A network-based e-commerce system

Abstract

The present invention discloses a network-based e-commerce system, which relates to the field of logistics selection technology. The system records the goods currently to be selected for logistics distribution as the first goods, records the goods type corresponding to the first goods as the target goods type, and records each goods corresponding to the target goods type that have passed logistics distribution as the second goods; obtains the order status data of the first goods and the order status data of each second goods, calculates the similarity value of the order status data of the first goods and the order status data of each second goods, and filters out the target logistics type set; obtains the order distribution status data corresponding to each target logistics type to calculate the logistics distribution priority coefficient; distributes the first goods with the target logistics type corresponding to the largest logistics distribution priority coefficient; in this way, a suitable logistics distribution company can be selected according to the actual status of the new order, thereby reducing the damage to the goods or the generation of additional costs during the distribution process, and reducing the losses of merchants and customers.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics selection, and in particular to an e-commerce system based on a network. Background Art

[0002] Web-based e-commerce systems typically refer to systems that facilitate the transaction, payment, and delivery of goods and services through internet platforms. These systems encompass not only basic functions such as product browsing, selection, and payment, but also encompass multiple aspects, including order management, customer information processing, inventory management, and delivery scheduling, significantly improving merchants' operational efficiency. In traditional e-commerce models, logistics and delivery typically rely on merchants' own selected courier companies. With the surge in order volume and the diversification of customer demands, a single logistics option can no longer meet the delivery requirements of all orders. In existing web-based e-commerce systems, merchants are increasingly choosing to partner with multiple logistics companies to ensure high-quality order delivery.

[0003] In the actual selection of logistics and delivery companies for orders, network-based e-commerce systems often default to selecting the main logistics company that the merchant cooperates with or a lower-cost logistics service provider for the logistics and delivery company for new orders. However, different orders have different statuses. If the appropriate logistics and delivery company is not selected based on the actual status of the new order, it may cause damage to the goods or incur additional costs during the delivery process, causing significant losses to both the merchant and the customer. Summary of the Invention

[0004] The purpose of the present invention is to solve the above-mentioned problem that if the appropriate logistics distribution company is not selected according to the actual status of the new order, it may cause damage to the goods or additional costs during the distribution process, causing great losses to both merchants and customers, and to provide a network-based e-commerce system.

[0005] The present invention provides a network-based e-commerce system, the system comprising:

[0006] Marking module: record the goods currently to be selected for logistics distribution as the first goods, record the goods type corresponding to the first goods as the target goods type, and record each goods corresponding to the target goods type that have passed logistics distribution as the second goods;

[0007] Screening module: obtains the order status data of the first product and the order status data of each second product, calculates the similarity value between the order status data of the first product and the order status data of each second product, and screens out a set of logistics types that can deliver the first product based on the similarity value, which is recorded as the target logistics type set;

[0008] Delivery priority module: For each target logistics type in the target logistics type set, the corresponding order delivery status data is obtained, and the logistics delivery priority coefficient of each target logistics type is calculated based on the order delivery status data and the corresponding similarity value; the order delivery status data includes the delivery efficiency coefficient, delivery time instability coefficient, damage coefficient and cost coefficient;

[0009] Distribution module: The target logistics type corresponding to the largest logistics distribution priority coefficient in the target logistics type set is used as the final logistics type, and the first cargo is distributed.

[0010] Optionally, similarity values between the order status data of the first goods and the order status data of each second goods are calculated, and a set of logistics types that can deliver the first goods is screened based on the similarity values, which is recorded as a target logistics type set including:

[0011] Obtaining order status data of the first goods and order status data of each second goods, and preprocessing the order status data of the first goods and the order status data of each second goods;

[0012] Construct a distance matrix where each element is the distance between the order status data point of the first product and the order status data of the second product;

[0013] Calculating the shortest path in the distance matrix using a dynamic programming algorithm to find the best match between the order status data sequence of the first product and the order status data sequence of the second product;

[0014] Aligning the order status data sequence of the first product with the order status data sequence of the second product based on the calculated shortest path;

[0015] Calculating a similarity value between the aligned order status data sequence of the first product and the order status data sequence of the second product; and using the calculated similarity value as the similarity value corresponding to the second product;

[0016] According to the similarity values of each second cargo, a set of logistics types that can deliver the first cargo is screened out and recorded as a target logistics type set.

[0017] Optionally, a set of logistics types that can deliver the first goods is screened out based on the similarity values of each second goods, and is recorded as a target logistics type set, including:

[0018] Comparing the similarity value of each second product with a preset similarity value threshold, if the similarity value is not less than the preset similarity value threshold, the corresponding second product is recorded as a qualified product;

[0019] Each logistics type corresponding to the logistics distribution of the goods is recorded as the logistics type of the first deliverable goods, and all logistics types of the first deliverable goods are integrated to obtain a target logistics type set.

[0020] Optionally, the delivery efficiency coefficient includes;

[0021] Get the total number of times the second cargo is delivered for each target logistics type, and get the actual total time spent on each delivery and the corresponding preset total time spent, and mark the actual total time and the corresponding preset total time spent as D respectively. m and S m , m represents the sequence number of the number of times the second cargo is delivered by the target logistics type, m = 1, 2, 3, 4, ..., n, n represents the total number of times the second cargo is delivered by the target logistics type, n is a positive integer;

[0022] According to D m and S m Calculate the distribution efficiency coefficient corresponding to the target logistics type. The calculation formula is: Where DF is the distribution efficiency coefficient.

[0023] Optionally, the delivery time instability coefficient includes:

[0024] Get the total number of times the second cargo is delivered for each target logistics type, and get the total time actually spent on each delivery, and mark the total time actually spent as D m , m represents the sequence number of the number of times the second cargo is delivered by the target logistics type, m = 1, 2, 3, 4, ..., n, n represents the total number of times the second cargo is delivered by the target logistics type, n is a positive integer;

[0025] Calculate the average total time actually spent The calculation formula is:

[0026] Calculate the delivery time instability coefficient of the target logistics type. The calculation formula is:

[0027]

[0028] Among them, HT is the delivery time instability coefficient of the target logistics type.

[0029] Optionally, the damage factor includes:

[0030] Obtain the total number of times the second cargo is delivered for each target logistics type, and the total number of times the cargo is damaged after each delivery. Divide the total number of times the cargo is damaged by the total number of times the second cargo is delivered to obtain the damage rate of the delivered cargo for the target logistics type.

[0031] Obtain the preset damage severity score corresponding to each damaged cargo, and calculate the average of the preset damage severity scores as the damage severity value of the target logistics type;

[0032] The damage coefficient of the target logistics type is calculated based on the damage rate and damage severity value of the distribution goods of the target logistics type. The calculation formula is: ER = a1×zx+a2×cv, where ER is the damage coefficient of the target logistics type, zx and cv are the damage rate and damage severity value of the distribution goods of the target logistics type, a1 and a2 are the preset proportional coefficients of zx and cv, respectively, and a1 and a2 are both greater than 0.

[0033] Optionally, calculating the logistics delivery priority coefficient of each target logistics type based on the order delivery status data and the corresponding similarity value includes:

[0034] Obtain the total number of times the second cargo is delivered for each target logistics type, calculate the average cost of each delivery, and use the average cost as the cost coefficient of the target logistics type;

[0035] The logistics distribution priority coefficient of each target logistics type is calculated based on the distribution efficiency coefficient, distribution time instability coefficient, damage coefficient, cost coefficient and corresponding similarity values. The calculation formula is:

[0036]

[0037] Where Ycv is the logistics distribution priority coefficient, DF, HT, ER, RF, and KU are the distribution efficiency coefficient, distribution time instability coefficient, damage coefficient, and cost coefficient and their corresponding similarity values, respectively. f1, f2, f3, f4, and f5 are the preset proportional coefficients of DF, HT, ER, RF, and KU, respectively, and f1, f2, f3, f4, and f5 are all greater than 0.

[0038] Beneficial effects of the present invention:

[0039] 1. The present invention proposes a network-based e-commerce system, which records the goods currently to be selected for logistics distribution as first goods, records the goods type corresponding to the first goods as target goods type, and records each goods corresponding to the target goods type that have passed logistics distribution as second goods; obtains the order status data of the first goods and the order status data of each second goods, calculates the similarity value of the order status data of the first goods and the order status data of each second goods, and screens out a set of logistics types that can deliver the first goods based on the similarity value, and records it as the target logistics type set; this method can select the logistics method that best matches the current order based on the actual performance of historical orders, which not only improves the accuracy of the selection, but also ensures that the logistics distribution plan is more in line with actual needs, thereby improving distribution efficiency and cargo safety.

[0040] 2. For each target logistics type in the target logistics type set, obtain its corresponding order delivery status data, and calculate the logistics delivery priority coefficient of each target logistics type based on the order delivery status data and the corresponding similarity value; the target logistics type corresponding to the largest logistics delivery priority coefficient in the target logistics type set is used as the final logistics type, and the first cargo is delivered; in this way, the appropriate logistics delivery company can be selected according to the actual status of the new order, thereby reducing cargo damage or additional costs during the delivery process, and reducing losses for merchants and customers. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention will be further described below with reference to the accompanying drawings.

[0042] Figure 1 A framework diagram of a network-based e-commerce system. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0044] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0045] The embodiment of the present invention provides a network-based e-commerce system. Figure 1 , Figure 1 A framework diagram of a network-based e-commerce system provided in an embodiment of the present invention, the system comprising:

[0046] Marking module: record the goods currently to be selected for logistics distribution as the first goods, record the goods type corresponding to the first goods as the target goods type, and record each goods corresponding to the target goods type that have passed logistics distribution as the second goods;

[0047] Screening module: obtains the order status data of the first product and the order status data of each second product, calculates the similarity value between the order status data of the first product and the order status data of each second product, and screens out a set of logistics types that can deliver the first product based on the similarity value, which is recorded as the target logistics type set;

[0048] Delivery priority module: For each target logistics type in the target logistics type set, the corresponding order delivery status data is obtained, and the logistics delivery priority coefficient of each target logistics type is calculated based on the order delivery status data and the corresponding similarity value; the order delivery status data includes the delivery efficiency coefficient, delivery time instability coefficient, damage coefficient and cost coefficient;

[0049] Distribution module: The target logistics type corresponding to the largest logistics distribution priority coefficient in the target logistics type set is used as the final logistics type, and the first cargo is distributed.

[0050] Based on the network-based e-commerce system provided by the embodiment of the present invention, through the above method, it can select a suitable logistics distribution company according to the actual status of the new order, reduce the damage to goods or the generation of additional costs during the distribution process, and reduce the losses of merchants and customers.

[0051] In one embodiment, similarity values are calculated between the order status data of the first goods and the order status data of each second goods, and a set of logistics types that can deliver the first goods is screened based on the similarity values. The target logistics type set includes:

[0052] Obtaining order status data of the first goods and order status data of each second goods, and preprocessing the order status data of the first goods and the order status data of each second goods;

[0053] Construct a distance matrix where each element is the distance between the order status data point of the first product and the order status data of the second product;

[0054] Calculating the shortest path in the distance matrix using a dynamic programming algorithm to find the best match between the order status data sequence of the first product and the order status data sequence of the second product;

[0055] Aligning the order status data sequence of the first product with the order status data sequence of the second product based on the calculated shortest path;

[0056] Calculating a similarity value between the aligned order status data sequence of the first product and the order status data sequence of the second product; and using the calculated similarity value as the similarity value corresponding to the second product;

[0057] According to the similarity values of each second cargo, a set of logistics types that can deliver the first cargo is screened out and recorded as a target logistics type set.

[0058] In one embodiment, a set of logistics types that can deliver the first goods is screened based on the similarity values of each second goods, and is recorded as a target logistics type set, including:

[0059] Comparing the similarity value of each second product with a preset similarity value threshold, if the similarity value is not less than the preset similarity value threshold, the corresponding second product is recorded as a qualified product;

[0060] Each logistics type that meets the logistics distribution corresponding to the goods is recorded as the logistics type of the first deliverable goods, and all logistics types that can be delivered to the first deliverable goods are integrated to obtain a target logistics type set.

[0061] It should be noted that the order status data may include the length of delivery time of the goods in the order, the volume and weight of the goods, the total delivery distance, the ambient temperature and humidity during delivery, etc. These data not only help to evaluate the complexity of logistics distribution, but also reflect the special needs that the goods may encounter during transportation. For example, if the goods are perishable, the system will pay special attention to changes in temperature and humidity during the delivery process to ensure that they are transported in a suitable environment to avoid damage. In addition, the total delivery distance and the weight of the goods also directly affect the cost and timeliness requirements of transportation, so these factors are particularly important when choosing the most suitable logistics company. By comprehensively analyzing these order status data, the system can accurately match the appropriate logistics type to ensure that the goods are delivered on time and safely; the selection of specific order status data can be determined by professional personnel based on the actual situation, and there is no specific limitation here;

[0062] It should be noted that there are many ways to preprocess the order status data and calculate similarity values, depending on the type and complexity of the order status data. During the data preprocessing stage, the original data can be standardized, for example, the weight, volume, delivery time and other data of the goods can be converted into a unified dimension to facilitate subsequent comparison and calculation. In addition, data cleaning is also an important step, including processing missing values, outliers and noise data to ensure the accuracy and completeness of the data. For similarity calculation, methods such as dynamic time warping (DTW) or Euclidean distance can be used to measure the similarity between order status data. Dynamic time warping (DTW) is particularly suitable for time series data because it can find the best match between different time points, thereby reducing errors caused by time offsets. This method can efficiently evaluate the similarity between different orders, thereby more accurately selecting the most suitable logistics distribution method.

[0063] It should be noted that the order status data for the first product can be obtained from the order management system of the e-commerce platform. Usually, after a customer places an order, the system will automatically generate and record all information related to the order, including the type of goods, delivery requirements, order status, timeliness, destination, etc. The order status data for the second product can be obtained from historical order data. This data is usually stored in the system's database or data warehouse and contains relevant information during the execution of orders for similar goods or similar delivery requirements, such as delivery time, distance, cost, environmental conditions, etc. This historical data can be extracted using data mining and analysis tools and used to compare with the status of the current order to help the system select the most appropriate logistics and delivery method;

[0064] It should be noted that the preset similarity value threshold is set by professionals based on actual conditions and will not be limited or elaborated on in detail;

[0065] In one implementation, a dynamic programming algorithm is used to optimally align and match order status data, effectively eliminating potential time and data sequence discrepancies during order execution. This approach accurately captures the similarities in logistics requirements across orders, avoiding the potential biases introduced by simple direct comparisons. By calculating similarity values, the system can select the logistics method that best matches the current order based on the actual performance of historical orders. This not only improves the accuracy of the selection but also ensures that the logistics and delivery solutions are more aligned with actual needs, enhancing delivery efficiency and cargo safety. Therefore, this approach, based on dynamic programming and similarity calculation, can identify the optimal logistics and delivery solution within complex order environments, effectively enhancing the system's intelligence.

[0066] In one implementation, the benefit of the target logistics type set selected using the above method is that it can accurately match the most suitable logistics company and delivery method based on the specific status data of the order. By considering factors such as cargo volume, weight, delivery time, and environmental requirements, the system can select logistics types that have performed well in historical orders and can meet specific needs. This not only improves the timeliness and safety of delivery, reduces cargo damage or delays caused by improper logistics, but also reduces costs because the system selects the most cost-effective logistics service provider. Furthermore, intelligent screening based on historical data can continuously optimize logistics selection strategies, improve delivery efficiency and customer satisfaction, and thus enhance merchants' operational capabilities and market competitiveness.

[0067] In one embodiment, the delivery efficiency coefficient includes:

[0068] Get the total number of times the second cargo is delivered for each target logistics type, and get the actual total time spent on each delivery and the corresponding preset total time spent, and mark the actual total time and the corresponding preset total time spent as D respectively.m and S m , m represents the sequence number of the number of times the second cargo is delivered by the target logistics type, m = 1, 2, 3, 4, ..., n, n represents the total number of times the second cargo is delivered by the target logistics type, n is a positive integer;

[0069] According to D m and S m Calculate the distribution efficiency coefficient corresponding to the target logistics type. The calculation formula is: Where DF is the distribution efficiency coefficient.

[0070] It should be noted that the total number of times the target logistics type delivers the second cargo, as well as the actual time and preset time for each delivery, can be obtained in real time through synchronization with the interface of the logistics company system; specifically, the system will obtain the delivery time requirements (preset time) of the order from the order management module, and at the same time obtain the actual time for each delivery from the tracking system of the logistics service provider; after each delivery is completed, the logistics company will provide real-time feedback of relevant delivery information (such as delivery duration, start and end time, etc.) to the e-commerce platform, and the system will record this data for subsequent analysis; in addition, the preset time is determined according to the actual situation and is not limited or elaborated.

[0071] It should be noted that the delivery efficiency coefficient of a target logistics type refers to the difference between the actual delivery time and the preset delivery time for each target logistics type. The smaller the delivery efficiency coefficient, that is, the smaller the difference between the actual delivery time and the preset delivery time for the target logistics type, the more suitable the target logistics type is for the delivery of the current selected logistics type, namely, the first cargo, and the higher its delivery selection priority. This is because a smaller delivery efficiency coefficient means that the difference between the actual delivery time and the preset delivery time for the target logistics type is smaller, indicating that the logistics company can accurately and consistently deliver within the expected time. This accuracy and stability are crucial for merchants, especially in terms of efficiency and cost control. For example, the preset delivery time is often set based on multiple factors such as order type, destination, and traffic conditions, and can effectively predict delivery time. If the target logistics type can accurately match this time expectation, it indicates high efficiency and reliability in planning and execution, effectively avoiding delays and additional costs, thereby improving the efficiency of the overall logistics process and customer satisfaction. Therefore, logistics types with lower delivery efficiency coefficients usually mean higher service quality and can better meet the timeliness requirements of order delivery. Merchants will also give priority to these logistics types to improve the accuracy and efficiency of transportation, ensure on-time delivery and reduce potential risks.

[0072] In one implementation, analyzing the delivery efficiency coefficient of the target logistics type is beneficial in selecting the target logistics type for the delivery of the currently selected goods. This helps the system accurately evaluate each logistics type's performance in historical orders, particularly its time control capabilities. Logistics types with lower delivery efficiency coefficients generally indicate better control over delivery times during execution, avoiding additional costs or customer dissatisfaction caused by delays. By selecting a logistics type with a lower delivery efficiency coefficient, merchants can improve delivery accuracy and reliability, ensuring that goods arrive at their destination on time and reducing losses caused by inaccurate transit times. This selection also helps optimize transportation resources, reduce unnecessary expenses, and improve the overall efficiency of logistics operations. Ultimately, this selection based on the delivery efficiency coefficient can enhance the customer experience, strengthen merchants' competitiveness, and further promote customer loyalty and repurchase rates.

[0073] In one embodiment, the delivery time instability coefficient includes:

[0074] Get the total number of times the second cargo is delivered for each target logistics type, and get the total time actually spent on each delivery, and mark the total time actually spent as D m , m represents the sequence number of the number of times the second cargo is delivered by the target logistics type, m = 1, 2, 3, 4, ..., n, n represents the total number of times the second cargo is delivered by the target logistics type, n is a positive integer;

[0075] Calculate the average total time actually spent The calculation formula is:

[0076] Calculate the delivery time instability coefficient of the target logistics type. The calculation formula is:

[0077]

[0078] Among them, HT is the delivery time instability coefficient of the target logistics type.

[0079] It should be noted that the delivery time instability coefficient of a target logistics type refers to the degree of instability in the actual delivery time taken by each target logistics type for each second shipment. The higher the instability, i.e., the larger the delivery time instability coefficient of a target logistics type, the less suitable the target logistics type is for the current shipment, i.e., the first shipment, and the lower its priority for delivery selection. This is because a larger delivery time instability coefficient indicates greater fluctuations in the delivery time taken by the target logistics type, which generally indicates that the logistics company exhibits significant time instability, potentially leading to delivery time variations due to various factors. For e-commerce systems, especially those handling high-value or time-sensitive shipments, unstable delivery times can lead to customer dissatisfaction and even negatively impact customer experience and brand reputation. If the delivery time of a target logistics type exhibits significant uncertainty, merchants face increased risk. This inability to accurately predict when shipments will arrive at their destination can lead to delays, stockouts, or other service failures, impacting overall operational efficiency. Therefore, the system should prioritize logistics types with greater time stability and less volatility to ensure on-time delivery and avoid potential customer loss and additional costs caused by instability.

[0080] In one implementation, analyzing the delivery time instability coefficient of the target logistics type has the benefit of selecting the target logistics type for the delivery of the goods currently to be selected, i.e., a certain goods, in that it can help merchants avoid selecting logistics service providers that have large time fluctuations in the delivery process. The stability of delivery time is crucial to customer satisfaction, especially when orders need to be delivered within a specific time limit. The uncertainty of delivery time may lead to customer dissatisfaction and complaints, affecting the merchant's reputation. By selecting a logistics type with a smaller delivery time instability coefficient, merchants can ensure that the timeliness of the delivery process is more reliable, reduce the additional costs and risks caused by delivery delays or uncertainties, and improve the reliability and controllability of the overall delivery; ultimately, stable delivery time can improve customer experience and reduce volatility in logistics operations.

[0081] In one embodiment, the damage coefficients include:

[0082] Obtain the total number of times the second cargo is delivered for each target logistics type, and the total number of times the cargo is damaged after each delivery. Divide the total number of times the cargo is damaged by the total number of times the second cargo is delivered to obtain the damage rate of the delivered cargo for the target logistics type.

[0083] Obtain the preset damage severity score corresponding to each damaged cargo, and calculate the average of the preset damage severity scores as the damage severity value of the target logistics type;

[0084] The damage coefficient of the target logistics type is calculated based on the damage rate and damage severity value of the distribution goods of the target logistics type. The calculation formula is: ER = a1×zx+a2×cv, where ER is the damage coefficient of the target logistics type, zx and cv are the damage rate and damage severity value of the distribution goods of the target logistics type, a1 and a2 are the preset proportional coefficients of zx and cv, respectively, and a1 and a2 are both greater than 0.

[0085] It should be noted that a1 and a2 are set by professionals based on actual circumstances. Generally, the sum of a1 and a2 is 1. For example, a1 and a2 can be 0.5 and 0.5 respectively, and the specific values are not limited. In addition, before calculating the damage coefficient, the damage rate and damage severity value of the distributed goods of each target logistics type need to be normalized. Commonly used normalization methods include Min-Max normalization and Z-Score standardization. The specific method is selected and determined by professionals based on actual circumstances and is not limited.

[0086] It should be noted that data on the total number of deliveries and the total number of damaged goods can be obtained through logistics management systems or warehouse management systems. These systems record the details of each delivery, including whether damage occurred and the number of damages. Secondly, the data for the damage severity score is usually derived from the evaluation of each damage incident. Staff or automated detection systems can assign a severity score to each damage incident based on the actual damage situation. This score can be based on a preset standard, categorized and assessed according to the degree of damage (e.g., minor damage, moderate damage, severe damage, etc.); the specific degree of damage and damage severity score are determined by professionals based on the actual situation and are not limited or elaborated upon.

[0087] It should be noted that the damage coefficient refers to the probability and severity of damage to the second shipment when delivered by the target logistics type. A higher probability and severity of damage, i.e., a higher damage coefficient, indicates that the target logistics type is less suitable for the delivery of the first shipment, the current shipment being selected, and its priority for delivery is lower. This is because a higher damage coefficient for a target logistics type indicates a higher probability and severity of damage during delivery. Damaged goods can lead to additional losses and unnecessary expenses during delivery, such as replacements, customer compensation, or subsequent processing, all of which increase costs for both merchants and consumers. More importantly, damaged goods can negatively impact the customer experience, potentially leading to customer dissatisfaction, brand damage, and customer churn. Therefore, logistics types with higher damage coefficients carry higher risks and should be prioritized lower. To ensure safe and intact delivery of goods, minimizing damage risk is a key consideration when selecting logistics and delivery services. For this reason, merchants tend to choose logistics types with lower damage coefficients to minimize the probability of damage during delivery and avoid the additional financial losses and customer complaints caused by severe damage.

[0088] In one implementation, analyzing the damage coefficient of the target logistics type has the following advantages for selecting the target logistics type for the delivery of the currently selected goods: logistics types with higher damage coefficients indicate a greater probability of damage and severity during delivery, which directly impacts the safety of the goods and the merchant's reputation. Therefore, by calculating and analyzing the damage coefficient, it is possible to effectively select logistics methods that offer lower risks and ensure the safe delivery of goods. Selecting a logistics type with a lower damage coefficient means that the logistics service provider is better able to protect the goods during transportation, reducing the likelihood of damage and ensuring that customers receive their goods intact. This not only improves customer satisfaction but also reduces compensation and after-sales costs for the merchant. Furthermore, analyzing the damage coefficient helps optimize logistics distribution strategies, avoiding logistics companies with high damage risks, thereby reducing potential economic losses during the logistics process and improving overall delivery efficiency and reliability.

[0089] In one embodiment, calculating the logistics delivery priority coefficient of each target logistics type based on the order delivery status data and the corresponding similarity value includes:

[0090] Obtain the total number of times the second cargo is delivered for each target logistics type, calculate the average cost of each delivery, and use the average cost as the cost coefficient of the target logistics type;

[0091] The logistics distribution priority coefficient of each target logistics type is calculated based on the distribution efficiency coefficient, distribution time instability coefficient, damage coefficient, cost coefficient and corresponding similarity values. The calculation formula is:

[0092]

[0093] Where Ycv is the logistics distribution priority coefficient, DF, HT, ER, RF, and KU are the distribution efficiency coefficient, distribution time instability coefficient, damage coefficient, and cost coefficient and their corresponding similarity values, respectively. f1, f2, f3, f4, and f5 are the preset proportional coefficients of DF, HT, ER, RF, and KU, respectively, and f1, f2, f3, f4, and f5 are all greater than 0.

[0094] It should be noted that f1, f2, f3, f4, and f5 are set by professionals according to actual circumstances. Generally, the sum of f1, f2, f3, f4, and f5 is 1. For example, f1, f2, f3, f4, and f5 can be 0.2, 0.2, 0.2, 0.2, and 0.2, respectively. They can also be other numbers and are not limited to any specific number. In addition, before calculating the logistics distribution priority coefficient, it is necessary to normalize the distribution efficiency coefficient, distribution time instability coefficient, damage coefficient, cost coefficient, and corresponding similarity values. Commonly used normalization methods include Min-Max normalization, Z-Score standardization, etc. The specific method is selected and determined by professionals according to the actual situation and is not limited to any specific number.

[0095] It's important to note that the cost coefficient calculation involves the costs of each delivery, which can be obtained by analyzing historical delivery records. These costs typically include transportation costs, labor costs, fuel costs, vehicle maintenance costs, storage fees, and so on. Specifically, merchants can extract the actual costs of each delivery from logistics management systems, transportation service providers' bills, and contract data, or partner with third-party logistics companies to obtain delivery cost data. Furthermore, by regularly tracking and recording various expenses, merchants can accurately calculate the average cost per delivery, thereby deriving the cost coefficient for each target logistics type.

[0096] It should be noted that, as shown in the above calculation expression, the smaller the target logistics type's delivery efficiency coefficient, delivery time instability coefficient, damage coefficient, and cost coefficient, and the larger the corresponding similarity value, the more suitable the target logistics type is for delivering the first cargo. This is because a smaller delivery efficiency coefficient indicates that the target logistics type can complete the delivery task with minimal time deviation, demonstrating its efficiency; a smaller delivery time instability coefficient indicates that time fluctuations during the delivery process are minimal, ensuring logistics stability and predictability; a smaller damage coefficient indicates that the logistics type has a low risk of damage to the cargo during delivery, ensuring cargo safety; and a smaller cost coefficient indicates that the target logistics type can complete delivery at a lower cost, helping to save the company's logistics expenses. A larger similarity value indicates that the logistics type's historical delivery status is more closely aligned with the current cargo delivery needs. In summary, when these factors are all favorable, the target logistics type performs well in terms of delivery efficiency, stability, safety, and cost-effectiveness, making it more suitable for the current cargo delivery.

[0097] In one implementation, the target logistics type with the highest logistics distribution priority coefficient among the target logistics types is selected as the final logistics type for delivery of the first shipment. This approach benefits from selecting the logistics type with the highest logistics distribution priority coefficient within the target logistics type set, ensuring that the selected logistics method performs optimally in terms of delivery efficiency, stability, safety, and cost. By comprehensively considering key factors such as delivery efficiency, time stability, damage risk, and delivery cost, delivery quality is optimized, avoiding bias from a single factor. Selecting this logistics type effectively reduces risks during delivery, improves cargo safety, and improves overall logistics operational efficiency by reducing costs, ensuring the optimal delivery of the first shipment. Applying this approach not only improves customer satisfaction but also conserves logistics resources for the company, enabling long-term sustainable development.

[0098] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be used to artificially limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A network-based e-commerce system, characterized in that: The system includes: Marking module: record the goods currently to be selected for logistics distribution as the first goods, record the goods type corresponding to the first goods as the target goods type, and record each goods corresponding to the target goods type that have passed logistics distribution as the second goods; Screening module: obtains the order status data of the first product and the order status data of each second product, calculates the similarity value between the order status data of the first product and the order status data of each second product, and screens out a set of logistics types that can deliver the first product based on the similarity value, which is recorded as the target logistics type set; Delivery priority module: For each target logistics type in the target logistics type set, the corresponding order delivery status data is obtained, and the logistics delivery priority coefficient of each target logistics type is calculated based on the order delivery status data and the corresponding similarity value; the order delivery status data includes the delivery efficiency coefficient, delivery time instability coefficient, damage coefficient and cost coefficient; Distribution module: The target logistics type corresponding to the largest logistics distribution priority coefficient in the target logistics type set is used as the final logistics type, and the first cargo is distributed; The logistics distribution priority coefficients of each target logistics type are calculated based on the order delivery status data and the corresponding similarity values, including: Obtain the total number of times the second cargo is delivered for each target logistics type, calculate the average cost of each delivery, and use the average cost as the cost coefficient of the target logistics type; The logistics distribution priority coefficient of each target logistics type is calculated based on the distribution efficiency coefficient, distribution time instability coefficient, damage coefficient, cost coefficient and corresponding similarity values. The calculation formula is: Where Ycv is the logistics distribution priority coefficient, DF, HT, ER, RF, and KU are the distribution efficiency coefficient, distribution time instability coefficient, damage coefficient, and cost coefficient and their corresponding similarity values, respectively. f1, f2, f3, f4, and f5 are the preset proportional coefficients of DF, HT, ER, RF, and KU, respectively, and f1, f2, f3, f4, and f5 are all greater than 0.

2. A network-based e-commerce system according to claim 1, characterized in that: Calculate the similarity between the order status data of the first product and the order status data of each second product, and filter out the logistics type set that can deliver the first product based on the similarity, which is recorded as the target logistics type set, including: Obtaining order status data of the first goods and order status data of each second goods, and preprocessing the order status data of the first goods and the order status data of each second goods; Construct a distance matrix where each element is the distance between the order status data point of the first product and the order status data of the second product; Calculating the shortest path in the distance matrix using a dynamic programming algorithm to find the best match between the order status data sequence of the first product and the order status data sequence of the second product; Aligning the order status data sequence of the first product with the order status data sequence of the second product based on the calculated shortest path; Calculating a similarity value between the aligned order status data sequence of the first product and the order status data sequence of the second product; and using the calculated similarity value as the similarity value corresponding to the second product; According to the similarity values of each second cargo, a set of logistics types that can deliver the first cargo is screened out and recorded as a target logistics type set.

3. A network-based e-commerce system according to claim 2, characterized in that: Based on the similarity values of each second cargo, a set of logistics types that can deliver the first cargo is selected, which is recorded as the target logistics type set and includes: Comparing the similarity value of each second product with a preset similarity value threshold, if the similarity value is not less than the preset similarity value threshold, the corresponding second product is recorded as a qualified product; Each logistics type corresponding to the logistics distribution of the goods is recorded as the logistics type of the first deliverable goods, and all logistics types of the first deliverable goods are integrated to obtain a target logistics type set.

4. A network-based e-commerce system according to claim 1, characterized in that: The distribution efficiency coefficient includes; Get the total number of times the second cargo is delivered for each target logistics type, and get the actual total time spent on each delivery and the corresponding preset total time spent, and mark the actual total time and the corresponding preset total time spent as D respectively. m and S m , m represents the sequence number of the number of times the second cargo is delivered by the target logistics type, m = 1, 2, 3, 4, ..., n, n represents the total number of times the second cargo is delivered by the target logistics type, n is a positive integer; According to D m and S m Calculate the distribution efficiency coefficient corresponding to the target logistics type. The calculation formula is: Where DF is the distribution efficiency coefficient.

5. A network-based e-commerce system according to claim 1, characterized in that: The delivery time instability coefficient includes: Get the total number of times the second cargo is delivered for each target logistics type, and get the total time actually spent on each delivery, and mark the total time actually spent as D m , m represents the sequence number of the number of times the second cargo is delivered by the target logistics type, m = 1, 2, 3, 4, ..., n, n represents the total number of times the second cargo is delivered by the target logistics type, n is a positive integer; Calculate the average total time actually spent The calculation formula is: Calculate the delivery time instability coefficient of the target logistics type. The calculation formula is: Among them, HT is the delivery time instability coefficient of the target logistics type.

6. A network-based e-commerce system according to claim 1, characterized in that: Damage factors include: Obtain the total number of times the second cargo is delivered for each target logistics type, and the total number of times the cargo is damaged after each delivery. Divide the total number of times the cargo is damaged by the total number of times the second cargo is delivered to obtain the damage rate of the delivered cargo for the target logistics type. Obtain the preset damage severity score corresponding to each damaged cargo, and calculate the average of the preset damage severity scores as the damage severity value of the target logistics type; The damage coefficient of the target logistics type is calculated based on the damage rate and damage severity value of the distribution goods of the target logistics type. The calculation formula is: ER = a1×zx+a2×cv, where ER is the damage coefficient of the target logistics type, zx and cv are the damage rate and damage severity value of the distribution goods of the target logistics type, a1 and a2 are the preset proportional coefficients of zx and cv, respectively, and a1 and a2 are both greater than 0.

Images