logistics information refers to the logistics status that should occur at a certain moment calculated based on the logistics data in the historical data.

[0023] Furthermore, the latest logistics information is verified against the expected logistics information, that is, the currently acquired logistics status is compared with the logistics status that should occur. When the latest logistics information is consistent with the expected logistics information, it means that the transportation of goods is normal and there is no delay, and no subsequent processing is performed; if the latest logistics information lags behind the expected logistics information, it is determined that the logistics information is delayed, and the delay time is further calculated, and the delay node is determined.

[0024] The step of "retrieve expected logistics information according to the transportation route and accumulated transportation time of the target batch of goods, and distinguish and verify the latest logistics information and the expected logistics information" in the method provided in the embodiment of the present application includes: Retrieving historical transportation data of the same transportation route, and retrieving logistics information according to the accumulated transportation time to obtain a historical logistics information set; Retrieve the historical logistics information with the largest proportion as the expected logistics information; Determine whether the latest logistics information is before the expected logistics information in the transportation route. If so, a delay occurs, and the predicted logistics information is used as a delay node. If not, no delay occurs. When a delay occurs, the average cumulative transportation time of the expected logistics information is calculated, and the difference between the average cumulative transportation time and the delay time is calculated to obtain the delay time.

[0025] In the embodiment of the present application, in order to accurately determine whether the transportation of goods is delayed, it is first necessary to retrieve the historical transportation data of the same transportation route, that is, to extract the logistics records of the same transportation route in the past. The logistics record data of the same transportation route can be stored in the logistics management system or supply chain data platform for easy call and analysis. The historical logistics information includes the time when the information of each logistics node in the same transportation route is updated.

[0026] Furthermore, according to the current cumulative transportation time, the logistics information is retrieved, and the updated logistics node information under the cumulative transportation time in the same transportation route in the logistics record is retrieved to obtain the historical logistics information set. For example, if the current cumulative transportation time of the goods is 36 hours, all the logistics status of the goods that have been under the same transportation route with a cumulative transportation time of 36 hours are retrieved. For example, the logistics status of the goods is B distribution station, and the historical logistics information of "36 hours-B distribution station" is formed to obtain the historical logistics information set, which may also include the historical logistics information of "36 hours-A transfer center". The historical logistics information set reflects the logistics status of the goods transportation when the cumulative transportation time is reached under the same transportation route before.

[0027] Furthermore, the historical logistics information with the largest appearance ratio in the historical logistics information set is counted as the expected logistics information, that is, the logistics status information that should be achieved when the cumulative transportation time arrives under normal transportation conditions. For example, the number of occurrences of each logistics information that is the same as the historical logistics information is counted, and the ratio to the total data volume of the historical logistics information set is calculated to obtain the appearance ratio, and then the historical logistics information with the largest appearance ratio is screened. For example, the appearance ratio of "36 hours-B distribution station" is 85%, and the appearance ratio of "36 hours-A transfer center" is 15%. The historical logistics information with the largest appearance ratio is "36 hours-B distribution station", which is used as the expected logistics information.

[0028] Furthermore, the latest logistics information is compared with the expected logistics information to determine whether the goods are delayed. Specifically, if the logistics node recorded in the latest logistics information is before the logistics node recorded in the expected logistics information, it means that the cargo transportation progress is lagging behind, and the predicted logistics information is used as the delayed node. If the logistics node recorded in the latest logistics information is after or consistent with the logistics node recorded in the expected logistics information, there is no delay and no subsequent processing is performed. For example, if the latest logistics information is "36 hours-A transfer center" and the expected logistics information is "36 hours-B distribution station", and the A transfer center is before the B distribution station in the transportation route, it is determined that a delay has occurred.

[0029] Furthermore, when a delay occurs, the delay time is calculated. Specifically, the average cumulative transportation time of the expected logistics information is extracted, that is, based on the logistics records of the same transportation route in the previous period, the average of the historical cumulative transportation time for the goods to reach the logistics nodes in the expected logistics information is calculated as the average cumulative transportation time, and then the difference between the average cumulative transportation time and the cumulative transportation time obtained by the current monitoring is calculated to obtain the delay time.

[0030] For example, if in the logistics records of the same transportation route, the average cumulative transportation time for the goods to arrive at the B distribution station is 35 hours, and the current cumulative transportation time is 36 hours, then the delay time is 36-35=1 hour.

[0031] In the embodiment of the present application, by matching the logistics status based on historical data, counting the expected logistics nodes, comparing the current logistics status, and calculating the delay time, it is possible to accurately identify the delay of goods, provide reliable delay warnings for supply chain managers, and provide key data support for subsequent logistics forecasts and information compensation push, thereby optimizing logistics scheduling and improving the overall operating efficiency of the supply chain.

[0032] S20: Obtaining demand update frequency and logistics query frequency for the same type of goods by multiple users associated with the target batch of goods.

[0033] In the embodiment of the present application, the supply chain involves multiple users associated with the target batch of goods, such as wholesalers, retailers, and users who purchase goods at the terminal. According to the purchase relationship of the target batch of goods in the supply chain, the multiple users associated with the target batch of goods are obtained.

[0034] Different users pay different attention to logistics information push. Therefore, it is necessary to obtain the demand update frequency and logistics query frequency of multiple users for the same type of goods, reflect the user's attention to the target batch of goods, conduct subsequent logistics information prediction and push, and improve the adaptability of logistics information push processing.

[0035] Step S20 in the method provided in the embodiment of the present application includes: Acquire multiple users associated with the target batch of goods in the supply chain; Acquire the frequencies of the multiple users purchasing the same type of goods as multiple demand update frequencies; The number of times the multiple users have performed logistics queries in a historical period is obtained, and multiple logistics query frequencies are calculated.

[0036] In the embodiment of the present application, it is first necessary to obtain multiple users associated with the target batch of goods in the supply chain. Associated users refer to enterprises or individuals that have a transaction relationship with the batch of goods, including manufacturers, distributors, retailers and end customers. The embodiment of the present application centralizes the logistics information push of the supply chain on one platform, and all users in the supply chain can receive the logistics information of the goods, which is convenient for formulating production, procurement, sales and purchase plans, and improving the efficiency and transparency of the supply chain.

[0037] Furthermore, the frequency of multiple users purchasing the same type of goods is analyzed and used as multiple demand update frequencies. The demand update frequency refers to the ratio of the number of purchases of the same type of goods by the user within a certain period of time (such as day, week, month) to the time. For example, if a user purchases 50 items in a month, the demand update frequency is 50 items / month. In this way, based on the purchase data of users on the platform, multiple demand update frequencies of multiple users are calculated. The demand update frequency reflects the user's purchasing power for goods. The larger the demand update frequency, the more important the user is and the more accurate logistics information push is needed.

[0038] Furthermore, the number of times multiple users have conducted logistics inquiries on the same type of goods in the historical period is obtained, and multiple logistics query frequencies are calculated accordingly. Logistics query frequency refers to the number of times a user accesses the logistics system and inquires about the order status within a certain period (such as day, week, month). For example, based on the platform, if a user has inquired about logistics information 10 times in a month, the number of logistics queries is divided by the unit time, and the logistics query frequency is 10 times / month. In this way, multiple logistics query frequencies of multiple users are obtained. The logistics query frequency reflects the user's attention to logistics information. In order to improve the user experience, the greater the logistics query frequency, the more accurate logistics information push needs to be provided.

[0039] Through the above steps, the demand update frequency and logistics query frequency of multiple users for the same type of goods are obtained, which can be used to judge the degree of attention paid by different users to logistics information, and then to carry out subsequent logistics information forecasting and push to improve adaptability.

[0040] S30: According to multiple demand update frequencies, multiple logistics query frequencies and delay times, logistics prediction parameters are configured, and the arrival time of the delayed node is predicted to obtain predicted logistics node information.

[0041] In an embodiment of the present application, in the process of pushing supply chain logistics information, in order to improve the accuracy and timeliness of logistics information, it is necessary to configure logistics prediction parameters according to multiple demand update frequencies, multiple logistics query frequencies and delay times, and then predict the arrival time of the delayed node, and finally obtain the predicted logistics node information.

[0042] Among them, the logistics prediction parameters are configured according to multiple demand update frequencies, multiple logistics query frequencies and delay times. Personalized logistics information prediction can be performed based on the objective degree of delay and the subjective degree of user concern about the logistics of goods. The logistics prediction parameters are specifically the computing power used for logistics information prediction, which can improve the applicability and accuracy of logistics information prediction while saving some computing power resources.

[0043] By predicting logistics node information and pushing it, the timeliness of logistics information updates can be guaranteed. At the same time, predictions through machine learning can also ensure the accuracy of logistics information, thereby improving user experience.

[0044] Step S30 in the method provided in the embodiment of the present application includes: Obtaining a maximum demand update frequency, respectively calculating ratios of the multiple demand update frequencies to the maximum demand update frequency, and obtaining multiple demand prediction coefficients; Obtaining a maximum logistics query frequency, respectively calculating ratios of the multiple logistics query frequencies to the maximum logistics query frequency, and obtaining multiple query prediction coefficients; Obtaining a maximum delay time, calculating a ratio of the delay time to the maximum delay time, and obtaining a time prediction coefficient; Respectively calculating the average values ​​of the multiple demand prediction coefficients, the multiple query prediction coefficients and the time prediction coefficient to obtain multiple logistics prediction parameters; training a logistics predictor including multiple logistics prediction paths; According to the multiple logistics prediction parameters, multiple configuration path quantities are calculated, and multiple groups of logistics prediction paths are randomly selected. The delay nodes are input, predicted outputs are output and averaged to obtain multiple predicted logistics node information.

[0045] In an embodiment of the present application, in the process of supply chain logistics information prediction, in order to improve the accuracy and pertinence of logistics information prediction, logistics prediction parameters of multiple users are configured and calculated according to multiple demand update frequencies, multiple logistics query frequencies and delay times, and a logistics predictor is trained to perform logistics node prediction.

[0046] In the embodiment of the present application, the maximum demand update frequency is first obtained, that is, the maximum value among multiple demand update frequencies of multiple users. The maximum demand update frequency reflects that the user has the largest purchase amount of goods and is the most important.

[0047] Furthermore, the ratios of multiple demand update frequencies to the maximum demand update frequency are calculated to obtain multiple demand forecast coefficients. The demand forecast coefficient reflects the proportion of computing resources configured according to the demand update frequency in the predicted logistics information. The greater the demand update frequency, the greater the user's purchase of goods, and the greater the demand forecast coefficient. For example, if the maximum demand update frequency is 100 pieces / month and a user's demand update frequency is 50 pieces / month, the demand forecast coefficient is 50 / 100=0.5.

[0048] Furthermore, the maximum value among multiple logistics query frequencies is obtained as the maximum logistics query frequency, and then the ratio of multiple logistics query frequencies to the maximum logistics query frequency is calculated to obtain multiple query prediction coefficients. The query prediction coefficient reflects the proportion of computing resources configured according to the logistics query frequency in the predicted logistics information. The greater the logistics query frequency, the greater the user's attention to logistics, and the greater the query prediction coefficient. For example, if the maximum logistics query frequency is 30 times / month and a user's logistics query frequency is 10 times / month, then the query prediction coefficient is 10 / 30=0.33.

[0049] Furthermore, the maximum delay time is obtained, specifically, the maximum delay time when logistics information delay occurs in the same transportation route in the historical time. The delay time can be calculated by the method in the aforementioned content. The greater the delay time, the more serious the logistics information delay, and the more accurate logistics information prediction and push are needed to ensure user experience, that is, more computing power resources are configured for logistics information prediction. Specifically, the ratio of the current delay time to the maximum delay time is calculated to obtain the time prediction coefficient. The time prediction coefficient reflects the proportion of computing power resources configured according to the current delay time in the predicted logistics information. For example, if the maximum delay time is 5 hours and the current delay time is 1 hour, the time prediction coefficient is 1 / 5=0.2.

[0050] The final logistics information prediction computing power resource ratio is configured based on the objective delay time, as well as the subjective demand update frequency and logistics query frequency of multiple users. Specifically, the average of each user's demand prediction coefficient, query prediction coefficient and time prediction coefficient is calculated to obtain the logistics prediction parameters. In this way, multiple logistics prediction parameters of multiple users are calculated. For example, if a user's demand prediction coefficient, query prediction coefficient and time prediction coefficient are 0.5, 0.33 and 0.2 respectively, the logistics prediction parameter is 0.34.

[0051] In this way, according to the current logistics delay situation, the user's demand update frequency, and the logistics query frequency, the computing power resource ratio and logistics prediction parameters for predicting logistics information are configured. When the delay is serious and the user's attention to the logistics of the goods is high, a larger computing power resource ratio is configured to improve the accuracy of the logistics prediction. When the delay is small and the user's attention to the logistics of the goods is low, a smaller computing power resource ratio is configured to save computing power consumption and improve the efficiency of logistics prediction and push.

[0052] In an embodiment of the present application, a logistics predictor including multiple logistics prediction paths is trained to predict logistics information.

[0053] The step of "training a logistics predictor including multiple logistics prediction paths" in the method of the embodiment of the present application includes: According to the transportation data of the same transportation route in the historical time, a set of sample delay nodes is collected, and the actual arrival time of different sample delay nodes is collected and marked as a set of sample logistics node information; Performing data division on the sample delay node set and the sample logistics node information set to obtain a plurality of logistics prediction training data; The multiple logistics prediction training data are used, and machine learning is used to train multiple logistics node prediction paths to obtain a logistics node predictor.

[0054] In an embodiment of the present application, the delay of logistics information may be caused by data transmission failure or low efficiency of information entry. At this time, the logistics node predictor is trained according to historical transportation data to predict the logistics information for push, which can ensure the timeliness of the logistics information push and the accuracy of the logistics information.

[0055] Specifically, based on the transportation data of the same transportation route in the historical time, the delay nodes when the logistics information is delayed are collected, such as A transfer center, B distribution station, wholesaler warehouse, etc., to obtain a sample delay node set. In addition, the actual arrival time of different sample delay nodes and the average of the cumulative transportation time from the start of transportation to the arrival of the sample delay node are collected, and marked as sample logistics node information. For example, according to the logistics driver's records or later tracing, the actual arrival time of the sample delay node is obtained. For example, the actual arrival time of the A transfer center is 12 hours, which is marked as the sample logistics node information to obtain a sample logistics node information set.

[0056] Furthermore, the sample delay node set and the sample logistics node information set are divided into data, specifically, evenly divided, to generate multiple logistics prediction training data with the same data volume, each of which includes multiple sample delay nodes and corresponding sample logistics node information. For example, 10,000 data are divided into 10 logistics prediction training data, each of which includes 1,000 sample delay nodes and corresponding sample logistics node information.

[0057] Furthermore, machine learning of ensemble learning is adopted to train multiple logistics node prediction paths using multiple logistics prediction training data. For example, a feedforward neural network is adopted to train multiple logistics node prediction paths.

[0058] Taking the training process of one of the logistics node prediction paths as an example, the training steps of multiple logistics node prediction paths are explained. According to the feedforward neural network, the logistics node prediction path is constructed, including the input layer, hidden layer and output layer. The input feature of the input layer is the delay node, and the output feature of the output layer is the predicted logistics node arrival time. The linear activation function is used, and the hidden layer includes multiple neurons using the ReLU activation function. During the training process, the sample delay node is input to obtain the output logistics node information, that is, the actual arrival time of the predicted sample delay node, and the error calculated with the actual sample logistics node information is used as the loss. The loss function uses the mean square error MSE loss function, as shown in the following formula: ; Among them, L is the loss, N is the amount of data in each logistics prediction training data, The logistics node information of the logistics node prediction path predicted after the input of the i-th sample delay node, is the real logistics node information of the i-th sample. According to the error loss back propagation, adjust the network parameters and reduce the loss until the loss is less than the requirement, for example, less than 6%, then the training is completed. In this way, the training obtains multiple logistics node prediction paths.

[0059] Combine multiple trained logistics node prediction paths to obtain a trained logistics node predictor.

[0060] The logistics node predictor that is finally trained can predict the logistics node information based on the delayed nodes, that is, the actual cumulative running time to reach the delayed nodes, and realize the logistics information prediction when the logistics information is delayed, so as to avoid the logistics information delay caused by the delay in logistics data collection and update, and improve the timeliness of logistics information push.

[0061] Furthermore, based on multiple logistics prediction parameters and the number of multiple logistics prediction paths, the number of multiple configuration paths is calculated to adjust the proportion of computing power resources for predicting logistics information.

[0062] Specifically, each logistics prediction parameter is multiplied by the number of multiple logistics prediction paths and rounded to obtain the number of configured paths. For example, if the logistics prediction parameter is 0.34 and the number of multiple logistics prediction paths is 10, the number of configured paths is 0.34*10 rounded to 4. That is, 4 logistics prediction paths are used to predict logistics information. The larger the logistics prediction parameter, the larger the number of configured logistics prediction paths and the higher the accuracy of predicted logistics information.

[0063] Furthermore, according to the number of multiple configured paths, multiple groups of logistics prediction paths are randomly selected, and the number of each group of logistics prediction paths corresponds to the number of configured paths. The current delay node is input, and the prediction output obtains multiple groups of predicted logistics node information, specifically the cumulative transportation time when the multiple groups of delay nodes arrive. Further, the mean of the multiple groups of predicted logistics node information is calculated to obtain multiple predicted logistics node information, that is, the predicted logistics node information to be pushed to multiple users.

[0064] Among them, since the training data of multiple logistics prediction paths are different, the predicted logistics node information output by multiple logistics prediction path predictions may also be different. Calculating the mean of the predicted logistics node information output by different logistics prediction path predictions can improve the accuracy of the predicted logistics node information.

[0065] The embodiment of the present application calculates the number of multiple configuration paths, that is, determines the number of predicted paths according to logistics prediction parameters, and performs prediction calculations, thereby obtaining more accurate logistics node update information and improving the timeliness, accuracy and reliability of supply chain logistics information.

[0066] S40: According to multiple demand update frequencies, multiple logistics query frequencies and delay times, error marking is performed on the predicted logistics node information respectively, multiple logistics information is generated, and update push is performed.

[0067] In the embodiment of the present application, due to the time interval for updating and synchronizing logistics data, there are certain errors in the logistics information, and there are also certain errors in the logistics information prediction in the embodiment of the present application. Therefore, when pushing logistics information, it is necessary to mark the errors of the predicted logistics node information to provide a reference for users.

[0068] Among them, according to multiple demand update frequencies, multiple logistics query frequencies and delay times, the predicted logistics node information is respectively marked with errors, so as to make different error marks according to the degree of attention and delay of different users to logistics. For example, if the user pays high attention to logistics and the current logistics delay is large, in order to avoid affecting the user's decision and ensure the correctness of the predicted logistics information push, an error mark with a larger error range is made. On the contrary, if the user pays less attention to logistics and the current logistics delay is small, an error mark with a smaller error range is made to improve the user experience.

[0069] Step S40 in the method provided in the embodiment of the present application includes: The logistics error range is calculated based on the arrival time of the delayed node in the same transportation route in the historical time; According to the multiple logistics prediction parameters, error adaptability calculation is performed on the logistics error range to obtain multiple logistics error identification information; The multiple logistics error identification information is used to identify errors in the predicted logistics node information, generate logistics information, and push updates.

[0070] In the embodiment of the present application, according to the historical transportation logistics records of the same transportation route, all arrival times of the delayed node are obtained, that is, the total cumulative transportation time to the delayed node in the logistics transportation, and the error range is calculated. For example, the mean of the arrival time of the delayed node is calculated, for example, 10 hours, and the minimum arrival time and the maximum arrival time are extracted, for example, 9 hours and 11.5 hours, and the maximum error range is calculated, then the maximum error range is 13-10=1.5 hours, and the logistics error range is ±11.5 hours.

[0071] According to multiple logistics forecast parameters, the logistics error range is calculated by error adaptability. Specifically, multiple logistics forecast parameters are multiplied by the maximum logistics error range to obtain multiple logistics error identification information. For example, if the logistics forecast parameter is 0.34 and the logistics error range is ±3 hours, the logistics error identification information calculated by error adaptability is 0.34*±1.5 hours=±0.51 hours. In this way, multiple logistics error identification information is calculated.

[0072] Furthermore, multiple logistics error identification information is used to identify the errors of the predicted logistics node information respectively, and then multiple logistics information is generated and updated and pushed to multiple users.

[0073] Among them, according to the cumulative transportation time of the predicted delay node in the predicted logistics node information, combined with the current actual cumulative transportation time, the logistics time information of the target batch of goods arriving at the delay node is generated, which is specifically the difference between the current time minus the predicted logistics node information and the current actual cumulative transportation time. For example, the current actual cumulative transportation time is 12 hours, the current time is 17:00, and the arrival time of the predicted delay node in the predicted logistics node information is 10 hours, then the time for the target batch of goods to arrive at the delay node is 17:00-2 hours=15:00, then it is generated: the logistics information that predicts that the goods will arrive at the delay node (such as A transfer center) at 15:00, and the time error is ±0.51 hours, is pushed to the corresponding user. Optionally, it can also be: the logistics information that predicts that the goods will arrive at the delay node (such as A transfer center) at 15:00, and the time error is ±30.6 minutes.

[0074] In an embodiment of the present application, when pushing logistics information, it can also be marked that the logistics information is generated based on artificial intelligence prediction to provide a prompt.

[0075] A supply chain logistics information update push method provided by an embodiment of the present invention has at least the following technical effects: The embodiment of the present invention can monitor the transportation status of goods in real time by obtaining the cumulative transportation time and the latest logistics information of the target batch of goods in transportation, and provide accurate basic data for logistics information push. When there is a delay in the push of logistics information, the delay node and delay time can be automatically extracted, the abnormal logistics status can be accurately identified, and the impact of information lag on user decision-making can be avoided. Further combined with the demand update frequency and logistics query frequency of multiple users for similar goods, the logistics information can be predicted and pushed to meet the needs of different users. The present invention also configures logistics prediction parameters to predict logistics nodes. When the logistics data is lagging, historical data and machine learning algorithms can be used to infer the time when the goods arrive at the delay node, so that the logistics information push no longer depends on a single logistics data source, effectively reducing the uncertainty caused by information delay. Combined with the error identification mechanism, it is ensured that the pushed information is more valuable for reference, and the user experience and supply chain management efficiency are improved. In summary, the present invention can effectively solve the problem of delayed logistics information push in the prior art and improve the timeliness of logistics information.

[0076] Embodiment 2, as Figure 2 As shown, the invention concept is the same as that of the supply chain logistics information update push method in embodiment 1. The embodiment of the present invention further provides a supply chain logistics information update push system. The explanation of the supply chain logistics information update push method in embodiment 1 is also applicable to a supply chain logistics information update push system, which includes: The delay information acquisition module 11 is used to obtain the cumulative transportation time and the latest logistics information of the target batch of goods in transportation in the supply chain logistics, and extract the delay node and delay time when there is a delay in the push of logistics information; A user information acquisition module 12 is used to obtain the demand update frequency and logistics query frequency of multiple users associated with the target batch of goods for the same type of goods; The logistics information prediction module 13 is used to configure logistics prediction parameters according to multiple demand update frequencies, multiple logistics query frequencies and delay times, perform arrival time prediction of the delay node, and obtain predicted logistics node information; The logistics information push module 14 is used to mark the errors of the predicted logistics node information according to multiple demand update frequencies, multiple logistics query frequencies and delay times, generate multiple logistics information, and push updates.

[0077] Furthermore, the delay information acquisition module 11 is also used for: in supply chain logistics, obtaining the transportation route of the target batch of goods in transportation, and monitoring and obtaining the cumulative transportation time and the latest logistics information of the target batch of goods according to a preset monitoring frequency; retrieving and obtaining the expected logistics information based on the transportation route and cumulative transportation time of the target batch of goods, distinguishing and verifying the latest logistics information and the expected logistics information, and when the latest logistics information is delayed, calculating the delay time and using the expected logistics information as the delay node; when the latest logistics information is not delayed, no subsequent processing is performed.

[0078] Among them, according to the transportation route and cumulative transportation time of the target batch of goods, the expected logistics information is retrieved, and the latest logistics information and the expected logistics information are distinguished and verified, including: retrieving historical transportation data of the same transportation route, and retrieving logistics information according to the cumulative transportation time to obtain a historical logistics information set; retrieving historical logistics information with the largest appearance ratio as the expected logistics information; judging whether the latest logistics information is before the expected logistics information within the transportation route, if so, a delay occurs, and the predicted logistics information is used as the delay node, if not, no delay occurs; when a delay occurs, the average cumulative transportation time of the expected logistics information is calculated, and the difference with the cumulative transportation time is calculated to obtain the delay time.

[0079] Furthermore, the user information acquisition module 12 is also used to: obtain multiple users associated with the target batch of goods in the supply chain; obtain the frequency of the multiple users purchasing the same type of goods as multiple demand update frequencies; obtain the number of times the multiple users conduct logistics queries in a historical period, and calculate multiple logistics query frequencies.

[0080] Furthermore, the logistics information prediction module 13 is also used to: obtain the maximum demand update frequency, calculate the ratios of the multiple demand update frequencies to the maximum demand update frequency, and obtain multiple demand prediction coefficients; obtain the maximum logistics query frequency, calculate the ratios of the multiple logistics query frequencies to the maximum logistics query frequency, and obtain multiple query prediction coefficients; obtain the maximum delay time, calculate the ratio of the delay time to the maximum delay time, and obtain a time prediction coefficient; calculate the average of the multiple demand prediction coefficients, the multiple query prediction coefficients and the time prediction coefficient, and obtain multiple logistics prediction parameters; train a logistics predictor including multiple logistics prediction paths; calculate and obtain multiple configuration path quantities based on the multiple logistics prediction parameters, randomly select multiple groups of logistics prediction paths, input the delay nodes, predict the outputs and average them to obtain multiple predicted logistics node information.

[0081] Among them, the training includes a logistics predictor for multiple logistics prediction paths, including: collecting a sample delay node set based on the transportation data of the same transportation route in historical time, and collecting the actual arrival time of different sample delay nodes, marking them as a sample logistics node information set; dividing the sample delay node set and the sample logistics node information set to obtain multiple logistics prediction training data; using the multiple logistics prediction training data and machine learning to train multiple logistics node prediction paths to obtain a logistics node predictor.

[0082] Furthermore, the logistics information push module 14 is also used to: calculate the logistics error range based on the arrival time of the delayed node in the same transportation route within the historical time; perform error adaptability calculation on the logistics error range based on the multiple logistics prediction parameters to obtain multiple logistics error identification information; use the multiple logistics error identification information to perform error identification on the predicted logistics node information, generate logistics information, and push updates.

[0083] Embodiment 3, the embodiment of the present application provides a supply chain logistics information update push platform, including a supply chain logistics information update push system in embodiment 2.

[0084] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0085] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention 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.

[0086] The present invention is described with reference to the flow diagrams and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flow diagram and / or block diagram, as well as the combination of flows and / or blocks in the flow diagram 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 computer, 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 flow diagram and / or block diagram. Figure 1 flow or flows and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0087] 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 work in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the flow Figure 1 flow or flows and / or boxes Figure 1 A function specified in one or more boxes.

[0088] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the flow Figure 1 flow or flows and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0089] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0090] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A supply chain logistics information update push method, characterized in that: The method comprises: In supply chain logistics, the cumulative transportation time and the latest logistics information of the target batch of goods in transportation are obtained. When there is a delay in pushing logistics information, the delay node and delay time are extracted; Obtaining the demand update frequency and logistics query frequency of multiple users associated with the target batch of goods for the same type of goods; According to multiple demand update frequencies, multiple logistics query frequencies and delay times, logistics prediction parameters are configured to predict the arrival time of the delay node and obtain predicted logistics node information; According to multiple demand update frequencies, multiple logistics query frequencies and delay times, the predicted logistics node information is respectively marked with errors, multiple logistics information is generated, and updated and pushed.

2. The supply chain logistics information update and push method according to claim 1, characterized in that: In supply chain logistics, the cumulative transportation time and the latest logistics information of the target batch of goods in transportation are obtained. When there is a delay in pushing logistics information, the delay node and delay time are extracted, including: In supply chain logistics, obtain the transportation route of the target batch of goods in transit, and monitor and obtain the cumulative transportation time and the latest logistics information of the target batch of goods according to the preset monitoring frequency; According to the transportation route and accumulated transportation time of the target batch of goods, the expected logistics information is retrieved and obtained, the latest logistics information and the expected logistics information are distinguished and verified, and when the latest logistics information is delayed, the delay time is calculated and the expected logistics information is used as the delay node; When there is no delay in the latest logistics information, no subsequent processing will be performed.

3. The supply chain logistics information update and push method according to claim 2, characterized in that: According to the transportation route and cumulative transportation time of the target batch of goods, the expected logistics information is retrieved and the latest logistics information and the expected logistics information are distinguished and verified, including: Retrieving historical transportation data of the same transportation route, and retrieving logistics information according to the accumulated transportation time to obtain a historical logistics information set; Retrieve the historical logistics information with the largest proportion as the expected logistics information; Determine whether the latest logistics information is before the expected logistics information in the transportation route. If so, a delay occurs, and the predicted logistics information is used as a delay node. If not, no delay occurs. When a delay occurs, the average cumulative transportation time of the expected logistics information is calculated, and the difference between the average cumulative transportation time and the delay time is calculated to obtain the delay time.

4. The supply chain logistics information update and push method according to claim 1, characterized in that: Obtaining the demand update frequency and logistics query frequency of the same type of goods by multiple users associated with the target batch of goods, including: Acquire multiple users associated with the target batch of goods in the supply chain; Acquire the frequencies of the multiple users purchasing the same type of goods as multiple demand update frequencies; The number of times the multiple users have performed logistics queries in a historical period is obtained, and multiple logistics query frequencies are calculated.

5. The supply chain logistics information update and push method according to claim 1, characterized in that: According to multiple demand update frequencies, multiple logistics query frequencies and delay times, configure logistics forecast parameters, perform logistics node forecasts, and obtain forecast logistics node information, including: Obtaining a maximum demand update frequency, respectively calculating ratios of the multiple demand update frequencies to the maximum demand update frequency, and obtaining multiple demand prediction coefficients; Obtaining a maximum logistics query frequency, respectively calculating ratios of the multiple logistics query frequencies to the maximum logistics query frequency, and obtaining multiple query prediction coefficients; Obtaining a maximum delay time, calculating a ratio of the delay time to the maximum delay time, and obtaining a time prediction coefficient; Respectively calculating the average values ​​of the multiple demand prediction coefficients, the multiple query prediction coefficients and the time prediction coefficient to obtain multiple logistics prediction parameters; training a logistics predictor including multiple logistics prediction paths; According to the multiple logistics prediction parameters, multiple configuration path quantities are calculated, and multiple groups of logistics prediction paths are randomly selected. The delay nodes are input, predicted outputs are output and averaged to obtain multiple predicted logistics node information.

6. The supply chain logistics information update and push method according to claim 1, characterized in that: Training a logistics predictor that includes multiple logistics prediction paths, including: According to the transportation data of the same transportation route in the historical time, a set of sample delay nodes is collected, and the actual arrival time of different sample delay nodes is collected and marked as a set of sample logistics node information; Performing data division on the sample delay node set and the sample logistics node information set to obtain a plurality of logistics prediction training data; The multiple logistics prediction training data are used, and machine learning is used to train multiple logistics node prediction paths to obtain a logistics node predictor.

7. The supply chain logistics information update and push method according to claim 5, characterized in that: According to multiple demand update frequencies, multiple logistics query frequencies and delay times, the predicted logistics node information is marked with errors, multiple logistics information is generated, and updated and pushed, including: The logistics error range is calculated based on the arrival time of the delayed node in the same transportation route in the historical time; According to the multiple logistics prediction parameters, error adaptability calculation is performed on the logistics error range to obtain multiple logistics error identification information; The multiple logistics error identification information is used to identify errors in the predicted logistics node information, generate logistics information, and push updates.

8. A supply chain logistics information update push system, characterized in that: The system is used to execute the method according to any one of claims 1 to 7, and the system comprises: The delay information acquisition module is used to obtain the cumulative transportation time and the latest logistics information of the target batch of goods in transportation in the supply chain logistics, and extract the delay node and delay time when there is a delay in the push of logistics information; A user information acquisition module, used to obtain the demand update frequency and logistics query frequency of multiple users associated with the target batch of goods for the same type of goods; The logistics information prediction module is used to configure logistics prediction parameters according to multiple demand update frequencies, multiple logistics query frequencies and delay times, predict the arrival time of the delay node, and obtain predicted logistics node information; The logistics information push module is used to identify errors in the predicted logistics node information according to multiple demand update frequencies, multiple logistics query frequencies and delay times, generate multiple logistics information, and push updates.

9. A supply chain logistics information update push platform, characterized in that: The platform includes a supply chain logistics information update push system as described in claim 8.

Images