Logistics supply chain collaborative optimization method and system based on AI large model

Abstract

The present disclosure provides an AI large model-based logistics supply chain collaborative optimization method and system, relating to the technical field of data processing. The method comprises: collecting logistics data of each circulation dimension in each logistics link of the logistics supply chain of a target product; determining the corresponding circulation dimension mutation degree based on the logistics data under each circulation dimension, and determining the circulation dimension mutation degree of the target circulation dimension as the user demand mutation degree of the target product; determining the initial link impact degree of each logistics link based on the circulation dimension mutation degree and the user demand mutation degree; correcting the initial link impact degree based on the position information of each logistics link on the logistics supply chain to obtain the target link impact degree; identifying a high-pressure link and correcting the logistics data of its associated links based on the target link impact degree thereof; inputting the corrected logistics data into a logistics large model to realize optimization of the logistics supply chain. The present disclosure realizes collaborative optimization of each link of the logistics supply chain.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a logistics supply chain collaborative optimization method and system based on an AI large model. BACKGROUND

[0002] The logistics supply chain refers to a full-chain network structure system starting from raw material procurement, passing through production and manufacturing, warehousing and transportation, distribution and retail, and ending with product delivery to consumers. Through the coordination and integration of logistics, information flow and capital flow, seamless connection of each link of the supply chain is realized to meet customer demand and improve overall efficiency and competitiveness.

[0003] In related technologies, when a new product is put into the market through the above logistics supply chain, the related enterprises on the logistics supply chain do not understand the real supply and demand relationship of the new product. Therefore, an AI (Artificial Intelligence) technology is usually used to build a large model, and the supply and demand relationship of the new product is predicted through the large model, and the logistics supply chain is optimized accordingly.

[0004] However, due to the lack of prior market experience about the new product, when the supply and demand relationship of the new product is predicted by the above method, the matching degree between the logistics supply chain and the new product is insufficient, which leads to the fact that when the upstream and downstream supply and demand relationship of the new product logistics supply chain changes, the logistics between each link of the logistics supply chain cannot respond quickly, and the sudden change of the new product cannot be coped with, and thus the entire logistics supply chain cannot be optimized according to the real-time demand. SUMMARY

[0005] Therefore, the present application provides a logistics supply chain collaborative optimization method and system based on an AI large model to solve the problem that in related technologies, due to the lack of prior market experience about the new product, when the supply and demand relationship of the new product is predicted, the matching degree between the logistics supply chain and the new product is insufficient, which leads to the fact that when the upstream and downstream supply and demand relationship of the new product logistics supply chain changes, the logistics between each link of the logistics supply chain cannot respond quickly, and the sudden change of the new product cannot be coped with, and thus the entire logistics supply chain cannot be optimized according to the real-time demand.

[0006] According to a first aspect of the present application, a logistics supply chain collaborative optimization method based on an AI large model is provided, and the technical solution is as follows:

[0007] Collecting logistics data of each flow dimension in each logistics link of the logistics supply chain of the target product;

[0008] determine a target flow dimension in the flow dimensions, and determine the flow dimension mutation degree of the target flow dimension as a user demand mutation degree of the target product;

[0009] determine an initial link impact degree of each logistics link based on the flow dimension mutation degree and the user demand mutation degree;

[0010] correct the initial link impact degree based on position information of each logistics link on the logistics supply chain to obtain a target link impact degree of each logistics link;

[0011] identify a high-pressure link in the logistics links, and correct logistics data of associated links of the high-pressure link based on the target link impact degree of the high-pressure link;

[0012] input the corrected logistics data into a logistics large model to optimize the logistics supply chain through the logistics large model.

[0013] For example, the determination of the flow dimension mutation degree corresponding to each flow dimension based on the logistics data under each flow dimension, the determination of the target flow dimension in the flow dimensions, and the determination of the flow dimension mutation degree of the target flow dimension as the user demand mutation degree of the target product include: obtaining a change amount of the logistics data in a current period and a previous period of each flow dimension, calculating a change amount difference between the change amount of the logistics data in the current period and the change amount of the logistics data in the previous period, denoted as the flow dimension mutation degree corresponding to the flow dimension; determining a product inventory amount of the target product in the target flow dimension as the target product inventory amount, and determining the flow dimension mutation degree corresponding to the product inventory amount as the user demand mutation degree.

[0014] For example, the determination of the initial link impact degree of each logistics link based on the flow dimension mutation degree and the user demand mutation degree includes: for each logistics link, determining a link supply flow rate of the logistics link based on the logistics data under each flow dimension corresponding to the logistics link; determining a user mutation correlation strength of each flow dimension corresponding to the logistics link based on the flow dimension mutation degree of each flow dimension corresponding to the logistics link, the link supply flow rate, and the user demand mutation degree; and determining the initial link impact degree of each logistics link based on the user mutation correlation strength of each flow dimension corresponding to each logistics link and the flow dimension mutation degree; wherein the initial link impact degree is used to describe the impact degree of the current logistics link on the next logistics link under the user demand mutation degree.

[0015] Illustratively, the user mutation correlation strength of each of the circulation dimensions corresponding to the logistics link is determined based on the circulation dimension mutation degree, the link supply circulation rate and the user demand mutation degree of each of the circulation dimensions corresponding to the logistics link, comprising: calculating the sum of the circulation dimension mutation degrees of each of the circulation dimensions corresponding to the logistics link; for each of the circulation dimensions, calculating the ratio of the circulation dimension mutation degree of the circulation dimension to the sum of the circulation dimension mutation degrees, and calculating the product of the ratio and the link supply circulation rate and the user demand mutation degree to obtain the user mutation correlation strength of the circulation dimension.

[0016] Illustratively, the initial link impact degree of each of the logistics links is determined based on the user mutation correlation strength and the circulation dimension mutation degree of each of the circulation dimensions corresponding to each of the logistics links, comprising: for each of the logistics links, a sequence formed by the user mutation correlation strength of each of the circulation dimensions corresponding to the logistics link is recorded as a first sequence; for the next logistics link of the logistics link, a sequence formed by the circulation dimension mutation degree of each of the circulation dimensions corresponding to the next logistics link is recorded as a second sequence; matching the user mutation correlation strength in the first sequence and the circulation dimension mutation degree in the second sequence, and determining the impact correlation influence strength between the user mutation correlation strength and the circulation dimension mutation degree corresponding thereto; for each of the circulation dimensions of the logistics link, based on the user mutation correlation strength of the circulation dimension and the impact correlation influence strength corresponding thereto, the dimension impact strength of the circulation dimension is determined; the average value of the dimension impact strength of each of the circulation dimensions in the logistics link is calculated and recorded as the initial link impact degree of the logistics link.

[0017] Illustratively, the target link impact degree of each of the logistics links is obtained by correcting the initial link impact degree based on the position information of each of the logistics links on the logistics supply chain, comprising: determining a first core link in the order direction of the logistics supply chain and a second core link in the reverse order direction; for each of the logistics links, determining the first number of links between the logistics link and the first core link, and the second number of links between the logistics link and the second core link; determining a first distance influence degree based on the first number of links and the total number of logistics links in the logistics supply chain; determining a second distance influence degree based on the second number of links and the total number of logistics links in the logistics supply chain; determining the target link impact degree based on the first distance influence degree, the second distance influence degree and the initial link impact degree.

[0018] Illustratively, the identification of the high-pressure link in the logistics link, the correction of the logistics data of the associated link based on the target link impact degree of the high-pressure link comprises: when the user demand mutation degree is greater than a first preset threshold, triggering the logistics optimization alarm; when the logistics optimization alarm is triggered, identifying the logistics link with a target link impact degree greater than a second preset threshold as the high-pressure link; calculating the difference between the target link impact degree of the high-pressure link and the second preset threshold, adjusting the logistics data of the associated link until the difference between the target link impact degree of the high-pressure link and the second preset threshold meets a preset condition.

[0019] Illustratively, the adjustment of the logistics data of the associated link until the difference between the target link impact degree of the high-pressure link and the second preset threshold meets a preset condition comprises: when the high-pressure link is a non-continuous distribution, iteratively adjusting the logistics data of the previous logistics link of the high-pressure link until the difference between the target link impact degree of the high-pressure link and the second preset threshold meets the preset condition; when the high-pressure link is a continuous distribution, iteratively adjusting the logistics data of the multiple logistics links before the continuous distribution in turn until the difference between the target link impact degree of the high-pressure link in the continuous distribution and the second preset threshold meets the preset condition.

[0020] According to a second aspect of the present disclosure, a logistics supply chain collaborative optimization system based on an AI large model is provided, comprising:

[0021] A data acquisition module is configured to acquire logistics data of each flow dimension in each logistics link of a logistics supply chain of a target product;

[0022] A data processing module is configured to determine a flow dimension mutation degree of a corresponding flow dimension based on the logistics data under each flow dimension, determine a target flow dimension in the flow dimension, and determine the flow dimension mutation degree of the target flow dimension as a user demand mutation degree of the target product;

[0023] The data processing module is further configured to determine an initial link impact degree of each logistics link based on the flow dimension mutation degree and the user demand mutation degree;

[0024] The data processing module is further configured to correct the initial link impact degree based on position information of each logistics link on the logistics supply chain to obtain a target link impact degree of each logistics link;

[0025] A logistics optimization module is configured to identify a high-pressure link in the logistics link, and correct the logistics data of an associated link based on the target link impact degree of the high-pressure link.

[0026] The logistics optimization module is further configured to input the corrected logistics data into a logistics large model to optimize the logistics supply chain by the logistics large model.

[0027] Exemplarily, the AI large model-based logistics supply chain collaborative optimization system implements any of the above AI large model-based logistics supply chain collaborative optimization methods in a manner based on an agent workflow and a large language model; the agent workflow includes a planner, an executor, a memory, and a reflector; wherein: the planner is configured to generate an execution plan for determining the target link impact degree of each logistics link on the logistics supply chain; the executor is configured to determine the target link impact degree of each logistics link by executing the execution plan; and correct the logistics data of the associated link based on the target link impact degree of the high-pressure link; the memory is configured to store the target link impact degree of each logistics link, each intermediate data generated when calculating the target link impact degree, and the adjustment record of the logistics data; the reflector is configured to identify the high-pressure link in the logistics link; and generate an adjustment strategy for the logistics data of the associated link of the high-pressure link.

[0028] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory storing a program, wherein the program comprises instructions which, when executed by the processor, cause the processor to perform the above AI large model-based logistics supply chain collaborative optimization method.

[0029] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are configured to cause the computer to perform the above AI large model-based logistics supply chain collaborative optimization method.

[0030] The present application can have the following partial or all beneficial effects:

[0031] In the AI large model-based logistics supply chain collaborative optimization method provided by the present application, the logistics data of the target product at each circulation dimension on the logistics supply chain thereof is collected, and the circulation dimension mutation degree and the user demand mutation degree are determined based on the logistics data, so that the initial link impact degree of each logistics link can be determined according to the circulation dimension mutation degree and the user demand mutation degree, and the initial link impact degree is corrected based on the position of each logistics link on the logistics supply chain. The target link impact degree obtained by correction can capture the implicit correlation between each logistics link, so that when the user demand mutates, the impact conduction path can be identified in time, and the efficiency of logistics supply chain collaborative optimization is improved. Further, after obtaining the target link impact degree, the present application can also identify the high-pressure link based on the target link impact degree, and relieve the pressure of the high-pressure link caused by the mutation of the user demand by adjusting the logistics data of the associated link of the high-pressure link. In addition, by inputting the corrected logistics data into the pre-trained logistics large model, the logistics large model can also be used to quickly adapt the logistics supply chain of new products of different categories.

[0032] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0034] Figure 1 A flow chart of the AI large model-based logistics supply chain collaborative optimization method according to an exemplary embodiment of the present disclosure is shown;

[0035] Figure 2 A schematic block diagram of the AI large model-based logistics supply chain collaborative optimization system according to an exemplary embodiment of the present disclosure is shown;

[0036] Figure 3 A structural block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0037] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined inventive purpose, the specific implementation, structure, features and effects of the AI large model-based logistics supply chain collaborative optimization method and system according to the present application are described in detail below in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0039] The specific scheme of the AI large model-based logistics supply chain collaborative optimization method and system provided by the present application is specifically described below in combination with the drawings.

[0040] Please refer to Figure 1 , which shows the method flowchart of the AI large model-based logistics supply chain collaborative optimization method provided by one embodiment of the present application, as Figure 1 shown, the AI large model-based logistics supply chain collaborative optimization method specifically includes the following steps:

[0041] S110: Collecting logistics data of each circulation dimension in each logistics link on the logistics supply chain of the target product;

[0042] S120: Determining the circulation dimension mutation degree of the corresponding circulation dimension based on the logistics data under each circulation dimension, determining the target circulation dimension in the circulation dimension, and determining the circulation dimension mutation degree of the target circulation dimension as the user demand mutation degree of the target product;

[0043] S130: Determining the initial link impact degree of each logistics link based on the circulation dimension mutation degree and the user demand mutation degree;

[0044] S140: Correcting the initial link impact degree based on the position information of each logistics link on the logistics supply chain to obtain the target link impact degree of each logistics link;

[0045] S150: Identifying the high-pressure link in the logistics link, and correcting the logistics data of the associated link of the high-pressure link based on the target link impact degree of the high-pressure link;

[0046] S160: Inputting the corrected logistics data into the logistics large model to optimize the logistics supply chain through the logistics large model.

[0047] The present application can determine the initial link impact degree of each logistics link according to the circulation dimension mutation degree and the user demand mutation degree, and correct the initial link impact degree based on the position of each logistics link in the logistics supply chain. The target link impact degree obtained by correction can capture the implicit correlation between each logistics link, so as to timely identify the impact conduction path when the user demand mutates, and improve the efficiency of logistics supply chain collaborative optimization. Further, after obtaining the above target link impact degree, the present application can also identify high-pressure links based on the target link impact degree, and relieve the pressure of high-pressure links caused by user demand mutation by adjusting the logistics data of the associated links of the high-pressure links. In addition, the present application can also quickly adapt the logistics supply chain of different categories of new products by inputting the corrected logistics data into the pre-trained logistics large model.

[0048] Next, each step of the above logistics supply chain collaborative optimization method based on AI large model will be described in detail:

[0049] In step S110, the logistics data of each circulation dimension in each logistics link of the target product in the logistics supply chain is collected.

[0050] In the embodiment of the present application, the above logistics supply chain refers to a full-chain network structure system starting from raw material procurement, passing through production and manufacturing, warehousing and transportation, distribution and retail, and ending with delivering products to consumers.

[0051] In the embodiment of the present application, the above target product is any product put into the user through the above logistics supply chain. For example, the target product can be a new product put into the user through the above logistics supply chain, and the above logistics supply chain collaborative optimization method based on AI large model is used to optimize each link of the logistics supply chain according to the supply and demand relationship between the user and the new product.

[0052] In the embodiment of the present application, the above logistics link is a logistics node experienced in the process of putting the target product into the above logistics supply chain. Specifically, in an actual logistics scenario, the above logistics supply chain can include logistics links such as procurement link, production link, delivery link and sales link.

[0053] In the embodiments of the present application, the flow dimension refers to a perspective for analyzing the flow of goods, services, funds, information and other elements of the target product in the logistics supply chain. For example, the procurement link can include the flow dimensions of raw material inventory and procurement lead time; the production link can include the flow dimensions of product inventory and product yield; the delivery link can include the flow dimensions of truck assembly and truck transportation time; and the sales link can include the flow dimensions of product inventory and customer satisfaction rate.

[0054] In the embodiments of the present application, the logistics data of each flow dimension in each logistics link of the logistics supply chain of the target product can be periodically collected. For example, taking the logistics supply chain of the target product as an example, the logistics data of each flow dimension in each logistics link of the logistics supply chain of the target product can be collected as follows: for each logistics link in the logistics supply chain, the corresponding single-day starting dimension logistics data is recorded once before daily operation, and the corresponding single-day end dimension logistics data is recorded once after operation. Specifically, taking the product inventory of the sales link as an example, the starting product inventory is recorded before the start of sales each day, and the end product inventory is recorded after the end of sales.

[0055] Further, in order to facilitate subsequent processing, the embodiments of the present application can also perform standardization processing on the collected data after collecting the logistics data of each flow dimension, which is implemented as follows: the collected logistics data of each flow dimension is dimensionally unified, and its value is normalized to the range of [0, 1]. Specifically, the range can be normalized based on the maximum and minimum values in each dimension using range standardization, which is a technical means familiar to those skilled in the art, and other normalization methods can also be selected, which are not limited and elaborated here.

[0056] In step S120, the flow dimension mutation degree of the corresponding flow dimension is determined based on the logistics data in each flow dimension, the target flow dimension in the flow dimension is determined, and the flow dimension mutation degree of the target flow dimension is determined as the user demand mutation degree of the target product.

[0057] In the embodiments of the present application, the user demand mutation degree is a quantitative index for measuring the degree of change in the demand of the target product by the user. Specifically, since the sales link can more directly represent the degree of demand for the target product by the user than other logistics links, when the demand relationship between the user and the target product changes, the sales link will directly be affected by the demand relationship change and produce obvious logistics data fluctuations, so the above-mentioned target circulation dimension can take the product inventory of the target product in the sales link as the circulation dimension. For example, the user demand mutation degree can be determined by the change in the product inventory of the target product in the sales link in the current period and the adjacent previous period, and the greater the value, the more obvious the change in the demand of the target product by the user.

[0058] In the embodiments of the present application, the circulation dimension mutation degree is a quantitative index for measuring the degree of change in the logistics data of each circulation dimension in the logistics supply chain, reflecting the potential impact ability of the change in the logistics data of the corresponding circulation dimension on the next logistics link. For example, the circulation dimension mutation degree can be determined by the change in the logistics data of the corresponding circulation dimension in the current period and the adjacent previous period, and the greater the value, the more significant the change in the information of the circulation dimension in the current period, and the stronger the impact effect on the next logistics link.

[0059] For example, the above-mentioned determination of the circulation dimension mutation degree of the corresponding circulation dimension based on the logistics data in each circulation dimension, the determination of the target circulation dimension in the circulation dimension, and the determination of the circulation dimension mutation degree of the target circulation dimension as the user demand mutation degree of the target product can be implemented as follows: obtaining the change in the logistics data of each circulation dimension in the current period and the previous period, calculating the change value difference between the change in the logistics data of the current period and the change in the logistics data of the previous period, denoted as the circulation dimension mutation degree of the corresponding circulation dimension; determining the product inventory of the target product in the circulation dimension as the target circulation dimension, and determining the circulation dimension mutation degree corresponding to the product inventory as the user demand mutation degree.

[0060] Specifically, taking each day as a period, the circulation dimension mutation degree of the circulation dimension i can be calculated by the following formula:

[0061]

[0062] Wherein, is the circulation dimension mutation degree of the circulation dimension i; is the change in the logistics data of the circulation dimension in the day, that is, the difference between the single-day start dimension logistics data and the single-day end dimension logistics data corresponding to the circulation dimension i in the day; a change amount of the logistics data of the circulation dimension in the day before the day, that is, a difference between the single-day start dimension logistics data and the single-day end dimension logistics data corresponding to the circulation dimension i of the day before; is a normalization function.

[0063] Correspondingly, the above user demand mutation degree can be calculated by the following formula:

[0064]

[0065] wherein, is a user demand mutation degree of the target product by the user; is a change amount of the product inventory of the target product in the sales link in the day; the above is a change amount of the product inventory of the target product in the sales link in the day before the day; is a normalization function.

[0066] In step S130, the initial link impact degree of each logistics link is determined based on the circulation dimension mutation degree and the user demand mutation degree.

[0067] In the embodiment of the present application, the above initial link impact degree refers to the degree of impact pressure caused by each logistics link in the logistics supply chain when responding to the change of user demand on the target product on the logistics supply state of the next logistics link when the user demand on the target product changes. The greater the value of the initial link impact degree, the stronger the impact of the corresponding logistics link on the next logistics link.

[0068] Exemplarily, the above determination of the initial link impact degree of each logistics link based on the circulation dimension mutation degree and the user demand mutation degree can be implemented as follows: for each logistics link, the link supply circulation rate of the logistics link is determined based on the logistics data under each circulation dimension corresponding to the logistics link; the user mutation correlation strength of each circulation dimension corresponding to the logistics link is determined based on the circulation dimension mutation degree, the link supply circulation rate and the user demand mutation degree of each circulation dimension corresponding to the logistics link; the initial link impact degree of each logistics link is determined based on the user mutation correlation strength and the circulation dimension mutation degree of each circulation dimension corresponding to each logistics link; wherein the initial link impact degree is used to describe the impact degree of the current logistics link on the next logistics link under the user demand mutation degree.

[0069] In the embodiment of the present application, the above link supply circulation rate is a quantitative index for measuring the efficiency of goods circulation of each logistics link in the logistics supply chain in the current period. The link supply circulation rate can be determined by comprehensively considering the change amount of the logistics data under multiple circulation dimensions corresponding to the logistics link, and reflects the goods circulation characteristics of the corresponding logistics link itself.

[0070] Specifically, taking any logistics link in the logistics supply chain as an example, the link supply circulation rate in the current cycle can be calculated using the following formula:

[0071]

[0072] in, The supply and circulation rate of the aforementioned logistics links within the current cycle; This indicates the number of circulation dimensions included in the above logistics process; For the above logistics links The data for each circulation dimension at the beginning of the current cycle; For the above logistics links The data for each circulation dimension at the end of the current cycle; the supply circulation rate of the above-mentioned links. The larger the value, the better the cargo flow in the above logistics links during the current cycle.

[0073] In the embodiments of this application, the aforementioned user mutation correlation strength refers to the degree of correlation between the change of the corresponding circulation dimension in any logistics link in the logistics supply chain and the change of user demand when the user demand changes. It is used to measure the impact strength of the circulation dimension on the next logistics link when responding to changes in user demand.

[0074] For example, the determination of the user mutation correlation strength of each circulation dimension corresponding to the logistics link based on the circulation dimension mutation degree, link supply circulation rate and user demand mutation degree of each circulation dimension corresponding to the logistics link can be achieved as follows: calculate the sum of the circulation dimension mutation degrees of each circulation dimension corresponding to the logistics link; for each circulation dimension, calculate the ratio of the circulation dimension mutation degree of the circulation dimension to the sum of the circulation dimension mutation degrees, and calculate its product with the link supply circulation rate and user demand mutation degree to obtain the user mutation correlation strength of the circulation dimension.

[0075] Specifically, taking any logistics link in the aforementioned logistics supply chain as an example, the user mutation correlation strength corresponding to any circulation dimension in that logistics link can be calculated using the following formula:

[0076]

[0077] in, The correlation strength of user mutations in the circulation dimension of the above logistics links within the current period; This represents the degree of mutation in the circulation dimension corresponding to that circulation dimension. This indicates the number of circulation dimensions included in the above logistics process; This is the sum of the mutation degrees of all circulation dimensions included in the current logistics process; The ratio of the flow dimension mutation degree of the flow dimension to the total of all dimension mutation degrees of the logistics link to which the flow dimension belongs, reflects the relative change intensity of the flow dimension in the logistics link to which the flow dimension belongs. The link supply flow rate of the logistics link; The link flow mutation degree of the logistics link in the current period, the greater the value, the more obvious the change of the characteristics of the logistics link itself. The user demand mutation degree of the user demand change of the target product; in addition, the greater the value of The greater the value, the more obvious the change of the characteristics of the logistics link unique to the flow dimension when the flow dimension represents the user demand change, reflecting the greater the impact intensity of the flow dimension on the next logistics link.

[0078] In the embodiments of the present application, after determining the link supply flow rate and the user mutation correlation intensity through the above process, the user mutation correlation intensity of each flow dimension corresponding to the logistics link is determined based on the flow dimension mutation degree of each flow dimension corresponding to the logistics link, the link supply flow rate and the user demand mutation degree; the initial link impact degree of each logistics link can be determined based on the user mutation correlation intensity of each flow dimension corresponding to each logistics link and the flow dimension mutation degree of each flow dimension corresponding to each logistics link as follows: for each logistics link, a sequence formed by the user mutation correlation intensity of each flow dimension corresponding to the logistics link is recorded as a first sequence; for the next logistics link of the logistics link, a sequence formed by the flow dimension mutation degree of each flow dimension corresponding to the next logistics link is recorded as a second sequence; the user mutation correlation intensity and the flow dimension mutation degree in the second sequence are matched, and the impact correlation influence intensity between the user mutation correlation intensity and the flow dimension mutation degree corresponding to the user mutation correlation intensity is determined; for each flow dimension of the logistics link, the dimension impact intensity of the flow dimension is determined based on the user mutation correlation intensity of the flow dimension and the impact correlation influence intensity corresponding to the user mutation correlation intensity; the average value of the dimension impact intensity of each flow dimension in the logistics link is calculated, which is recorded as the initial link impact degree of the logistics link.

[0079] Specifically, according to the influence order of the user demand relationship of the target product on the logistics supply chain: the sales link-the delivery link-the production link-the procurement link, when the user demand relationship of the target product changes, the previous logistics link gradually penetrates into the next logistics link; taking the above sales link as the current logistics link and the above delivery link as the next logistics link as an example, the initial link impact degree of the sales link can be determined according to the above method as follows, which specifically includes the following steps:

[0080] S1: assuming that the sales link includes two flow dimensions, the user mutation correlation intensity corresponding to the two flow dimensions is D1 and D2 respectively, and the first sequence is L1={D1, D2}.

[0081] S2: Assuming that the above delivery link includes three flow dimensions, and the flow dimension mutation degrees corresponding to the three flow dimensions are C1, C2 and C3 respectively, the above second sequence is L2={C1, C2, C3}.

[0082] S3: The matching of the user mutation association strength in the above first sequence and the flow dimension mutation degree in the second sequence can be realized based on a dynamic time warping matching algorithm (Dynamic Time Warping Matching, DTW).

[0083] Specifically, the first sequence of the sales link and the second sequence of the delivery link can be dynamically matched by the above-mentioned DTW matching to determine the optimal corresponding relationship between elements in the first sequence and the second sequence. For example, for the time sequence of the logistics data of a certain flow dimension in the first sequence and the time sequence of the logistics data corresponding to the flow dimension in the second sequence, the two elements with the closest distance in the two time sequences can be taken as the optimal corresponding relationship. Thus, the impact transmission effect between the above-mentioned flow dimension in the first sequence and the flow dimension matched in the second sequence can be quantified based on the minimum distance value, i.e. the impact association influence strength of the flow dimension in the above-mentioned first sequence. Specifically, the impact association influence strength can take the inverse proportional value of the minimum distance.

[0084] It should be noted that the matching relationship between the flow dimension corresponding to the first sequence and the flow dimension corresponding to the second sequence can be determined based on the actual scene, and a certain flow dimension corresponding to the first sequence may have multiple matching flow dimensions in the second sequence. For example, in the actual logistics scene, the product inventory quantity of the above-mentioned sales link may have an impact on the truck assembly quantity and truck transportation time of the delivery link, so when determining the matching relationship, a one-to-many correspondence relationship of {product inventory quantity, truck assembly quantity}, {product inventory quantity, truck transportation time} can be determined.

[0085] S4: In the current period, taking any one flow dimension in the above-mentioned sales link as an example, after determining the impact association influence strength between the flow dimension and the flow dimension in the delivery link that has a matching relationship, the dimension impact strength of the flow dimension can be calculated by the following formula :

[0086]

[0087] wherein, is the comprehensive impact strength of any one flow dimension in the current logistics link on the corresponding flow dimension of the next logistics link when the user's demand for the target product changes. the number of the flow dimensions that have a matching relationship with the flow dimension in the next logistics link; the impact correlation influence strength of the flow dimension to the first impact correlation dimension in the next logistics link; In the above formula, the greater the value of the dimension impact strength , the greater the degree of the flow dimension affecting the flow dimensions in the next logistics link when responding to the sudden change in user demand, and the greater the flow dimension as the main link to transmit the impact force of user demand in the logistics link to the corresponding dimension in the next logistics link when the user demand impacts the logistics link.

[0088] S5: Calculate the average value of the dimension impact strength corresponding to all flow dimensions in the current logistics link as the initial link impact degree of the current logistics link to the next logistics link . The greater the value of the initial link impact degree , the greater the impact pressure on the logistics supply state of the next logistics link when the current logistics link responds to the sudden change in user demand.

[0089] In step S140, the initial link impact degree is corrected based on the position information of each logistics link on the logistics supply chain to obtain the target link impact degree of each logistics link.

[0090] Since the change in user demand for the target product directly affects the sales link on the logistics supply chain, when performing collaborative optimization of the logistics supply chain, the embodiments of the present application usually determine the initial link impact degree of the previous logistics link to the next logistics link in the order of sales link-delivery link-production link-purchasing link through the method of steps S110 to S130.

[0091] However, in the actual logistics scenario, when the user demand for the target product changes, in addition to the process of gradually transmitting the impact on the logistics state from the previous logistics link to the next logistics link in the above order, the logistics supply chain also simultaneously transmits the impact on the logistics state from the previous logistics link to the next logistics link in the order of the logistics supply link (purchasing link-production link-delivery link-sales link), so in addition to obtaining the initial link impact degree on the sales link-delivery link-production link-purchasing link through the above process, the initial link impact degree can also be corrected based on the position information of each logistics link on the logistics supply chain to obtain the target link impact degree of each logistics link.

[0092] Exemplarily, the above-mentioned correcting the initial link impact degree based on the position information of each logistics link on the logistics supply chain to obtain the target link impact degree of each logistics link can be implemented as follows: determining a first core link in the order direction of the logistics supply chain and a second core link in the reverse order direction; for each logistics link, determining the first number of intervals between the logistics link and the first core link, and the second number of intervals between the logistics link and the second core link; determining the first distance influence degree based on the first number of intervals and the total number of logistics links in the logistics supply chain; determining the second distance influence degree based on the second number of intervals and the total number of logistics links in the logistics supply chain; determining the target link impact degree based on the first distance influence degree, the second distance influence degree, and the initial link impact degree.

[0093] In one specific embodiment of the present application, after the above-mentioned logistics supply chain determines the initial link impact degree of each logistics link on the next logistics link in the order of the sales link-delivery link-production link-purchasing link, the above-mentioned determination of the target link impact degree can be implemented as follows:

[0094] S1: The purchasing link plays a core role in the order direction of the logistics supply chain (purchasing link-production link-delivery link-sales link), and the sales link plays a core role in the reverse order direction of the logistics supply chain (sales link-delivery link-production link-purchasing link), so in this step, the purchasing link in the order direction of the logistics supply chain is determined as the first core link, and the sales link in the reverse order direction of the logistics supply chain is determined as the second core link.

[0095] S2: Taking any one of the above-mentioned sales link-delivery link-production link-purchasing link as an example, the distance influence degree of the logistics link and the sales link / purchasing link is calculated ; wherein, represents the number of intervals between the logistics link and the sales / purchasing link; represents the total number of links in the logistics supply chain.

[0096] S3: After determining the first distance influence degree of the above-mentioned logistics link and the sales link , and the second distance influence degree of the above-mentioned logistics link and the purchasing link , the target link impact degree of the logistics link can be calculated by the following formula:

[0097]

[0098] wherein, is the target link impact degree of the logistics link on the next logistics link; is the first distance influence degree of the logistics link and the sales link (i.e. the first core link). represents the second distance influence degree of the logistics link to the purchasing link (i.e., the second core link); in addition, in the above formula, The greater the value of the greater, the weaker the influence of the logistics supply chain on the supply state of the logistics link in the logistics supply sequence (purchasing link-production link-delivery link-sales link), and the stronger the influence on the supply state of the logistics link in the impact effect influence sequence (sales link-delivery link-production link-purchasing link). is the initial link impact degree of the logistics link; is a normalization function.

[0099] In step S150, a high-pressure link in the logistics link is identified, and the logistics data of the associated link of the high-pressure link is corrected based on the target link impact degree of the high-pressure link.

[0100] In the embodiments of the present application, the high-pressure link refers to a logistics link in the logistics supply chain that needs to adjust the logistics state when the user demand for the target product changes due to the inability to withstand the mutation pressure. Illustratively, the logistics link with a target link impact degree exceeding a preset threshold can be identified as a high-pressure link.

[0101] In the embodiments of the present application, the associated link of the high-pressure link is a logistics link that has an impact on the logistics state of the high-pressure link, and the logistics pressure of the high-pressure link can be adjusted to a bearable range by adjusting the logistics state of the associated link.

[0102] Illustratively, the identification of the high-pressure link in the logistics link and the correction of the logistics data of the associated link of the high-pressure link based on the target link impact degree of the high-pressure link can be implemented as follows: when the user demand mutation degree is greater than a first preset threshold, triggering a logistics optimization alarm; when the logistics optimization alarm is triggered, identifying the logistics link with a target link impact degree greater than a second preset threshold as a high-pressure link; calculating the difference between the target link impact degree of the high-pressure link and the second preset threshold, adjusting the logistics data of the associated link until the difference between the target link impact degree of the high-pressure link and the second preset threshold meets a preset condition.

[0103] Wherein, according to the different distribution of high-pressure link, the above-mentioned adjustment of the logistics data of the associated link, until the difference between the target link impact degree of the high-pressure link and the second preset threshold value meets the preset condition can be realized as follows: when the high-pressure link is non-continuous distribution, the logistics data of the previous logistics link of the high-pressure link is iteratively adjusted until the difference between the target link impact degree of the high-pressure link and the second preset threshold value meets the preset condition; when the high-pressure link is continuous distribution, the logistics data of the logistics link before the continuous distribution is iteratively adjusted until the difference between the target link impact degree of the high-pressure link and the second preset threshold value meets the preset condition.

[0104] Next, in a specific embodiment, the process of identifying the high-pressure link in the logistics link and correcting the logistics data of the associated link of the high-pressure link based on the target link impact degree of the high-pressure link is described in detail:

[0105] In the present specific embodiment, when the user demand mutation degree of the target product of the user in the current period is greater than 0.5 (i.e. the value of the first preset threshold value is 0.5), the logistics link with a target link impact degree greater than 0.9 (i.e. the value of the second preset threshold value is 0.9) is set as the logistics supply high-pressure link in the current period; if there is no logistics supply high-pressure link in the current period, it proves that the logistics supply chain can operate normally and there is no need to optimize the logistics link for logistics coordination; if there is a logistics supply high-pressure link in the current period, it is further judged whether the logistics supply high-pressure link in the current period is a continuously distributed link in the logistics supply chain:

[0106] If the logistics supply high-pressure link is not adjacent to the continuously distributed link in the logistics supply chain, i.e. there is at least one normal logistics link before the logistics supply high-pressure link, in this case, the local pressure is distributed in the logistics supply chain by increasing the logistics supply capacity of the normal logistics link before the logistics supply high-pressure link, i.e. the pressure generated when the user demand for the new product mutates can be relieved, and the specific implementation is as follows: calculating the difference ΔF1 between the target link impact degree of the logistics supply high-pressure link and the second preset threshold value, the greater the value, the greater the degree of adjustment required by the logistics supply high-pressure link from the previous logistics link; modifying the logistics information data of the normal logistics link before the high-pressure link, and iteratively calculating the target link impact degree of the logistics supply high-pressure link to gradually reduce the target link impact degree; repeating the above process until the value of ΔF1 is 0.

[0107] If the logistics supply high-pressure links are adjacent and continuous in the logistics supply chain, for example, the logistics supply high-pressure links are distributed in the logistics supply chain as follows: normal logistics link 1-normal logistics link 2-logistics supply high-pressure link 3-logistics supply high-pressure link 4, the pressure borne by a normal logistics link (normal logistics link 2) before the continuously distributed logistics supply high-pressure links will be greater, and therefore multiple normal logistics links need to be adjusted synchronously to share the pressure of the logistics supply high-pressure links. The specific implementation is as follows: starting from a normal logistics link (normal logistics link 2) before the continuously distributed logistics supply high-pressure links, the logistics information data of the normal logistics link is corrected, and the target link impact degree of the logistics supply high-pressure links is calculated cumulatively, so that the target link impact degree of the logistics supply high-pressure links is gradually reduced until the difference between the logistics supply high-pressure links and the second preset threshold is 0. If the above target cannot be achieved through the current normal logistics link (normal logistics link 2), the above process is performed on the normal logistics link (for example, the above normal logistics link 1) before the current normal logistics link, and so on, until the difference between the logistics supply high-pressure links and the second preset threshold is 0.

[0108] In step S160, the corrected logistics data is input into the logistics big model to optimize the logistics supply chain through the logistics big model.

[0109] In the embodiments of the present application, after the logistics data of the associated links of the high-pressure link is adjusted through the above process to reduce the pressure of the high-pressure link to a bearable range, the adjusted logistics data can be input into the pre-trained logistics big model to realize the collaborative optimization of each logistics link in the logistics supply chain through the logistics big model.

[0110] Specifically, the logistics supply information data of each logistics link determined through the above process can be input into the above logistics big model as standard data to determine the logistics supply preparation of each logistics link in the next period logistics supply chain through the logistics big model, so that the logistics supply state of each logistics link in the next period meets the requirements of the above logistics supply information data.

[0111] This invention collects logistics data for a target product across various distribution dimensions within its supply chain. Based on this data, it determines the degree of change in distribution dimensions and user demand. This allows for the determination of the initial impact of each logistics link based on these factors, and the adjustment of this initial impact based on the position of each link within the supply chain. The corrected impact of the target link captures implicit connections between logistics links, enabling timely identification of impact transmission paths when user demand changes abruptly, thus improving the efficiency of collaborative optimization within the supply chain. Furthermore, after obtaining the target impact, this invention can identify high-pressure links and alleviate their pressure caused by sudden changes in user demand by adjusting the logistics data of related links. Additionally, by inputting the corrected logistics data into a pre-trained logistics model, this invention can quickly adapt the logistics supply chain for new products of different categories.

[0112] The foregoing primarily describes the solutions provided by the embodiments of the present invention from a methodological perspective. To achieve the aforementioned functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, the present invention can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0113] Correspondingly, this disclosure also provides a logistics supply chain collaborative optimization system based on an AI large model, referencing... Figure 2 As shown, the AI-based large-scale model-based logistics supply chain collaborative optimization system 200 may include a data acquisition module 210, a data processing module 220, and a logistics optimization module 230, wherein:

[0114] The data acquisition module 210 is used to collect logistics data of various circulation dimensions in each logistics link of the target product's logistics supply chain;

[0115] The data processing module 220 is used to determine the mutation degree of the corresponding circulation dimension based on the logistics data under each circulation dimension, determine the target circulation dimension in the circulation dimension, and determine the mutation degree of the target circulation dimension as the mutation degree of user demand for the target product.

[0116] The data processing module 220 is also used to determine the initial impact of each logistics link based on the mutation degree of circulation dimension and the mutation degree of user demand.

[0117] The data processing module 220 is further configured to correct the initial link impact degree of each logistics link based on the position information of the logistics link on the logistics supply chain, to obtain a target link impact degree of each logistics link.

[0118] The logistics optimization module 230 is configured to identify a high-pressure link in the logistics link, and correct logistics data of an associated link based on the target link impact degree of the high-pressure link.

[0119] The logistics optimization module 230 is further configured to input the corrected logistics data into a logistics large model, to optimize the logistics supply chain by using the logistics large model.

[0120] In the embodiments of the present application, the data processing module is specifically configured to: obtain a change amount of logistics data of a current period and a previous period in each circulation dimension, calculate a change amount difference between the change amount of the logistics data of the current period and the change amount of the logistics data of the previous period, denoted as a circulation dimension mutation degree corresponding to the circulation dimension; determine a product inventory amount of a target product in the circulation dimension as a target circulation dimension, and determine a circulation dimension mutation degree corresponding to the product inventory amount as a user demand mutation degree.

[0121] In the embodiments of the present application, the data processing module is specifically configured to: for each logistics link, determine a link supply circulation rate of the logistics link based on logistics data in each circulation dimension corresponding to the logistics link; determine a user mutation correlation strength of each circulation dimension corresponding to the logistics link based on the circulation dimension mutation degree of each circulation dimension, the link supply circulation rate, and the user demand mutation degree; and determine an initial link impact degree of each logistics link based on the user mutation correlation strength of each circulation dimension corresponding to each logistics link and the circulation dimension mutation degree; wherein the initial link impact degree is used to describe the impact degree of the current logistics link on the next logistics link under the user demand mutation degree.

[0122] In the embodiments of the present application, the data processing module is specifically configured to: calculate a sum of the circulation dimension mutation degrees of each circulation dimension corresponding to the logistics link; for each circulation dimension, calculate a ratio of the circulation dimension mutation degree of the circulation dimension to the sum of the circulation dimension mutation degrees, and calculate a product of the ratio, the link supply circulation rate, and the user demand mutation degree, to obtain a user mutation correlation strength of the circulation dimension.

[0123] In the embodiment of the present application, the data processing module is specifically configured to: for each logistics link, record a sequence formed by the user mutation correlation strength of each flow dimension corresponding to the logistics link as a first sequence; for a next logistics link of the logistics link, record a sequence formed by the flow dimension mutation degree of each flow dimension corresponding to the next logistics link as a second sequence; match the user mutation correlation strength in the first sequence and the flow dimension mutation degree in the second sequence, and determine the impact correlation influence strength between the user mutation correlation strength and the flow dimension mutation degree corresponding thereto; for each flow dimension of the logistics link, determine the dimension impact strength of the flow dimension based on the user mutation correlation strength of the flow dimension and the impact correlation influence strength corresponding thereto; and calculate the average value of the dimension impact strengths of each flow dimension in the logistics link, and record the average value as the initial link impact degree of the logistics link.

[0124] In the embodiment of the present application, the data processing module is specifically configured to: determine a first core link in the order direction of the logistics supply chain and a second core link in the reverse order direction; for each logistics link, determine the first link interval number between the logistics link and the first core link, and the second link interval number between the logistics link and the second core link; determine the first distance influence degree based on the first link interval number and the total number of logistics links in the logistics supply chain; determine the second distance influence degree based on the second link interval number and the total number of logistics links in the logistics supply chain; and determine the target link impact degree based on the first distance influence degree, the second distance influence degree and the initial link impact degree.

[0125] In the embodiment of the present application, the logistics optimization module is specifically configured to: when the user demand mutation degree is greater than a first preset threshold, trigger a logistics optimization alarm; when the logistics optimization alarm is triggered, identify a logistics link with a target link impact degree greater than a second preset threshold as a high-pressure link; calculate the difference between the target link impact degree of the high-pressure link and the second preset threshold, and adjust the logistics data of the associated link until the difference between the target link impact degree of the high-pressure link and the second preset threshold meets a preset condition.

[0126] In the embodiment of the present application, the logistics optimization module is specifically configured to: when the high-pressure link is a non-continuous distribution, iteratively adjust the logistics data of a previous logistics link of the high-pressure link until the difference between the target link impact degree of the high-pressure link and the second preset threshold meets the preset condition; and when the high-pressure link is a continuous distribution, iteratively adjust the logistics data of a plurality of logistics links before the continuous distribution in turn until the difference between the target link impact degree of the high-pressure link of the continuous distribution and the second preset threshold meets the preset condition.

[0127] In an embodiment of the present application, the above-mentioned AI large model-based logistics supply chain collaborative optimization system can implement the above-mentioned AI large model-based logistics supply chain collaborative optimization method based on an agent workflow and a large language model. Specifically, the agent workflow includes a planner, an executor, a memory, and a reflector. The planner is configured to generate an execution plan for determining the target link impact of each logistics link on the logistics supply chain. The executor is configured to determine the target link impact of each logistics link by executing the execution plan and correct the logistics data of the associated link based on the target link impact of the high-pressure link. The memory is configured to store the target link impact of each logistics link, the intermediate data generated when calculating the target link impact, and the adjustment record of the logistics data. The reflector is configured to identify the high-pressure link in the logistics link and generate an adjustment strategy for the logistics data of the associated link of the high-pressure link.

[0128] In the specific scenario of the AI large model-based logistics supply chain collaborative optimization method and system provided in the embodiments of the present application, the process flow of steps S110 to S140 in the above-mentioned AI large model-based logistics supply chain collaborative optimization method for calculating the target link impact is the execution plan of each sub-step that the planner needs to generate. After the execution plan including each sub-step is generated by the planner, the specific execution of the execution plan is implemented by the executor, and the data and calculation results in the calculation process are stored in the memory. The target link impact calculated is input into the reflector. After receiving the target link impact, the reflector can identify the high-pressure link in the logistics link based on the target link impact and generate an adjustment strategy for the logistics data of the associated link of the high-pressure link. The adjustment strategy has been described in detail in the corresponding position of the above-mentioned AI large model-based logistics supply chain collaborative optimization method, and therefore will not be described here.

[0129] The specific implementation details of the above-mentioned AI large model-based logistics supply chain collaborative optimization system have been described in detail in the corresponding position of the AI large model-based logistics supply chain collaborative optimization method, and therefore will not be described here.

[0130] Figure 3 FIG. 1 is a structural schematic diagram of an electronic device according to an embodiment of the present disclosure. Figure 3 FIG. 2 is a structural schematic diagram of an electronic device according to an embodiment of the present disclosure. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the present disclosure.

[0131] As shown in FIG. 1, the electronic device 100 includes a processor 110, a memory 120, a communication interface 130, and a bus 140. Figure 3As shown, the electronic device 300 can include a processing device (e.g., a central processor, a graphics processor, etc.) 301 that can perform various appropriate actions and processes to implement the valve regulation method of embodiments as described in the present disclosure according to programs stored in a read-only memory (ROM) 302 or loaded into a random access memory (RAM) 303 from a storage device 308. Various programs and data required by the electronic device 300 to operate are also stored in the RAM 303. The processing device 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0132] Generally, the following devices can be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 can allow the electronic device 300 to communicate wirelessly or wired with other devices to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that not all of the shown devices are required to be implemented or present. More or fewer devices can alternatively be implemented or present.

[0133] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts, thereby implementing the valve regulation method as described above. In such embodiments, the computer program can be downloaded and installed from a network through the communication devices 309, or installed from the storage devices 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the methods of embodiments of the present disclosure are performed.

[0134] It should be noted that the computer-readable medium described above can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium, for example, can be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any suitable combination of the foregoing. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program used by or in connection with an instruction execution system, apparatus or device. In the disclosure, the computer-readable signal medium can include a data signal propagated in baseband or propagated as a carrier wave in a propagated data signal, in which the computer-readable program code is contained. Such a propagated data signal can take many forms, including but not limited to, an electromagnetic signal, an optical signal, or any suitable combination of the foregoing. The computer-readable signal medium can also be any computer-readable medium that can send, propagate or transfer the program for use by or in connection with an instruction execution system, apparatus or device. The program code contained in the computer-readable medium can be transmitted by any suitable medium, including but not limited to, wire, cable, RF (radio frequency), etc., or any suitable combination of the foregoing.

[0135] In some embodiments, the client, server, or both can communicate using any current known or future developed network protocol, such as HTTP (Hyper Text Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), the Internet, and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any current known or future developed networks.

[0136] The computer-readable medium described above can be included in the electronic device described above; or can exist separately from the electronic device, and is not assembled into the electronic device.

[0137] The computer-readable medium described above carries one or more programs, when the one or more programs are executed by the electronic device, the electronic device:

[0138] Collect logistics data from various circulation dimensions in each logistics link of the target product's logistics supply chain;

[0139] Based on logistics data under each circulation dimension, determine the circulation dimension mutation degree of the corresponding circulation dimension, determine the target circulation dimension in the circulation dimension, and determine the circulation dimension mutation degree of the target circulation dimension as the user demand mutation degree of the target product.

[0140] The initial impact of each logistics link is determined based on the degree of change in circulation dimension and the degree of change in user demand.

[0141] The initial impact of each logistics link is corrected based on the location information of each logistics link in the logistics supply chain, and the target impact of each logistics link is obtained.

[0142] Identify high-pressure links in the logistics process, and correct the logistics data of related links based on the impact of the target link in the high-pressure link.

[0143] The corrected logistics data is input into the logistics big data model to optimize the logistics supply chain.

[0144] Optionally, when one or more of the above-described procedures are executed by the electronic device, the electronic device may also perform other steps described in the above embodiments.

[0145] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0146] The computer program product of the first aspect can include one or more non-transitory computer-readable media storing instructions that, when executed, cause one or more processors to perform the operations of the method of the first aspect. The computer program product of the first aspect can include a computer-readable medium storing instructions that, when executed, cause one or more processors to perform the operations of the method of the first aspect.

[0147] The units described in the embodiments of the present disclosure can be implemented by software, or by hardware, or by a combination of software and hardware. In some cases, the names of the units do not constitute a limitation on the units themselves.

[0148] The functions described in this document can be implemented in part or in whole in hardware, firmware, software, or any combination thereof. For example, the functions can be implemented in hardware, such as through one or more ASICs modified in entirety for a particular design, or through one or more ASICs having some circuits modified. Said one or more ASICs can comprise one or more processors operating in a vacuum, in air, in a liquid, or in some other medium. In software, the functions can be stored as one or more instructions and / or data on non-transitory computer-readable storage medium or media including, without limitation, memory 104, 204, 304, 404, 504, 604, 704, 804, 904, 1004, 1104, 1204, 1304, and / or 1404 in the aforementioned system, server or device, etc. In that case, the computer-readable storage medium or media can include RAM, ROM, programmable ROM (e.g., EEPROM, erasable programmable ROM, electrically erasable programmable ROM, programmable read-only memory (PROM), etc.), magnetic RAM, flash memory, magnetic disks, optical disks, tape, solid-state memory devices, etc. The aforementioned system, server or device, etc. of this disclosure can include non-transitory computer-readable storage media. The instructions can include, for example, instructions 106, 206, 306, 406, 506, 606, 706, 806, 906, 1006, 1106, 1206, 1306, and / or 1406 or some or all of the modules in the aforementioned system, server or device, etc. The instructions can also include, without limitation, machine code, assembly language code, high-level languages, interpreted code, code written in C, C++, Java, Visual Basic, Python, and / or other suitable programming languages.

[0149] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of a computer program code, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0150] The above description is only preferred embodiments of the present disclosure and the explanation of the technical principles of the present disclosure. It should be understood by those skilled in the art that the disclosed scope of the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by the combinations of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the technical solutions formed by the mutual replacement of the above features and the technical features with similar functions disclosed in the present disclosure (but not limited to) can be used.

[0151] Further, while operations are depicted in a particular order, this should not be understood as requiring the operations to be performed in the particular order shown or in sequential order, and that certain operations can be performed in parallel or concurrently with other operations disclosed herein. In addition, the various elements of the disclosed examples can be combined in a single example or implemented in a separate example. Further, it will be understood that the various features of the disclosed examples can be combined in any suitable sub-combination.

Claims

1. A collaborative optimization method for logistics supply chain based on AI large-scale models, characterized in that the method... include: Collect logistics data from various circulation dimensions in each logistics link of the target product's logistics supply chain; Obtain the change in logistics data for the current period and the previous period under each circulation dimension, calculate the difference between the change in logistics data for the current period and the change in logistics data for the previous period, and record it as the circulation dimension mutation degree of the corresponding circulation dimension; determine the product inventory of the target product in the circulation dimension as the target circulation dimension, and determine the circulation dimension mutation degree corresponding to the product inventory as the user demand mutation degree. For each logistics link, based on logistics data under each circulation dimension corresponding to the logistics link, the link supply circulation rate is determined. The link supply circulation rate is a quantitative indicator that measures the efficiency of goods circulation in each logistics link in the above logistics supply chain within the current cycle. Based on the circulation dimension mutation degree, link supply circulation rate, and user demand mutation degree corresponding to each circulation dimension of the logistics link, the user mutation correlation strength of each circulation dimension corresponding to the logistics link is determined. The user mutation correlation strength is used to measure the impact strength of the circulation dimension on the next logistics link when responding to changes in user demand. Based on the user mutation correlation strength and circulation dimension mutation degree corresponding to each logistics link, the initial link impact degree of each logistics link is determined. The initial link impact degree is used to describe the degree of impact of the current logistics link on the next logistics link under the user demand mutation degree. Based on the location information of each logistics link in the logistics supply chain, the initial link impact is corrected to obtain the target link impact for each logistics link: The first core link in the sequential direction and the second core link in the reverse direction of the logistics supply chain are determined; for each logistics link, the number of first link intervals between the logistics link and the first core link, and the number of second link intervals between the logistics link and the second core link are determined; the first distance impact is determined based on the number of first link intervals and the total number of logistics links in the logistics supply chain; the second distance impact is determined based on the number of second link intervals and the total number of logistics links in the logistics supply chain; the target link impact is determined based on the first distance impact, the second distance impact, and the initial link impact. When the sudden change in user demand exceeds the first preset threshold, a logistics optimization alarm is triggered. When the logistics optimization alarm is triggered, the logistics links with the impact of the target link exceeding the second preset threshold are identified as high-pressure links. The difference between the impact of the target link of the high-pressure link and the second preset threshold is calculated, and the logistics data of the related links are adjusted until the difference between the impact of the target link of the high-pressure link and the second preset threshold meets the preset conditions. The adjusted logistics data is input into a pre-trained logistics model, which enables collaborative optimization of various logistics links in the logistics supply chain.

2. The logistics supply chain collaborative optimization method based on AI large model according to claim 1, characterized in that, Based on the variability of each circulation dimension, the supply circulation rate of each link, and the variability of user demand in the logistics process, the correlation strength of user mutations in each circulation dimension corresponding to the logistics process is determined, including: Calculate the sum of the abrupt changes in the circulation dimensions corresponding to each circulation dimension in the logistics process; For each circulation dimension, the ratio of the circulation dimension mutation degree to the sum of circulation dimension mutation degrees is calculated, and its product with the link supply circulation rate and user demand mutation degree is calculated to obtain the user mutation correlation strength of the circulation dimension.

3. The logistics supply chain collaborative optimization method based on AI large model according to claim 1, characterized in that, Based on the strength of user mutation correlation and the degree of mutation in each circulation dimension corresponding to each logistics link, the initial impact of each logistics link is determined, including: For each logistics link, the sequence formed by the user mutation correlation strength of each circulation dimension corresponding to the logistics link is denoted as the first sequence; For the next logistics link in the logistics process, the sequence formed by the mutation degree of each circulation dimension corresponding to the next logistics link is denoted as the second sequence. For the time series of logistics data of a certain circulation dimension in the first sequence and the time series of logistics data of the corresponding circulation dimension in the second sequence, the two elements closest to each other in the two time series are taken as the optimal correspondence. Based on this distance, the impact transmission effect between the above circulation dimension in the first sequence and the circulation dimension that matches in the second sequence is quantified, and the impact correlation influence intensity corresponding to the circulation dimension in the first sequence is obtained. For each circulation dimension in the logistics process, the dimensional impact intensity of the circulation dimension is determined based on the user mutation correlation strength of the circulation dimension and its corresponding impact correlation strength. The dimensional impact intensity is the comprehensive impact intensity of any circulation dimension in the current logistics process on the corresponding circulation dimension in the next logistics process when the user's demand for the target product changes. The average value of the dimensional impact intensity of each circulation dimension in the logistics process is calculated and denoted as the initial impact intensity of the logistics process.

4. The logistics supply chain collaborative optimization method based on AI large model according to claim 1, characterized in that, Adjust the logistics data of related links until the difference between the impact intensity of the target link in the high-pressure link and the second preset threshold meets the preset conditions, including: When the high-pressure links are discontinuously distributed, the logistics data of the logistics link preceding the high-pressure link is iteratively adjusted until the difference between the impact of the target link of the high-pressure link and the second preset threshold meets the preset condition. When the high-pressure links are continuously distributed, the logistics data of multiple logistics links before the continuous distribution are iteratively adjusted until the difference between the impact degree of the target link of the continuously distributed high-pressure links and the second preset threshold meets the preset condition.

5. A logistics supply chain collaborative optimization system based on an AI large-scale model, characterized in that, The system includes: The data acquisition module is used to collect logistics data from various circulation dimensions in each logistics link of the target product's logistics supply chain; The data processing module is used to obtain the change in logistics data for the current period and the previous period under each circulation dimension, calculate the difference between the change in logistics data for the current period and the change in logistics data for the previous period, and record it as the circulation dimension mutation degree of the corresponding circulation dimension; determine the product inventory of the target product in the circulation dimension as the target circulation dimension, and determine the circulation dimension mutation degree corresponding to the product inventory as the user demand mutation degree. The data processing module is also used to determine the supply-flow rate of each logistics link based on the logistics data of each circulation dimension corresponding to the logistics link. The supply-flow rate is a quantitative indicator that measures the efficiency of goods circulation in each logistics link in the current cycle of the above-mentioned logistics supply chain. Based on the circulation dimension mutation degree, link supply-flow rate and user demand mutation degree of each circulation dimension corresponding to the logistics link, the module determines the user mutation correlation strength of each circulation dimension corresponding to the logistics link. The user mutation correlation strength is used to measure the impact strength of the circulation dimension on the next logistics link when responding to changes in user demand. Based on the user mutation correlation strength and circulation dimension mutation degree of each circulation dimension corresponding to each logistics link, the module determines the initial link impact degree of each logistics link. The initial link impact degree is used to describe the degree of impact of the current logistics link on the next logistics link under the user demand mutation degree. The data processing module is also used to correct the initial impact degree of each logistics link based on the location information of each logistics link in the logistics supply chain, and obtain the target impact degree of each logistics link: determine the first core link in the sequential direction of the logistics supply chain and the second core link in the reverse direction; for each logistics link, determine the number of first link intervals between the logistics link and the first core link, and the number of second link intervals between the logistics link and the second core link; determine the first distance impact degree based on the number of first link intervals and the total number of logistics links in the logistics supply chain; determine the second distance impact degree based on the number of second link intervals and the total number of logistics links in the logistics supply chain; and determine the target impact degree based on the first distance impact degree, the second distance impact degree, and the initial impact degree of the logistics link. The logistics optimization module is used to trigger a logistics optimization alarm when the sudden change in user demand exceeds a first preset threshold. When the logistics optimization alarm is triggered, the logistics links with a target link impact exceeding a second preset threshold are identified as high-pressure links. The difference between the target link impact of the high-pressure link and the second preset threshold is calculated, and the logistics data of the related links are adjusted until the difference between the target link impact of the high-pressure link and the second preset threshold meets the preset conditions. The logistics optimization module is also used to input the adjusted logistics data into a pre-trained logistics model, which enables collaborative optimization of various logistics links in the logistics supply chain.

6. The logistics supply chain collaborative optimization system based on an AI large model according to claim 5, characterized in that, The AI-based large-scale model-based logistics supply chain collaborative optimization system implements the AI-based large-scale model-based logistics supply chain collaborative optimization method according to any one of claims 1 to 4, based on agent workflow and large language model; the agent workflow includes a planner, an executor, a memory, and a reflector; wherein: A planner is used to generate execution plans that determine the target impact level of each logistics link in the logistics supply chain. The actuator is used to determine the target impact level of each logistics link by executing the execution plan; and to correct the logistics data of related links based on the target impact level of the high-pressure link. The memory is used to store the target impact of each logistics link, the intermediate data generated when calculating the target impact, and the adjustment records of the logistics data. A reflector is used to identify high-pressure links in the logistics process; and to generate adjustment strategies for logistics data related to high-pressure links.

Images