Order trajectory prediction method, device and equipment based on SVM (Support Vector Machine) and storage medium

The method addresses the challenge of improving the prediction of arrival times in logistics networks by using SVM to analyze the arrival times and dynamically predict arrival times by updating durations based on real-time data analysis.

CN119067543APending Publication Date: 2024-12-03SHANGHAI DONGPU INFORMATION TECH CO LTD
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411114837.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Traditional methods for predicting order arrival times in logistics networks are inadequate in capturing dynamic changes, leading to inaccuracies and reduced customer satisfaction due to uncertainties from factors like weather and traffic.

Method used

A method and device using SVM to analyze real-time logistics data, including order information, transportation routes, weather, and traffic conditions, to dynamically predict arrival times by updating durations based on support vector machine analysis.

Benefits of technology

Improves prediction accuracy by adapting to real-time network dynamics, enhancing customer satisfaction through precise estimated arrival time delivery to improve customer satisfaction and operational efficiency, and reduce transportation costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067543A_ABST
    Figure CN119067543A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of information processing, in particular to an SVM-based order trajectory prediction method, device and equipment and a storage medium, logistics order data is collected in real time, the current time length is calculated according to the current real-time time and the earliest scanning time, and when the current time length is larger than the prediction time length, the prediction time length is calculated. A support vector machine algorithm is utilized to analyze logistics order standard data, various data in a logistics network are collected and analyzed in real time, so that the real-time state of the logistics network is comprehensively known, a dynamic change mode in the logistics network is automatically learned and identified, prediction duration is updated according to an analysis result, and the prediction time is updated. The order arrival time is dynamically predicted, the prediction accuracy is improved, the number of arrival days of each branch in the logistics track is distributed according to the latest duration, the predicted arrival time of each branch is sent to the client, the courier and the mobile terminal of each branch, the client, the courier and each branch can conveniently know the latest order state, and the client satisfaction degree is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to a method, device, equipment and storage medium for predicting order trajectories based on SVM. Background Art

[0002] In the logistics industry, accurate prediction of order arrival time is not only related to the smooth operation of the entire supply chain, but also a key factor in improving user satisfaction and maintaining corporate competitiveness. However, due to the complexity and variability of the logistics network, order arrival time is often full of uncertainty. This uncertainty may come from various factors, such as weather changes, traffic congestion, delays in cargo transportation, etc., which may directly affect the efficiency of logistics and user experience. Traditional time prediction methods, such as static analysis based on historical data or relying on manual experience judgment, can provide some guidance to a certain extent, but in the face of complex and changing logistics environments, their prediction results are often unsatisfactory. These methods are difficult to fully capture and respond to dynamic changes in the logistics network, resulting in a large deviation between the prediction results and the actual situation, thereby reducing customer satisfaction. Summary of the invention

[0003] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a method, device, equipment and storage medium for predicting order trajectories based on SVM, which can collect and analyze various data in the logistics network in real time, so as to fully understand the real-time status of the logistics network, automatically learn and identify the dynamic change patterns in the logistics network, dynamically predict the arrival time of orders, improve the prediction accuracy, and thus improve customer satisfaction.

[0004] The first aspect of the present invention provides a method for predicting an order trajectory based on SVM, comprising: collecting logistics order data in real time, performing data preprocessing on the logistics order data, and obtaining standard logistics order data; identifying the waybill number in the standard logistics order data and the logistics trajectory corresponding to the waybill number, and using the waybill number and the logistics trajectory to query the earliest scanning time and the predicted duration respectively; calculating the current duration according to the current real-time time and the earliest scanning time, and judging whether the current duration is greater than the predicted duration; if so, analyzing the standard logistics order data using a support vector machine algorithm to obtain an analysis result, and updating the predicted duration according to the analysis result to obtain the latest duration; allocating the arrival days of each outlet in the logistics trajectory according to the latest duration to obtain the estimated arrival time of each outlet, and sending the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet.

[0005] Optionally, in a first implementation method of the first aspect of the present invention, the real-time collection of logistics order data and the data preprocessing of the logistics order data to obtain logistics order standard data include: real-time collection of logistics order data, the logistics order data including order information, transportation routes, cargo status, weather conditions and traffic conditions; parsing the logistics order data to obtain parsed data; cleaning the logistics order data to obtain logistics order cleaned data; filling missing values ​​in the logistics order cleaned data to obtain logistics order filled data; and format conversion of the logistics order filled data based on a preset target standard format to obtain logistics order standard data.

[0006] Optionally, in a second implementation method of the first aspect of the present invention, the identifying the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number, and using the waybill number and the logistics track to query the earliest scanning time and the predicted duration respectively, includes: identifying the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number; using the waybill number to query the earliest scanning time; collecting historical order data, and calculating the historical average duration corresponding to the logistics track based on the historical order data; and determining the predicted duration based on the historical average duration.

[0007] Optionally, in a third implementation method of the first aspect of the present invention, the current duration is calculated based on the current real-time time and the earliest scanning time, and whether the current duration is greater than the predicted duration is determined, including: obtaining the current real-time time, determining the start time based on the earliest scanning time; calculating the current duration based on the current real-time time and the start time; and determining whether the current duration is greater than the predicted duration.

[0008] Optionally, in a fourth implementation method of the first aspect of the present invention, if so, then using a support vector machine algorithm to analyze the standard logistics order data to obtain an analysis result, and updating the predicted duration according to the analysis result to obtain the latest duration, including: if so, then using a support vector machine algorithm to analyze the standard logistics order data to obtain an analysis result, the analysis result including weather condition analysis data and traffic condition analysis data; generating a first overtime value according to the weather condition analysis data, and generating a second overtime value according to the traffic condition analysis data; updating the predicted duration according to the first overtime value and the second overtime value to obtain the latest duration.

[0009] Optionally, in a fifth implementation method of the first aspect of the present invention, the arrival days of each outlet in the logistics trajectory are allocated according to the latest duration to obtain an estimated arrival time of each outlet, and the estimated arrival time of each outlet is sent to customers, couriers and mobile terminals of each outlet, including: identifying each track node address in the logistics trajectory, and determining each outlet according to the track node address; identifying the historical collection and delivery time of each outlet corresponding to each outlet, and calculating the collection and delivery time ratio according to the historical collection and delivery time of each outlet; allocating the arrival days of each outlet based on the collection and delivery time ratio and the latest duration to obtain an estimated collection and delivery time for each outlet; calculating the estimated arrival time of each outlet based on the start time and the estimated collection and delivery time of each outlet; sending the estimated arrival time of each outlet to customers, couriers and mobile terminals of each outlet, so that the mobile terminal generates and displays an estimated arrival time page for each outlet.

[0010] Optionally, in a sixth implementation method of the first aspect of the present invention, the arrival days of each outlet in the logistics trajectory are allocated according to the latest duration to obtain the estimated arrival time of each outlet, and the estimated arrival time of each outlet is sent to the customer, the courier and the mobile terminal of each outlet, and it also includes: creating an actual arrival time listener, and using the actual arrival time listener to listen to the actual arrival time information of each outlet; when the actual arrival time listener listens to the actual arrival time information, the current actual arrival time information is obtained; the deviation value between the actual arrival time information and the estimated arrival time of each outlet is analyzed, and the deviation value is recorded to obtain deviation information; and the deviation information is uploaded to the blockchain.

[0011] The second aspect of the present invention provides an order trajectory prediction device based on SVM, including: a collection and processing module, which is used to collect logistics order data in real time, perform data preprocessing on the logistics order data, and obtain logistics order standard data; an identification and query module, which is used to identify the waybill number in the logistics order standard data and the logistics trajectory corresponding to the waybill number, and use the waybill number and the logistics trajectory to query the earliest scanning time and predicted duration respectively; a calculation and judgment module, which is used to calculate the current duration according to the current real-time time and the earliest scanning time, and judge whether the current duration is greater than the predicted duration; an analysis and update module, which is used to analyze the logistics order standard data using a support vector machine algorithm if so, to obtain an analysis result, and update the predicted duration according to the analysis result to obtain the latest duration; an allocation and sending module, which is used to allocate the arrival days of each outlet in the logistics trajectory according to the latest duration, obtain the estimated arrival time of each outlet, and send the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet.

[0012] Optionally, in a first implementation method of the second aspect of the present invention, the collection and processing module includes: a collection unit, used to collect logistics order data in real time, the logistics order data including order information, transportation routes, cargo status, weather conditions and traffic conditions; a parsing unit, used to parse the logistics order data to obtain parsed data; a cleaning unit, used to clean the logistics order data to obtain logistics order cleaned data; a filling unit, used to fill missing values ​​in the logistics order cleaned data to obtain logistics order filled data; a conversion unit, used to convert the format of the logistics order filled data based on a preset target standard format to obtain logistics order standard data.

[0013] Optionally, in a second implementation method of the second aspect of the present invention, the identification and query module includes: an identification unit, used to identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number; a query unit, used to use the waybill number to query the earliest scanning time; a collection and calculation unit, used to collect historical order data, and calculate the historical average duration corresponding to the logistics track based on the historical order data; a determination unit, used to determine the predicted duration based on the historical average duration.

[0014] Optionally, in a third implementation of the second aspect of the present invention, the calculation and judgment module includes: an acquisition and determination unit, used to obtain the current real-time time, and determine the start time according to the earliest scanning time; a first calculation unit, used to calculate the current duration according to the current real-time time and the start time; and a judgment unit, used to judge whether the current duration is greater than the predicted duration.

[0015] Optionally, in a fourth implementation method of the second aspect of the present invention, the analysis and update module includes: an analysis unit, used to, if yes, use a support vector machine algorithm to analyze the logistics order standard data to obtain an analysis result, wherein the analysis result includes weather condition analysis data and traffic condition analysis data; a generation unit, used to generate a first overtime value according to the weather condition analysis data, and to generate a second overtime value according to the traffic condition analysis data; an update unit, used to update the predicted duration according to the first overtime value and the second overtime value to obtain the latest duration.

[0016] Optionally, in a fifth implementation of the second aspect of the present invention, the allocation and sending module includes: an identification and determination unit, used to identify each track node address in the logistics track, and determine each outlet according to the track node address; an identification and calculation unit, used to identify the historical collection and delivery time of each outlet corresponding to each outlet, and calculate the collection and delivery time ratio according to the historical collection and delivery time of each outlet; an allocation unit, used to allocate the arrival days of the outlets based on the collection and delivery time ratio and the latest time, and obtain the estimated collection and delivery time of each outlet; a second calculation unit, used to calculate the estimated arrival time of each outlet according to the start time and the estimated collection and delivery time of each outlet; a sending unit, used to send the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet, so that the mobile terminal generates and displays the estimated arrival time page of each outlet.

[0017] Optionally, in a sixth implementation method of the second aspect of the present invention, it also includes: creating a monitoring module, used to create an actual arrival time listener, and using the actual arrival time listener to monitor the actual arrival time information of each outlet; an acquisition module, used to obtain the current actual arrival time information when the actual arrival time listener listens to the actual arrival time information; an analysis and recording module, used to analyze the deviation value between the actual arrival time information and the estimated arrival time of each outlet, and record the deviation value to obtain deviation information; an upload module, used to upload the deviation information to the blockchain.

[0018] The third aspect of the present invention provides an order trajectory prediction device based on SVM, and the order trajectory prediction device based on SVM comprises: a memory and at least one processor, wherein the memory stores instructions; at least one of the processors calls the instructions in the memory to enable the order trajectory prediction device based on SVM to perform each step of the order trajectory prediction method based on SVM described above.

[0019] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, and when the instructions are executed by a processor, the various steps of any of the above-mentioned methods for predicting order trajectories based on SVM are implemented.

[0020] In the technical solution of the present invention, logistics order data is collected in real time, and the current duration is calculated according to the current real-time time and the earliest scanning time. When the current duration is greater than the predicted duration, the support vector machine algorithm is used to analyze the logistics order standard data, and various data in the logistics network are collected and analyzed in real time to fully understand the real-time status of the logistics network, automatically learn and identify the dynamic change patterns in the logistics network, update the predicted duration according to the analysis results, dynamically predict the order arrival time, improve the prediction accuracy, allocate the arrival days of each outlet in the logistics trajectory according to the latest duration, and send the estimated arrival time of each outlet to customers, couriers and mobile terminals of each outlet, so as to facilitate customers, couriers and each outlet to understand the latest order status and improve customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A first flow chart of a method for predicting order trajectories based on SVM provided in an embodiment of the present invention;

[0022] Figure 2 A second flow chart of the order trajectory prediction method based on SVM provided in an embodiment of the present invention;

[0023] Figure 3 A third flow chart of the order trajectory prediction method based on SVM provided in an embodiment of the present invention;

[0024] Figure 4 A fourth flow chart of the order trajectory prediction method based on SVM provided in an embodiment of the present invention;

[0025] Figure 5 A schematic diagram of a structure of an order trajectory prediction device based on SVM provided in an embodiment of the present invention;

[0026] Figure 6 Another structural schematic diagram of an order trajectory prediction device based on SVM provided in an embodiment of the present invention;

[0027] Figure 7 A schematic diagram of the structure of an order trajectory prediction device based on SVM provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The present invention provides a method, device, equipment and storage medium for predicting order trajectories based on SVM, which collects and analyzes various data in the logistics network in real time to fully understand the real-time status of the logistics network, automatically learn and identify dynamic change patterns in the logistics network, dynamically predict the arrival time of orders, improve prediction accuracy, and thus improve customer satisfaction.

[0029] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 In one embodiment of the present invention, an embodiment of a method for predicting order trajectories based on SVM includes:

[0031] 101. Collect logistics order data in real time, pre-process the logistics order data, and obtain logistics order standard data;

[0032] In this embodiment, logistics order data can come from multiple channels, such as online order systems, logistics companies' APIs, sensor data, etc., and appropriate data collectors are selected according to the characteristics of the data source, which may include timed polling APIs, event-driven data stream processing, or sensor data capture, identify and remove duplicate order data, ensure that the data format meets the expected standards, detect and process outliers in the data, such as invalid order numbers or timeout timestamps, unify the timestamp formats from different sources to facilitate subsequent time analysis, standardize or geocode address information for subsequent trajectory analysis or route planning, integrate order data with other related data, such as customer information or product information, to enrich the data content, and mark the logistics nodes of the order (such as shipment, transit, and delivery) based on the scanning data or GPS data of the logistics service provider.

[0033] 102. Identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number, and use the waybill number and the logistics track to query the earliest scanning time and predicted duration respectively;

[0034] In this embodiment, the waybill number is usually a number that uniquely identifies an order in the logistics process. The waybill number is extracted from the order data using a regular expression or a string processing method. The logistics track contains the key nodes of the waybill in the logistics process, such as shipment, transit, and delivery information. According to the waybill number, the relevant information is queried by calling the API interface provided by the logistics service provider. The earliest scanning time is found from the returned logistics track data, which indicates the start time of the order, that is, the earliest shipment or scanning node time. The timestamp in the logistics track is converted into a comparable format, such as a UNIX timestamp or date and time format. According to the key nodes in the logistics track, such as the shipment and delivery nodes, the predicted logistics duration is calculated.

[0035] 103. Calculate the current duration according to the current real-time time and the earliest scanning time, and determine whether the current duration is greater than the predicted duration;

[0036] In this embodiment, a built-in function or library function of a programming language (such as Python) is used to obtain the timestamp of the current time, and the timestamp of the earliest scanning time is subtracted from the current timestamp to obtain the elapsed time difference, that is, the current duration. According to the previous analysis steps, the predicted logistics duration has been obtained. This predicted duration is a value calculated based on historical data. The calculated current duration is compared with the predicted duration to determine whether the logistics has timed out or is still within the expected range.

[0037] 104. If yes, the standard data of the logistics order is analyzed by using the support vector machine algorithm to obtain the analysis result, and the predicted duration is updated according to the analysis result to obtain the latest duration;

[0038] In this embodiment, when logistics has timed out, features suitable for SVM analysis are selected and extracted from the logistics order standard data, and the features of the logistics order standard data are analyzed using the support vector machine algorithm to obtain analysis results, and the predicted duration is updated based on the analysis results to obtain the latest duration.

[0039] 105. Allocate the arrival days of each outlet in the logistics track according to the latest duration, obtain the estimated arrival time of each outlet, and send the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet;

[0040] In this embodiment, the arrival days of each outlet in the logistics trajectory are allocated according to the latest duration to obtain the estimated arrival time of each outlet, and the estimated arrival time of each outlet is sent to the customer's mobile terminal to facilitate the customer to understand the latest order status. The estimated arrival time of each outlet is sent to the courier's mobile terminal so that they can adjust their itinerary and service arrangements according to the latest estimated arrival time. The estimated arrival time of each outlet is sent to the mobile terminal of each outlet so that the outlet staff can arrange the preparations for receiving and distributing the goods in a timely manner.

[0041] In an embodiment of the present invention, logistics order data is collected in real time, and the current duration is calculated based on the current real-time time and the earliest scanning time. When the current duration is greater than the predicted duration, the support vector machine algorithm is used to analyze the logistics order standard data, and various data in the logistics network are collected and analyzed in real time to fully understand the real-time status of the logistics network, automatically learn and identify dynamic change patterns in the logistics network, update the predicted duration based on the analysis results, dynamically predict the order arrival time, improve the prediction accuracy, allocate the arrival days of each outlet in the logistics trajectory based on the latest duration, and send the estimated arrival time of each outlet to customers, couriers and mobile terminals of each outlet, so that customers, couriers and each outlet can understand the latest order status conveniently and improve customer satisfaction.

[0042] See also Figure 2 The second embodiment of the order trajectory prediction method based on SVM in the embodiment of the present invention includes:

[0043] 201. Collect logistics order data in real time, including order information, transportation routes, cargo status, weather conditions and traffic conditions;

[0044] In this embodiment, logistics order data is collected in real time, including order information, transportation routes, cargo status, weather conditions and traffic conditions. Specifically, order number, shipping address, receiving address, estimated delivery time, etc. are collected from an order management system or an e-commerce platform; location information, route planning, estimated arrival time, etc. of each outlet or node are collected from a transportation management system or a GPS tracking device; cargo location and transportation status (such as loading, in transit, in delivery, received, etc.) are collected from a logistics tracking system or an RFID tag; real-time weather information of each node on the transportation route, such as temperature, precipitation, wind speed, etc., is collected from an API or a meteorological monitoring device of a meteorological service provider; real-time traffic flow, congestion, accident reports, etc. of a road section are collected from real-time traffic data, GPS data of a traffic management department or a third-party traffic service provider; real-time data from different data sources are integrated into a unified data platform or data warehouse for subsequent analysis and processing; streaming data processing technology (such as Apache Kafka, Spark Streaming, etc.) is used to process and analyze real-time data to ensure the timeliness and accuracy of the data.

[0045] 202. Parse the logistics order data to obtain parsed data;

[0046] In this embodiment, the logistics order data is parsed according to a predetermined data model. The specific parsing steps may include parsing the order number, shipping address and receiving address, extracting detailed information such as province, city, street, etc., parsing the estimated delivery time, converting it into a standard date and time format, and parsing cargo information such as type, quantity, weight, etc.

[0047] 203. Perform data cleaning on the logistics order data to obtain logistics order cleansing data;

[0048] In this embodiment, the logistics order data is cleaned to remove duplicate data in the logistics order data to obtain logistics order cleaned data.

[0049] 204. Fill missing values ​​in the logistics order cleansing data to obtain logistics order filling data;

[0050] In this embodiment, it is determined whether each field is of numeric type, text type, date type, etc., the proportion and position of missing values ​​in each field are analyzed, and a suitable filling strategy is selected according to the type of field and the distribution of missing values. The selected filling strategy is implemented for each field, the mean or median is calculated, and the missing values ​​are replaced. The most common text values ​​are counted and the missing values ​​are replaced. According to the forward or backward filling strategy, the missing date values ​​are replaced, and the basic statistics of each field are recounted, such as the mean, median, maximum value, minimum value, etc., to ensure that the data distribution after filling is reasonable, and the data after filling is logically verified according to the business logic to ensure that the filled value conforms to the actual situation.

[0051] 205. Based on the preset target standard format, the logistics order filling data is formatted to obtain the logistics order standard data;

[0052] In this embodiment, the preset target standard format is clarified, which includes the name of the field, the data type (numeric type, text type, date type, etc.), the field order, and the meaning and data constraints of each field (such as length limit, enumeration value, etc.), the current logistics order data is analyzed, its current format and field situation are understood, the name of each field and their meaning in the current data are determined, the data type of each field, such as numeric type, text type or date type, is confirmed, and whether there are error values, abnormal values ​​or situations that do not conform to the expected format in the data, the mapping relationship between the current data field and the target standard format field is determined, and the data type is converted or formatted according to the requirements of the target format, such as unifying the date format, adjusting the precision of the numeric field, etc. According to the mapping relationship, the field is renamed, and the data is converted to the requirements of the target format to ensure that each field meets the preset data format and constraints, and multiple source data fields are merged into one field of the target format, or one source data field is split into multiple target format fields.

[0053] 206. Identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number;

[0054] In this embodiment, the structure and field meanings of the logistics order data are understood. The logistics order data will contain core information such as the waybill number and the logistics track. The waybill number is a number that uniquely identifies a logistics order. The logistics track describes the various stages and events in the logistics process from order creation to delivery, such as scanning points, transportation status, location, etc. The relevant fields containing the waybill number and the logistics track are extracted from the original data, and a string matching algorithm (such as the KMP algorithm, the Boyer-Moore algorithm, etc.) is used to quickly locate and extract the waybill number, and the extracted waybill number is associated with the corresponding logistics track data.

[0055] 207. Use the waybill number to check the earliest scanning time;

[0056] In this embodiment, once the logistics order information of a specific waybill number is successfully queried, the next step is to extract the earliest scanning time therefrom, traverse the query results, compare the scanning time one by one, record and update the earliest scanning time found, and once the earliest scanning time is found, convert the earliest scanning time into the required date and time format.

[0057] 208. Collect historical order data, and calculate the historical average duration corresponding to the logistics track based on the historical order data;

[0058] In this embodiment, historical order data is collected, including relevant information of each order, such as waybill number, starting location, destination, shipping time, arrival time, etc. Each order has a starting location and a destination. This information will determine the starting and ending points of the logistics track and the specific transportation path of the order from the starting location to the destination. The time point when the order starts to be transported is the shipping time, and the time point when the order arrives at the destination is the arrival time. For each order, the time difference from the shipping time to the arrival time, that is, the logistics time, is calculated. The logistics time of all orders is counted, including the total time and the number of orders. The total logistics time is divided by the number of orders to obtain the average logistics time, which will be the historical average time of the logistics track.

[0059] 209. Determine the prediction duration based on the historical average duration;

[0060] In this embodiment, the historical average duration is directly used as the predicted duration for all future orders, or the historical average duration is weighted averaged according to different conditions (such as different seasons, different regions, different modes of transportation, etc.) to improve the accuracy of the prediction.

[0061] In the embodiments of the present invention, the real-time update and tracking capabilities of logistics information are guaranteed, which helps to respond to potential logistics problems and adjust plans in a timely manner, ensures the quality and consistency of data, cleans and fills data to remove noise and errors, fills missing values, and converts to a standard format, making the data easier to analyze and compare, thereby improving the accuracy of decision-making. Data-driven prediction capabilities can help optimize logistics planning and resource allocation, predict possible delays or early delivery in advance, thereby improving service levels and customer satisfaction, and significantly improving the efficiency of logistics management. Accurate data analysis and prediction capabilities help reduce transportation time and costs, optimize route planning and resource utilization, and reduce additional costs due to errors and uncertainties. Comprehensive logistics data analysis and processing are provided to support the basis for management in decision-making. From real-time data collection to historical data analysis, each step provides management with deeper insights and understanding, helping them to formulate more reasonable and effective logistics strategies and plans.

[0062] See also Figure 3 The third embodiment of the order trajectory prediction method based on SVM in the embodiment of the present invention includes:

[0063] 301. Obtain the current real time and determine the start time according to the earliest scanning time;

[0064] In this embodiment, the current real time is obtained through a time function provided by a programming language or an operating system, and the start time is determined according to the earliest scanning time.

[0065] 302. Calculate the current duration according to the current real time and the start time;

[0066] In this embodiment, the current real time and the predetermined start time are obtained, and the current duration, that is, the time interval from the start time to the current real time, is calculated.

[0067] 303. Determine whether the current duration is greater than the predicted duration;

[0068] In this embodiment, once the current duration is obtained, a comparison can be performed to determine whether it exceeds the predicted duration. Based on the comparison result, corresponding information can be output to indicate whether the current duration has exceeded the predicted duration.

[0069] 304. If yes, the logistics order standard data is analyzed using a support vector machine algorithm to obtain analysis results, which include weather condition analysis data and traffic condition analysis data;

[0070] In this embodiment, when the current duration is greater than the predicted duration, the trained support vector machine model is used to analyze the changes in order timeliness under different weather conditions. For example, whether the processing time of an order is extended in rainy or hot weather, and the impact of different traffic conditions (congestion, accidents, etc.) on the order delivery time are analyzed, so as to optimize the logistics route or make scheduling preparations in advance.

[0071] 305. Generate a first overtime value according to the weather condition analysis data, and generate a second overtime value according to the traffic condition analysis data;

[0072] In this embodiment, the relationship between weather data and timeliness is analyzed, the impact of weather factors on service timeliness is predicted, and the predicted weather impact is converted into a specific numerical value, namely the first overtime value, which can be an estimate of the expected time extension or increase in service processing time. The relationship between traffic data and service timeliness is analyzed, and the predicted traffic delays are converted into a second overtime value, which represents an estimate of the increase in service processing time due to poor traffic conditions.

[0073] 306. Update the predicted duration according to the first overtime value and the second overtime value to obtain the latest duration;

[0074] In this embodiment, the initial predicted duration is adjusted according to the value of the first overtime value. For example, if the first overtime value is a positive number, it means that the service time is extended due to weather factors, and the initial predicted duration can be increased by the corresponding amount of time. If the first overtime value is a negative number, it means that the weather is conducive to reducing the service time, and the initial predicted duration can be reduced accordingly. According to the value of the second overtime value, the initial predicted duration is adjusted again. When the second overtime value is a positive number, it means that the traffic conditions are not good and the service time should be extended; when it is a negative number, the service time is reduced accordingly. The initial predicted duration is added or subtracted from the values ​​after the influence of the above two overtime values ​​to obtain the final predicted duration. The calculation formula is: latest duration = predicted duration + first overtime value + second overtime value.

[0075] In the embodiment of the present invention, the impact of weather and traffic factors is directly reflected in the final service time prediction, which can more accurately predict the service time changes caused by weather and traffic factors, help to adjust and optimize service arrangements in real time, dynamically predict the order arrival time, and improve the prediction accuracy.

[0076] See also Figure 4 The fourth embodiment of the order trajectory prediction method based on SVM in the embodiment of the present invention includes:

[0077] 401. Identify the addresses of each track node in the logistics track, and determine each network point according to the track node address;

[0078] In this embodiment, address identification and standardization tools or services are used to extract the address information of each trajectory node. Specifically, the address text of each trajectory node is analyzed and processed, key information such as city, street, postal code, etc. is extracted, and the address information is converted into accurate geographic coordinates (latitude and longitude). Once the standardized address or geographic coordinates are obtained, the next step is to map each trajectory node to the corresponding logistics network or service center. A network information database is pre-established, and the nearest logistics network for each node can be queried based on the address or geographic coordinates. Using GIS tools, the geographic coordinates of each trajectory node are compared with the location of the known logistics network or service center to determine the closest network.

[0079] 402. Identify the historical pickup and dispatch time of each outlet corresponding to each outlet, and calculate the pickup and dispatch time ratio according to the historical pickup and dispatch time of each outlet;

[0080] In this embodiment, the average collection and delivery time of each outlet is calculated based on the historical collection and delivery data of each outlet. Specifically, for each collection and delivery, the time elapsed from the package to the successful delivery is calculated, and all collection and delivery times of each outlet are averaged to obtain the average collection and delivery time of the outlet. The proportion of the collection and delivery time of each outlet is calculated, that is, the proportion of the collection and delivery time of each outlet in the collection and delivery time of all outlets.

[0081] 403. Based on the ratio of the pickup and delivery time and the latest time, the arrival days of each outlet are allocated to obtain the estimated pickup and delivery time of each outlet;

[0082] In this embodiment, the ratio of the pickup and delivery time and the latest time have been calculated, and the arrival days of each outlet are allocated to obtain the estimated pickup and delivery time of each outlet.

[0083] 404. Calculate the estimated arrival time of each outlet based on the starting time and the estimated pickup and delivery time of each outlet;

[0084] In this embodiment, the estimated arrival time at each outlet is calculated based on the start time, i.e., the time when the package or cargo begins to be collected, and the estimated collection and delivery time of each outlet.

[0085] 405. Send the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet, so that the mobile terminal generates and displays the estimated arrival time page of each outlet;

[0086] In this embodiment, the calculated estimated arrival time of each outlet is integrated into a data set, and the data set is sent to the customer, the courier and the mobile terminal of each outlet. This can be achieved through email, SMS notification, instant message or application push. For the customer, the tracking information of the package or goods is sent, including the estimated arrival time, which can be displayed on the tracking page of the courier company or sent directly to the customer through SMS and other means. For the courier, a dedicated mobile application is used to manage and track the delivery tasks. In these applications, the estimated arrival time can be displayed as part of the task plan to help the courier effectively arrange the collection and delivery tasks. For each outlet, the mobile terminal can display a list of all packages to be collected and delivered, each package with its estimated arrival time, so that the outlet staff can optimize resource management and time scheduling and ensure that the estimated arrival time is updated in real time in the system to reflect any delays or changes. This can be achieved through automatic system updates.

[0087] 406. Create an actual arrival time listener, and use the actual arrival time listener to monitor the actual arrival time information of each network point;

[0088] In this embodiment, the function and purpose of the actual arrival time listener are defined. The main task of this listener is to obtain the actual arrival time data from various logistics transportation points (such as warehouses, sorting centers, transportation hubs, etc.). The listener needs to collect data from various logistics nodes in real time. These nodes may include logistics centers, transportation nodes or key points on the distribution route. Once the data is collected, the listener will transmit the actual arrival time information to the logistics management system.

[0089] 407. When the actual arrival time listener listens to the actual arrival time information, the current actual arrival time information is obtained;

[0090] In this embodiment, when the actual arrival time listener listens to the actual arrival time information, the actual arrival time listener receives the actual arrival time data from the logistics node or the transportation vehicle.

[0091] 408. Analyze the deviation between the actual arrival time information and the estimated arrival time of each network point, and record the deviation value to obtain deviation information;

[0092] In this embodiment, the time difference between the actual arrival time and the estimated arrival time, that is, the deviation value, is calculated, and the calculated deviation value is recorded in a special database table or data set. Each record usually includes an identifier of the transportation node or outlet (such as node ID, outlet number), estimated arrival time, actual arrival time, calculated deviation value, etc. The timestamp of each deviation information and the metadata of the data source are recorded to ensure the traceability and reliability of the data.

[0093] 409. Upload the deviation information to the blockchain;

[0094] In this embodiment, a smart contract is designed to support the storage and access of deviation information, and a suitable data structure is defined to store the key data of each deviation information, such as the estimated arrival time, the actual arrival time, the deviation value, etc. The collected deviation information is converted into a format acceptable to the smart contract, and the prepared data is uploaded to the blockchain using suitable tools and libraries, and the transaction is signed by a private key to ensure the integrity and authenticity of the data. Once the transaction is submitted to the blockchain, it is necessary to wait for confirmation from the network nodes to ensure that the transaction is correctly recorded on the chain. The smart contract on the blockchain will execute predefined logic to verify the uploaded deviation information, such as checking whether specific conditions or rules are met. The status and history of the uploaded deviation information can be viewed using a blockchain browser to ensure the transparency and traceability of the data, and the immutability and distributed characteristics of the blockchain can be used for auditing and analysis, such as analyzing the distribution pattern of deviation information or finding abnormal data points.

[0095] In the embodiment of the present invention, by calculating the historical data and the ratio of the pick-up and delivery time, the pick-up and delivery time and the arrival time of each outlet can be accurately predicted. This precise prediction helps to optimize logistics routes and resource allocation, improve transportation efficiency and customer satisfaction, and send the estimated arrival time to the mobile terminals of customers, couriers and outlets, ensuring real-time information transmission and transparency, which helps to notify relevant parties in advance, reduce communication costs, and improve the predictability of services and user experience. By creating an actual arrival time listener and analyzing the deviation value, real-time monitoring and feedback of the actual arrival time are achieved. This capability enables the system to respond and adjust quickly, handle any deviations in a timely manner, and ensure the smooth progress of the logistics transportation process. The deviation information is recorded and uploaded to the blockchain, ensuring the security and non-tamperability of the data. This technology ensures the integrity and credibility of the data, and provides a reliable basis for future audits and analysis. It not only provides real-time logistics information and prediction capabilities, but also can optimize future logistics strategies and decisions by analyzing deviation information. This comprehensive advantage can help enterprises improve operational efficiency, reduce costs, and improve customer service quality.

[0096] The above describes the order trajectory prediction method based on SVM in the embodiment of the present invention. The following describes the order trajectory prediction device based on SVM in the embodiment of the present invention. Figure 5 In one embodiment of the present invention, an order trajectory prediction device based on SVM includes:

[0097] The collection and processing module 501 is used to collect logistics order data in real time, perform data preprocessing on the logistics order data, and obtain logistics order standard data;

[0098] Identification query module 502, used to identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number, and use the waybill number and the logistics track to query the earliest scanning time and predicted duration respectively;

[0099] A calculation and judgment module 503 is used to calculate the current duration according to the current real time and the earliest scanning time, and to judge whether the current duration is greater than the predicted duration;

[0100] Analysis and update module 504, for analyzing the logistics order standard data using a support vector machine algorithm to obtain an analysis result, and updating the predicted duration according to the analysis result to obtain the latest duration;

[0101] The allocation and sending module 505 is used to allocate the arrival days of each outlet in the logistics track according to the latest duration, obtain the estimated arrival time of each outlet, and send the estimated arrival time of each outlet to the customer, courier and the mobile terminal of each outlet.

[0102] In this embodiment, logistics order data is collected in real time, and the current duration is calculated based on the current real-time time and the earliest scanning time. When the current duration is greater than the predicted duration, the support vector machine algorithm is used to analyze the logistics order standard data. Various data in the logistics network are collected and analyzed in real time to fully understand the real-time status of the logistics network, automatically learn and identify dynamic change patterns in the logistics network, update the predicted duration based on the analysis results, dynamically predict the order arrival time, improve the prediction accuracy, allocate the arrival days of each outlet in the logistics trajectory based on the latest duration, and send the estimated arrival time of each outlet to customers, couriers and mobile terminals of each outlet, so that customers, couriers and each outlet can understand the latest order status and improve customer satisfaction.

[0103] See also Figure 6 Another embodiment of the order trajectory prediction device based on SVM in the embodiment of the present invention includes:

[0104] The collection and processing module 501 is used to collect logistics order data in real time, perform data preprocessing on the logistics order data, and obtain logistics order standard data;

[0105] Identification query module 502, used to identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number, and use the waybill number and the logistics track to query the earliest scanning time and predicted duration respectively;

[0106] A calculation and judgment module 503 is used to calculate the current duration according to the current real time and the earliest scanning time, and to judge whether the current duration is greater than the predicted duration;

[0107] Analysis and update module 504, for analyzing the logistics order standard data using a support vector machine algorithm to obtain an analysis result, and updating the predicted duration according to the analysis result to obtain the latest duration;

[0108] The allocation and sending module 505 is used to allocate the arrival days of each outlet in the logistics track according to the latest duration, obtain the estimated arrival time of each outlet, and send the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet;

[0109] In this embodiment, the collection and processing module 501 includes: a collection unit 5011, which is used to collect logistics order data in real time, and the logistics order data includes order information, transportation routes, cargo status, weather conditions and traffic conditions; a parsing unit 5012, which is used to parse the logistics order data to obtain parsed data; a cleaning unit 5013, which is used to clean the logistics order data to obtain logistics order cleaned data; a filling unit 5014, which is used to fill missing values ​​in the logistics order cleaned data to obtain logistics order filled data; a conversion unit 5015, which is used to convert the format of the logistics order filled data based on a preset target standard format to obtain logistics order standard data.

[0110] In this embodiment, the identification query module 502 includes: an identification unit 5021, which is used to identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number; a query unit 5022, which is used to use the waybill number to query the earliest scanning time; a collection and calculation unit 5023, which is used to collect historical order data and calculate the historical average duration corresponding to the logistics track based on the historical order data; a determination unit 5024, which is used to determine the predicted duration based on the historical average duration.

[0111] In this embodiment, the calculation and judgment module 503 includes: an acquisition and determination unit 5031, used to obtain the current real-time time and determine the start time according to the earliest scanning time; a first calculation unit 5032, used to calculate the current duration according to the current real-time time and the start time; a judgment unit 5033, used to judge whether the current duration is greater than the predicted duration.

[0112] In this embodiment, the analysis and update module 504 includes: an analysis unit 5041, which is used to analyze the logistics order standard data using a support vector machine algorithm to obtain an analysis result, and the analysis result includes weather condition analysis data and traffic condition analysis data; a generation unit 5042, which is used to generate a first overtime value based on the weather condition analysis data, and to generate a second overtime value based on the traffic condition analysis data; an update unit 5043, which is used to update the predicted duration according to the first overtime value and the second overtime value to obtain the latest duration.

[0113] In this embodiment, the allocation and sending module 505 includes: an identification and determination unit 5051, which is used to identify each track node address in the logistics track, and determine each outlet according to the track node address; an identification and calculation unit 5052, which is used to identify the historical collection and delivery time of each outlet corresponding to each outlet, and calculate the collection and delivery time ratio according to the historical collection and delivery time of each outlet; an allocation unit 5053, which is used to allocate the arrival days of each outlet based on the collection and delivery time ratio and the latest time, and obtain the estimated collection and delivery time of each outlet; a second calculation unit 5054, which is used to calculate the estimated arrival time of each outlet according to the start time and the estimated collection and delivery time of each outlet; a sending unit 5055, which is used to send the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet, so that the mobile terminal generates and displays the estimated arrival time page of each outlet.

[0114] In this embodiment, it also includes: creating a monitoring module 506, which is used to create an actual arrival time monitor, and using the actual arrival time monitor to monitor the actual arrival time information of each outlet; an acquisition module 507, which is used to obtain the current actual arrival time information when the actual arrival time monitor monitors the actual arrival time information; an analysis and recording module 508, which is used to analyze the deviation value between the actual arrival time information and the estimated arrival time of each outlet, and record the deviation value to obtain the deviation information; an upload module 509, which is used to upload the deviation information to the blockchain.

[0115] above Figure 5 and Figure 6 The order trajectory prediction device based on SVM in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The order trajectory prediction device based on SVM in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0116] Figure 76 is a schematic diagram of the structure of a device for predicting order trajectories based on SVM provided by an embodiment of the present invention. The device 600 for predicting order trajectories based on SVM may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 610 (for example, one or more processors) and a memory 620, and one or more storage media 630 (for example, one or more mass storage devices) storing application programs 633 or data 632. Among them, the memory 620 and the storage medium 630 may be short-term storage or permanent storage. The program stored in the storage medium 630 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the device 600 for predicting order trajectories based on SVM. Furthermore, the processor 610 may be configured to communicate with the storage medium 630, and execute a series of instruction operations in the storage medium 630 on the device 600 for predicting order trajectories based on SVM, so as to implement the steps of the method for predicting order trajectories based on SVM provided in the above-mentioned method embodiments.

[0117] The order trajectory prediction device 600 based on SVM may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input and output interfaces 660, and / or one or more operating systems 631, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that Figure 7 The structure of the device for predicting order trajectories based on SVM shown does not constitute a limitation on the device for predicting order trajectories based on SVM, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0118] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the order trajectory prediction method based on SVM.

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

[0120] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0121] Finally, it should be noted that the above description is only a preferred example of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting order trajectories based on SVM. It is characterized in that include: Collecting logistics order data in real time, performing data preprocessing on the logistics order data, and obtaining logistics order standard data; Identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number, and use the waybill number and the logistics track to query the earliest scanning time and predicted duration respectively; Calculate the current duration according to the current real-time time and the earliest scanning time, and determine whether the current duration is greater than the predicted duration; If yes, the logistics order standard data is analyzed using a support vector machine algorithm to obtain an analysis result, and the predicted duration is updated according to the analysis result to obtain the latest duration; The arrival days of each outlet in the logistics track are allocated according to the latest duration to obtain the estimated arrival time of each outlet, and the estimated arrival time of each outlet is sent to the customer, the courier and the mobile terminal of each outlet.

2. The order trajectory prediction method based on SVM according to claim 1, It is characterized in that The real-time collection of logistics order data and data preprocessing of the logistics order data to obtain logistics order standard data include: Collecting logistics order data in real time, including order information, transportation routes, cargo status, weather conditions, and traffic conditions; Parsing the logistics order data to obtain parsed data; Performing data cleaning on the logistics order data to obtain logistics order cleansing data; Filling missing values ​​in the logistics order cleansing data to obtain logistics order filled data; Based on a preset target standard format, the logistics order filling data is formatted to obtain logistics order standard data.

3. The order trajectory prediction method based on SVM according to claim 1, It is characterized in that The identifying the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number, and using the waybill number and the logistics track to query the earliest scanning time and the predicted duration respectively, comprises: Identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number; Use the shipping number to check the earliest scanning time; Collect historical order data, and calculate the historical average duration corresponding to the logistics track according to the historical order data; The predicted duration is determined based on the historical average duration.

4. The order trajectory prediction method based on SVM according to claim 1, It is characterized in that The calculating the current duration according to the current real time and the earliest scanning time, and judging whether the current duration is greater than the predicted duration, comprises: Obtain the current real time, and determine the start time according to the earliest scan time; Calculate the current duration according to the current real time and the start time; Determine whether the current duration is greater than the predicted duration.

5. The order trajectory prediction method based on SVM according to claim 1, It is characterized in that If so, the logistics order standard data is analyzed using a support vector machine algorithm to obtain an analysis result, and the predicted duration is updated according to the analysis result to obtain the latest duration, including: If yes, the logistics order standard data is analyzed using a support vector machine algorithm to obtain analysis results, which include weather condition analysis data and traffic condition analysis data; generating a first overtime value according to the weather condition analysis data, and generating a second overtime value according to the traffic condition analysis data; The predicted duration is updated according to the first overtime value and the second overtime value to obtain a latest duration.

6. The order trajectory prediction method based on SVM according to claim 1, It is characterized in that The method allocates the arrival days of each outlet in the logistics track according to the latest duration to obtain the estimated arrival time of each outlet, and sends the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet, including: Identify each track node address in the logistics track, and determine each network point according to the track node address; Identify the historical pickup and delivery time of each outlet corresponding to each outlet, and calculate the pickup and delivery time ratio according to the historical pickup and delivery time of each outlet; Based on the pick-up and delivery time ratio and the latest time, the arrival days of each outlet are allocated to obtain the estimated pick-up and delivery time of each outlet; The estimated arrival time at each outlet is calculated based on the starting time and the estimated pickup and delivery time at each outlet; The estimated arrival time of each outlet is sent to the customer, the courier and the mobile terminal of each outlet, so that the mobile terminal generates and displays the estimated arrival time page of each outlet.

7. The order trajectory prediction method based on SVM according to claim 1, It is characterized in that After allocating the arrival days of each outlet in the logistics track according to the latest duration to obtain the estimated arrival time of each outlet, and sending the estimated arrival time of each outlet to the customer, the courier and the mobile terminal of each outlet, the method further includes: Creating an actual arrival time listener, and using the actual arrival time listener to monitor the actual arrival time information of each network point; When the actual arrival time listener listens to the actual arrival time information, the current actual arrival time information is acquired; Analyze the deviation between the actual arrival time information and the estimated arrival time of each network point, and record the deviation value to obtain deviation information; The deviation information is uploaded to the blockchain.

8. A device for predicting order trajectory based on SVM, It is characterized in that include: A collection and processing module is used to collect logistics order data in real time, perform data preprocessing on the logistics order data, and obtain logistics order standard data; An identification query module, used to identify the waybill number in the logistics order standard data and the logistics track corresponding to the waybill number, and use the waybill number and the logistics track to query the earliest scanning time and predicted duration respectively; A calculation and judgment module, used to calculate the current duration according to the current real time and the earliest scanning time, and to judge whether the current duration is greater than the predicted duration; An analysis and update module, for analyzing the logistics order standard data using a support vector machine algorithm to obtain an analysis result, and updating the predicted duration according to the analysis result to obtain the latest duration; The allocation and sending module is used to allocate the arrival days of each outlet in the logistics track according to the latest duration, obtain the estimated arrival time of each outlet, and send the estimated arrival time of each outlet to the customer, courier and the mobile terminal of each outlet.

9. A device for predicting order trajectories based on SVM. It is characterized in that The order trajectory prediction device based on SVM includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors calls the instructions in the memory so that the device for predicting order trajectories based on SVM executes the various steps of the method for predicting order trajectories based on SVM as described in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, It is characterized in that When the instructions are executed by the processor, the steps of the order trajectory prediction method based on SVM as described in any one of claims 1 to 7 are implemented.

Citation Information

Cited By

  • Logistics tracking efficiency analysis method and system

    CN119477143A

  • Method and device for monitoring and tracking operation timeliness of logistics personnel

    CN119990939A

  • Third-party transport capacity resource scheduling method and device, storage medium and server

    CN120851563A