Supply chain traceability system and method based on large model

By adopting large-scale model technology in the supply chain management system, establishing prediction models and time models, and calculating cargo transportation costs, the problem of low efficiency of manual estimation costs in traditional systems is solved, and fast and accurate transportation cost prediction and supply chain management are achieved. optimization.

CN119090512AInactive Publication Date: 2024-12-06SHANGHAI MICHI INFORMATION TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411090810.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In traditional supply chain management systems, the transportation cost estimates of different products require a lot of manual accounting, which is inefficient and prone to errors, especially when there are many types of goods.

Method used

Using a supply chain traceability system based on large models, we can obtain and classify historical cargo transportation cost data, establish a predictive model, calculate the cargo transportation cost, including cargo type matching degree, time cost and equipment tool cost, and calculate the vehicle return time when no load is available. fuel consumption realizes automated transportation cost prediction.

Benefits of technology

It improves the speed and accuracy of transportation cost calculation, reduces manual errors, and can reasonably plan shipment time, optimize supply chain management, and reduce transportation costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure BDA0004986510330000021
    Figure BDA0004986510330000021
  • Figure BDA0004986510330000022
    Figure BDA0004986510330000022
  • Figure BDA0004986510330000031
    Figure BDA0004986510330000031
Patent Text Reader

Abstract

The invention relates to a supply chain traceability system and method based on a large model, and the method specifically comprises the following steps: obtaining the cargo transportation cost of all different types of historical cargos in a supply chain, carrying out the classification extraction of all different types of historical cargos, and obtaining the data of a plurality of different types of cargos; calling data of all different types of goods, and then performing one-to-one correspondence with the goods transportation cost of all different types of goods in the historical supply chain; and based on the data of all different types of goods and the goods transportation cost of all different types of goods. According to the scheme, the cost result of each cargo is obtained based on statistical analysis of the cargo transportation cost of all different types of cargos in the supply chain history, and when the transportation cost of different types of cargos is calculated in the supply chain, the transportation cost of the different types of cargos is calculated based on the prediction model and the cargo types according to the cargo transportation quantity. Therefore, the transportation cost of the goods can be effectively predicted, and compared with manual calculation, the speed is high and the error is small.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain management, and in particular to a supply chain tracing system and method based on a large model. Background Art

[0002] The supply chain is a logistics network composed of suppliers, manufacturers, warehouses, distribution centers and channel dealers. The same enterprise may constitute different nodes of the logistics network, but in more cases, different enterprises constitute different nodes in this network. The supply chain is actually composed of all the node enterprises (or enterprise units) that have joined, among which there is generally a core enterprise. Driven by demand information, the node enterprises (or enterprise units) realize the continuous value-added of the entire supply chain through the division of labor and cooperation of supply chain functions (manufacturing, transshipment, distribution, retail, etc.). Therefore, the so-called supply chain management refers to the management method of product manufacturing, transshipment, distribution and sales that effectively organizes suppliers, manufacturers, warehouses, distribution centers and channel dealers to minimize the cost of the entire supply chain system under the condition of meeting a certain level of customer service.

[0003] Chinese patent CN202310083743.9 discloses a supply chain transportation traceability management system, including a logistics data room, a vehicle dispatch management platform, a mobile network and a positioning satellite. The upper end of the logistics data room publishes transportation requirements through the logistics client, and the transportation requirements transmit data to the vehicle dispatch management platform through the logistics data room. The vehicle dispatch management platform dispatches the driver for processing. After the driver receives the order, he logs into the system through his mobile phone and randomly generates a dynamic QR code. The code is scanned and authenticated by the camera at the sign-in place. The code scanning authentication includes information entry. By setting up vehicle-mounted positioning, full-process cargo tracking and trajectory playback are achieved. At the same time, the cargo traceability is supervised by one vehicle, one code, and one object, one code, and the whole traceability process is supervised. Strictly abide by the safety of hazardous chemicals transportation to carry out refined monitoring of safe cargo management and transportation management, which is convenient for real-time monitoring and search.

[0004] However, in supply chain transportation management, companies need to inquire about the actual cost of commodity transportation. Based on the cost, they can estimate the profit that the company can obtain after the product transportation is completed, and then make decisions on the shipment of products in the supply chain based on the profit. In traditional management systems, different products need to be manually estimated for the actual cost each time they are shipped. In actual shipment and transportation, the types of products are different, and the production costs and transportation costs are also different. If there are many types of goods during accounting, a lot of manual accounting is required, which is not only inefficient but also prone to errors. Summary of the invention

[0005] In view of the technical problem raised in the background technology that in traditional management systems, each time different products are shipped, the actual cost required needs to be estimated manually. In actual shipment and transportation, different types of products have different production costs and transportation costs. If there are many types of goods during accounting, a large amount of manual accounting is required, which is not only inefficient but also prone to errors. The present invention provides a supply chain traceability system and method based on a large model.

[0006] The technical solution adopted by the present invention is: a supply chain traceability system based on a large model, comprising:

[0007] The traceability module traces the goods in the supply chain and stores all the goods supply chain data;

[0008] The acquisition module is used to obtain the cargo transportation costs of all different types of cargoes in the history of the enterprise, and classify and extract all different types of cargoes in the history to obtain data of several different types of cargoes;

[0009] The retrieval module retrieves all different types of cargo data, and then matches them one by one with the historical cargo transportation costs of all different types of cargo;

[0010] A construction module is used to establish a prediction model based on all different types of cargo data and cargo transportation costs of all different types of cargo. The input of the prediction model is the cargo type, and the output of the prediction model is the cost result of each cargo.

[0011] The output module inputs the cargo type and cargo quantity into the prediction model, obtains the cost results, and recommends the final transportation cost based on the cost results.

[0012] Furthermore, based on all different types of cargo data and cargo transportation costs of all different types of cargo, a prediction model is established, specifically including:

[0013] Establish data on all different types of goods, and then match them one by one with the historical freight transportation costs of all different types of goods;

[0014] Based on the one-to-one correspondence between all different types of cargo data and the cargo transportation costs of all different types of cargo, statistical analysis is performed on the cargo transportation costs corresponding to each type of cargo, and the relevant weights between each type of cargo and the corresponding cargo transportation costs are obtained respectively;

[0015] Establish a sub-model, which is used to calculate the matching degree between the input cargo type and all different types of cargo in history;

[0016] A screening model is established. The screening model screens out one or more of all different types of historical cargo data that best matches the input cargo type based on the calculation results of the sub-model, and outputs them as cost results.

[0017] Furthermore, the calculation formula for the relevant weight between each type of goods and the cost of corresponding goods transportation is:

[0018]

[0019] In the formula, p n is the correlation weight between the nth type of goods and the corresponding transportation cost, x 总 is the total freight transportation cost of all different types of goods in history corresponding to the goods type, x n It is the total number of the nth historical freight transportation costs of all different types of freight corresponding to the freight type.

[0020] The further analysis design is as follows: The specific calculation formula of the sub-model is:

[0021]

[0022] Among them, Q j is the matching degree between the input j-th type of goods and the corresponding transportation cost, X j is the total number of historical cargo transportation cost data for the jth cargo type, X0 is the total number of cargo types input, and X 同 is the total cost of historical cargo transportation corresponding to the input cargo type data and the jth cargo type, p' l is the correlation weight between the j-th type of goods and the same features in the l-th input type of goods and the cost of the j-th historical goods transportation, and max() is the maximum value function.

[0023] Furthermore, the type and quantity of goods are input into the forecasting model to obtain the cost results, including:

[0024] Calculate the quantity of the input goods type and quantity;

[0025] Call the sub-model to calculate the matching degree between the input goods and the corresponding goods transportation cost based on several goods types;

[0026] Calling the screening model to screen out one or more historical cargo types whose matching degree with the input cargo type is greater than a preset value as the predicted historical cargo types;

[0027] The historical cost of the type of goods will be retrieved and then multiplied by the quantity of goods to obtain the transportation cost.

[0028] Furthermore, the cost of cargo transportation also includes the time cost calculation of the cargo, as follows:

[0029] Obtain the time cost of all different types of goods in the company's history;

[0030] Retrieve all different types of goods in history and make a one-to-one correspondence with the time costs of all different types of goods in history;

[0031] Based on all different types of goods and their corresponding time costs in history, a time model is established. The input of the time model is the type of goods, and the output of the time model is the time cost of each goods transportation;

[0032] Input the cargo type into the time model and obtain the time result.

[0033] Furthermore, based on all different types of goods in history and their corresponding time costs, the calculation formula for establishing the time model is as follows:

[0034]

[0035] Where: f t represents the time cost of goods, V k represents the value of the goods, p represents the cost coefficient of the goods, T represents the transportation time of the goods (hours), and the method for determining the cost coefficient of the goods holding is the Aford-bangs model. The Alford-bangs model considers the factors affecting the time value of the goods from multiple aspects. The Alford-bangs model can be expressed as p = p1 + p2 + p3 ... + p n Where p represents the coefficient of goods holding cost in a certain year (%); p1 to p n Indicates influencing factors.

[0036] Furthermore, the cost of transporting goods also includes the cost of equipment and tools used to transport the goods, as follows:

[0037] Calculate the average annual expenditure during the life of the equipment and tools;

[0038] Assume A is the original value of the equipment, B is the estimated net residual value of the equipment, λ is the added value of the equipment due to deterioration, T is the useful life of the equipment, and K is the average annual expenditure during the useful life of the equipment.

[0039] The calculation formula is as follows:

[0040]

[0041] Furthermore, the cost of transporting goods also includes the calculation of fuel consumption when the transport vehicle returns empty;

[0042] When the vehicle is a truck;

[0043] The formula for calculating the fuel consumption of a truck on its return trip when empty is:

[0044]

[0045] Where: D is the fuel consumption of the truck when it is empty, I is the fuel consumption of the truck when it is fully loaded, h is the wind resistance, which is taken as 0.7; V, V1, and V2 are the vehicle's own weight, cargo weight, and fully loaded cargo weight, respectively.

[0046] Furthermore, a supply chain traceability method based on a large model includes:

[0047] S1. Obtain the historical freight transportation costs of all different types of freight in the supply chain, and classify and extract all different types of freight in history to obtain data of several different types of freight;

[0048] S2. Retrieve all different types of cargo data, and then match them one by one with the cargo transportation costs of all different types of cargo in the historical supply chain;

[0049] S3. Based on all different types of cargo data and cargo transportation costs of all different types of cargo, a prediction model is established, wherein the input of the prediction model is the cargo type, and the output of the prediction model is the cost result of each cargo;

[0050] S4. Input the type and quantity of goods into the forecasting model to obtain the cost result;

[0051] S5. Based on the cost results, recommend the final transportation cost.

[0052] The beneficial effects of the present invention are:

[0053] 1. Compared with the prior art, in the present invention, this solution is based on the statistical analysis of the freight transportation costs of all different types of goods in the supply chain history to obtain the cost results of each commodity. The freight transportation costs of different types of goods in the supply chain are calculated based on the prediction model and the type of goods. Then, according to the quantity of goods transported, the freight transportation cost can be effectively predicted. Compared with manual calculation, the speed is fast and the error is small.

[0054] 2. Compared with the prior art, the present invention establishes a time model based on all different types of goods and the corresponding time costs in history. Therefore, in subsequent supply chain inquiries, it is possible to predict the approximate time cost of transporting different types of goods, and to reasonably plan the delivery time of different types of goods based on the time cost, so that the receiving company can receive the goods in time and increase customer satisfaction.

[0055] 3. Compared with the prior art, the present invention can calculate the cost of different types of equipment and tools used in cargo transportation, thereby adding to the cost of goods, which is used by enterprises to fully calculate the various costs used in the entire transportation process.

[0056] 4. Compared with the prior art, in the present invention, by calculating the fuel consumption of the vehicle (truck) when empty and using it for return trip calculation, the empty fuel consumption and the journey can be calculated into the cost of cargo transportation, thereby making it easier for enterprises to understand the monetary cost required for cargo transportation and facilitating supply chain management optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a flow chart of the present invention;

[0058] Figure 2 A specific flow chart for establishing a prediction model based on all different types of cargo data and cargo transportation costs of all different types of cargo in the present invention;

[0059] Figure 3 A specific flow chart for inputting the type and quantity of goods into the prediction model and obtaining the cost result in the present invention;

[0060] Figure 4 This is a specific flow chart for calculating the time cost of goods transportation in the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] Please refer to the attached Figure 1-Figure 4 In order to solve the problems existing in the background technology, the present application proposes the following technical solution: a supply chain traceability method based on a large model, specifically comprising the following steps:

[0063] S101, obtain the historical transportation costs of all different types of goods in the supply chain, and classify and extract all different types of goods in the history to obtain data of several different types of goods (goods types include wine, furniture, etc. For example, if the transportation cost of each bottle of red wine is 50 yuan, then the transportation quantity*50 is the transportation cost of the red wine)

[0064] S102, retrieve all different types of cargo data, and then match them one by one with the cargo transportation costs of all different types of cargo in the supply chain history;

[0065] S103, based on all different types of cargo data and cargo transportation costs of all different types of cargo, a prediction model is established, the input of the prediction model is the cargo type, and the output of the prediction model is the cost result of each cargo (the cargo type and the cargo quantity to be transported are input, and then multiplied with each other, the cargo transportation cost can be obtained through the prediction model);

[0066] S104, input the type and quantity of goods into the forecasting model to obtain the cost result;

[0067] S105, based on the cost result, recommend the final transportation cost;

[0068] This solution is based on the statistical analysis of the historical transportation costs of all different types of goods in the enterprise to obtain the cost results for each product. When the enterprise calculates the transportation costs of different types of goods, it is based on the prediction model and the type of goods, and then based on the quantity of goods transported, the transportation cost of the goods can be effectively predicted. Compared with manual calculation, it is fast and has a small error.

[0069] Further designs also include:

[0070] S101, based on all different types of cargo data and cargo transportation costs of all different types of cargo, establishing a prediction model specifically includes:

[0071] S102, establishing data of all different types of goods, and then making a one-to-one correspondence with the historical freight transportation costs of all different types of goods;

[0072] S103, based on a one-to-one correspondence between all different types of cargo data and cargo transportation costs of all different types of cargo, statistical analysis is performed on the cargo transportation costs corresponding to each type of cargo, and a correlation weight between each type of cargo and the corresponding cargo transportation cost is obtained respectively;

[0073] S104, establishing a sub-model, where the sub-model is used to calculate the matching degree between the input cargo type and all different types of cargo in history;

[0074] S105, establishing a screening model, which screens out one or more of all different types of historical cargo data that best matches the input cargo type based on the calculation results of the sub-model, and outputs the result as the cost result.

[0075] Based on the comprehensive statistical analysis of all different types of cargo data and the cargo transportation costs of all different types of cargo, it is finally possible to calculate the transportation cost according to the type of each cargo, and then calculate and sum the transportation costs of multiple cargoes, which is the basic cargo transportation cost required for this transportation. At the same time, by calculating the relevant weights between each cargo and the corresponding cargo transportation cost, they are used as the data basis for matching degree calculation in subsequent queries, thereby realizing the effective reuse of all different types of cargo data and improving the accuracy of cargo cost calculation.

[0076] The calculation formula for the relevant weight between each type of goods and the corresponding transportation cost is:

[0077]

[0078] In the formula, p n is the correlation weight between the nth type of goods and the corresponding transportation cost, x 总 is the total freight transportation cost of all different types of goods in history corresponding to the goods type, x n It is the total number of the nth historical freight transportation costs of all different types of freight corresponding to the freight type.

[0079] The following is a calculation formula for the relevant weight between each of the above-mentioned goods and the corresponding goods transportation cost in conjunction with a specific embodiment;

[0080] In some embodiments, the total number of all historical transportation costs corresponding to Goods A is 5000, of which 3985 historical transportation cost total data include Goods A, then the correlation weight between Goods A and the cost b of goods transportation is 3985 / 5000=0.797.

[0081] The specific calculation formula of the sub-model is:

[0082]

[0083] Among them, Q j is the matching degree between the input j-th type of goods and the corresponding transportation cost, X j is the total number of historical cargo transportation cost data for the jth cargo type, X0 is the total number of cargo types input, and X 同 is the total cost of historical cargo transportation corresponding to the input cargo type data and the jth cargo type, p' l is the correlation weight between the j-th type of goods and the same features in the l-th input type of goods and the cost of the j-th historical goods transportation, and max() is the maximum value function.

[0084] In this solution, the matching degree is calculated based on two indicators. The transportation cost can be accurately predicted through the matching degree between the j-th type of goods and the corresponding transportation cost.

[0085] In further design optimization, the type and quantity of goods are input into the prediction model to obtain the cost results, including:

[0086] S101, performing quantity calculation on the input goods type and goods quantity;

[0087] S102, calling the sub-model to calculate the matching degree between the input goods and the corresponding goods transportation cost based on several goods types;

[0088] S103, calling the screening model to screen out one or more historical cargo types whose matching degree with the input cargo type is greater than a preset value as the predicted historical cargo types;

[0089] S104, retrieve the historical cost of the goods type, and then multiply it by the quantity of goods to obtain the transportation cost.

[0090] In this solution, the transportation cost of the goods to be calculated is calculated through screening, and then the quantity is calculated. Finally, by calling the screening model, one or more historical goods types whose matching degree with the input goods type is greater than the preset value are screened out as the predicted historical goods types. Then, the corresponding historical goods transportation cost is retrieved, so that the transportation cost of the same type of goods that currently need to be transported can be calculated, so that the actual cost calculation can be performed.

[0091] To sum up, the advantages of the present invention are: it can effectively enable enterprises to calculate the transportation costs of different types of goods when conducting international transportation, accurately to the transportation cost of each product, and then multiply the transportation cost of each product by the transportation quantity to calculate the actual cost. At the same time, by retrieving the historical transportation costs of all different types of goods, it can achieve effective reuse of massive historical data.

[0092] In further design, time cost is also part of transportation cost:

[0093] The cost of cargo transportation also includes the time cost calculation of cargo, as follows:

[0094] S101, obtain the time cost of all different types of goods in the enterprise history;

[0095] S102, retrieve all different types of goods in history and make a one-to-one correspondence with the time costs of all different types of goods in history;

[0096] S103, based on all historical different types of goods and corresponding time costs, a time model is established, the input of the time model is the type of goods, and the output of the time model is the time cost of each goods transportation;

[0097] S104, input the cargo type into the time model and obtain the time result.

[0098] In this solution, by retrieving the time cost of transporting different types of goods, the approximate time required for transporting different types of goods can be calculated.

[0099] In the further design, based on all different types of goods in history and the corresponding time costs, the calculation formula for establishing the time model is as follows:

[0100]

[0101] Where: f t represents the time cost of goods, V k represents the value of the goods, p represents the cost coefficient of the goods, T represents the transportation time of the goods (hours), and the method for determining the cost coefficient of the goods holding is the Aford-bangs model. The Alford-bangs model considers the factors affecting the time value of the goods from multiple aspects. The Alford-bangs model can be expressed as p = p1 + p2 + p3 ... + p n Where p represents the coefficient of goods holding cost in a certain year (%); p1 to p n Indicates influencing factors.

[0102] Specific details: p1+p2+p3......+p n They are respectively represented as insurance, warehousing, taxes, transportation, removal, depreciation, interest, obsolescence, etc.

[0103] In summary, in this solution, a time model is established based on all different types of goods and the corresponding time costs in history. Therefore, when subsequent companies inquire, it is possible to predict the approximate time cost of transporting different types of goods. According to the time cost, the delivery time of different types of goods can be reasonably planned, so that the receiving companies can receive the goods in time and increase customer satisfaction.

[0104] In further design, the cost of cargo transportation also includes the cost calculation of equipment and tools for moving cargo, as follows:

[0105] Calculate the average annual cost of the equipment and tools during their life (forklifts are used to transport the tools onto carriers, so forklifts are also a cost)

[0106] Assume A is the original value of the equipment and tools, B is the estimated net residual value of the tools and tools (net residual value refers to the value of the fixed assets after the expiration of their useful life minus the fixed assets liquidation fee to be paid), λ is the deterioration added value of the tools and tools, T is the useful life of the equipment and tools, and K is the average annual expenditure during the useful life of the equipment and tools.

[0107] The calculation formula is as follows:

[0108]

[0109] In this solution, the usage costs of different types of equipment and tools can be calculated, which can be added to the cost of goods, allowing enterprises to fully calculate the various costs used in the entire transportation process.

[0110] In further design, the cost of cargo transportation also includes the calculation of fuel consumption when the transport vehicle returns empty;

[0111] When the vehicle is a truck;

[0112] The formula for calculating the fuel consumption of a truck on its return trip when empty is:

[0113]

[0114] Where: D is the fuel consumption of the truck when it is empty, I is the fuel consumption of the truck when it is fully loaded, h is the wind resistance, which is taken as 0.7; V, V1, and V2 are the vehicle's own weight, cargo weight, and fully loaded cargo weight, respectively.

[0115] In summary, in this solution, by calculating the fuel consumption of the empty truck for the return trip calculation, the empty fuel consumption and the trip can be calculated into the cost of cargo transportation, so that the enterprise can understand the monetary cost of cargo transportation and optimize management. It should be noted that when cargo transportation is shipping, sea transportation or rail transportation, the above fuel consumption calculation is not applicable.

[0116] A supply chain traceability system based on a large model, including:

[0117] The traceability module traces the goods in the supply chain and stores all the goods supply chain data;

[0118] The acquisition module is used to obtain the historical freight transportation costs of all different types of goods of the enterprise, and classify and extract all different types of historical goods to obtain data of several different types of goods (including the input goods types such as wine, furniture, etc.);

[0119] The retrieval module retrieves all different types of cargo data, and then matches them one by one with the historical cargo transportation costs of all different types of cargo;

[0120] The construction module builds a prediction model based on all different types of cargo data and the cargo transportation costs of all different types of cargo. The input of the prediction model is the cargo type, and the output of the prediction model is the cost result of each cargo. Based on the statistical analysis of the cargo transportation costs of all different types of cargo in the enterprise's history, the cost result of each cargo is obtained. When the enterprise calculates the transportation costs of different types of cargo, based on the prediction model and the cargo type, the transportation cost of the cargo can be effectively predicted according to the quantity of cargo transportation. Compared with manual calculation, the speed is fast and the error is small.

[0121] The output module inputs the cargo type and cargo quantity into the prediction model, obtains the cost results, and recommends the final transportation cost based on the cost results.

[0122] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A supply chain traceability system based on a large model, characterized in that: The specific steps include: The traceability module traces the goods in the supply chain and stores all the goods supply chain data; The acquisition module is used to obtain the cargo transportation costs of all different types of cargoes in the history of the enterprise, and classify and extract all different types of cargoes in the history to obtain data of several different types of cargoes; The retrieval module retrieves all different types of cargo data, and then matches them one by one with the historical cargo transportation costs of all different types of cargo; A construction module is used to establish a prediction model based on all different types of cargo data and cargo transportation costs of all different types of cargo. The input of the prediction model is the cargo type, and the output of the prediction model is the cost result of each cargo. The output module inputs the cargo type and cargo quantity into the prediction model, obtains the cost results, and recommends the final transportation cost based on the cost results.

2. A supply chain traceability system based on a large model according to claim 1, characterized in that: The establishment of a prediction model based on all different types of cargo data and cargo transportation costs of all different types of cargo specifically includes: Establish data on all different types of goods, and then match them one by one with the historical freight transportation costs of all different types of goods; Based on the one-to-one correspondence between all different types of cargo data and the cargo transportation costs of all different types of cargo, statistical analysis is performed on the cargo transportation costs corresponding to each type of cargo, and the relevant weights between each type of cargo and the corresponding cargo transportation costs are obtained respectively; Establish a sub-model, which is used to calculate the matching degree between the input cargo type and all different types of cargo in history; A screening model is established. The screening model screens out one or more of all different types of historical cargo data that best matches the input cargo type based on the calculation results of the sub-model, and outputs them as cost results.

3. A supply chain traceability system based on a large model according to claim 2, characterized in that: The calculation formula for the relevant weight between each type of goods and the cost of corresponding goods transportation is: In the formula, p n is the correlation weight between the nth type of goods and the corresponding transportation cost, x 总 is the total freight transportation cost of all different types of goods in history corresponding to the goods type, x n It is the total number of the nth historical freight transportation costs of all different types of freight corresponding to the freight type.

4. A supply chain traceability system based on a large model according to claim 3, characterized in that: The specific calculation formula of the sub-model is: Among them, Q j is the matching degree between the input j-th type of goods and the corresponding transportation cost, X j is the total number of historical cargo transportation cost data for the jth cargo type, X0 is the total number of cargo types input, and X 同 is the total cost of historical cargo transportation corresponding to the input cargo type data and the jth cargo type, p' l is the correlation weight between the j-th type of goods and the same features in the l-th input type of goods and the cost of the j-th historical goods transportation, and max() is the maximum value function.

5. A supply chain traceability system based on a large model according to claim 4, characterized in that: The inputting of the cargo type and cargo quantity into the prediction model to obtain the cost result specifically includes: Calculate the quantity of the input goods type and quantity; Call the sub-model to calculate the matching degree between the input goods and the corresponding goods transportation cost based on several goods types; Calling the screening model to screen out one or more historical cargo types whose matching degree with the input cargo type is greater than a preset value as the predicted historical cargo types; The historical cost of the type of goods will be retrieved and then multiplied by the quantity of goods to obtain the transportation cost.

6. A supply chain traceability system based on a large model according to claim 5, characterized in that: The cost of cargo transportation also includes the time cost calculation of cargo, which is as follows: Obtain the time cost of all different types of goods in the company's history; Retrieve all different types of goods in history and make a one-to-one correspondence with the time costs of all different types of goods in history; Based on all different types of goods and their corresponding time costs in history, a time model is established. The input of the time model is the type of goods, and the output of the time model is the time cost of each goods transportation; Input the cargo type into the time model and obtain the time result.

7. A supply chain traceability system based on a large model according to claim 6, characterized in that: The calculation formula for establishing the time model based on all different types of goods and the corresponding time costs in history is as follows: Where: f t represents the time cost of goods, V k represents the value of the goods, p represents the cost coefficient of the goods, T represents the transportation time of the goods (hours), and the method for determining the cost coefficient of the goods holding is the Aford-bangs model. The Alford-bangs model considers the factors affecting the time value of the goods from multiple aspects. The Alford-bangs model can be expressed as p = p1 + p2 + p3 ... + p n Where p represents the coefficient of goods holding cost in a certain year (%); p1 to p n Indicates influencing factors.

8. A supply chain traceability system based on a large model according to claim 7, characterized in that: The cost of cargo transportation also includes the cost of equipment and tools for handling cargo, as follows: Calculate the average annual expenditure during the life of the equipment and tools; Assume A is the original value of the equipment and tools, B is the estimated net residual value of the tools and tools, λ is the added value of the tools and tools due to deterioration, T is the service life of the equipment and tools, and K is the average annual expenditure during the service life of the equipment and tools. The calculation formula is as follows:

9. A supply chain traceability system based on a large model according to claim 1, characterized in that: The cost of cargo transportation also includes the calculation of fuel consumption when the transport vehicle returns empty; When the vehicle is a truck; The fuel consumption calculation formula for a truck returning empty is: Where: D is the fuel consumption of the truck when it is empty, I is the fuel consumption of the truck when it is fully loaded, h is the wind resistance, which is taken as 0.7; V, V1, and V2 are the vehicle's own weight, cargo weight, and fully loaded cargo weight, respectively.

10. A supply chain traceability method based on a large model, characterized in that: A supply chain traceability system based on a large model is used to implement the system as described in claims 1-9, comprising: S1. Obtain the historical freight transportation costs of all different types of freight in the supply chain, and classify and extract all different types of freight in history to obtain data of several different types of freight; S2. Retrieve all different types of cargo data, and then match them one by one with the cargo transportation costs of all different types of cargo in the historical supply chain; S3. Based on all different types of cargo data and cargo transportation costs of all different types of cargo, a prediction model is established, wherein the input of the prediction model is the cargo type, and the output of the prediction model is the cost result of each cargo; S4. Input the type and quantity of goods into the forecasting model to obtain the cost result; S5. Based on the cost results, recommend the final transportation cost.