An artificial intelligence-based intelligent optimization and management system and method for supply chain

Abstract

The present application provides an artificial intelligence-based intelligent optimization and management system and method for a supply chain. By obtaining the historical warehousing data set of a target fresh food supply chain platform, multiple warehousing data nodes are determined based on the historical warehousing data set, and all the warehousing data nodes are converted into a warehousing data node sequence. Then, the local response degree of each warehousing data node in the warehousing data node sequence is determined. The warehousing confluence decision domain is determined through all the local response degrees, and the confluence eigenvalue of the warehousing confluence decision domain is further determined. Furthermore, the abnormal monitoring quantification of the target fresh food supply chain platform is determined according to the confluence eigenvalue and the local response degrees of all the warehousing data nodes. Thus, the target fresh food supply chain platform is monitored and warned through the abnormal monitoring quantification, that is, the warehousing confluence warning for the semi-empty warehouse is realized.

Description

Technical Field

[0001] This application relates to the technical field of big data supply chain. More specifically, this application relates to a supply chain intelligent optimization and management system and method based on artificial intelligence. Background Art

[0002] Supply chain technology is a field that encompasses information technology and management methods, focusing on optimizing, coordinating, and improving the entire process from suppliers to end customers of products or services. These technologies cover aspects such as logistics, inventory management, production planning, demand forecasting, supplier relationship management, and data analysis, aiming to improve efficiency, reduce costs, enhance visibility, mitigate risks, and adapt to rapidly changing market demands, enabling enterprises to be more competitive.

[0003] The monitoring of the big data supply chain platform is based on data analysis and integration technologies to comprehensively understand, track, and optimize the entire supply chain ecosystem and ensure the smooth operation of the supply chain process. However, in existing fresh food supply chain platforms, half-empty warehouses can cause warehousing confluence in the fresh food supply chain. Specifically, warehousing confluence refers to the situation where fresh food products from other warehouses in the fresh food supply chain platform are simultaneously stored in the same half-empty warehouse, causing the half-empty warehouse to need to process a large number of fresh food products in a short period of time, resulting in unsmooth circulation of the supply chain route. Therefore, how to achieve early warning of warehousing confluence in half-empty warehouses has become a difficult problem faced by the industry. Summary of the Invention

[0004] This application provides a supply chain intelligent optimization and management system and method based on artificial intelligence, which can achieve early warning of warehousing confluence in half-empty warehouses.

[0005] In a first aspect, this application provides a supply chain management method based on artificial intelligence, including the following steps:

[0006] Obtain the historical warehousing data set of the target fresh food supply chain platform;

[0007] Determine multiple warehousing data nodes according to the historical warehousing data set, and convert all warehousing data nodes into a warehousing data node sequence;

[0008] Determine the local response degree of each warehousing data node in the warehousing data node sequence, and determine the warehousing confluence decision domain through all local response degrees;

[0009] Determine the confluence characteristic value of the warehousing confluence decision domain, and determine the abnormal monitoring quantification of the target fresh food supply chain platform according to the confluence characteristic value and the local response degrees of all warehousing data nodes;

[0010] Monitor and give early warning to the target fresh food supply chain platform through the abnormal monitoring quantification.

[0011] In some embodiments, converting all the storage data nodes into a storage data node sequence specifically includes:

[0012] Determine the multi-time scales of each storage data node;

[0013] Determine the storage data node sequence according to the multi-time scales of all the storage data nodes.

[0014] In some embodiments, determining the storage confluence decision domain through all the local response degrees specifically includes:

[0015] Suppress all the local response degrees to obtain a storage measure suppression sequence;

[0016] Determine the storage confluence decision domain according to the storage measure suppression sequence.

[0017] In some embodiments, suppressing all the local response degrees to obtain a storage measure suppression sequence specifically includes:

[0018] Determine the scaling suppression eigenvalue of each local response degree;

[0019] Determine the storage measure suppression sequence according to all the scaling suppression eigenvalues.

[0020] In some embodiments, determining the scaling suppression eigenvalue of each local response degree specifically includes:

[0021] Select a local response degree;

[0022] Scale the local response degree to obtain the measure scaling value of the local response degree;

[0023] Suppress the storage far-neighbor distance of the storage data node corresponding to the local response degree through the measure scaling value of the local response degree to obtain the scaling suppression eigenvalue of the local response degree.

[0024] Repeat the above steps to obtain the scaling suppression eigenvalues of the remaining local response degrees.

[0025] In some embodiments, determining the confluence eigenvalue of the storage confluence decision domain specifically includes:

[0026] Determine the confluence coefficient of the storage confluence decision domain;

[0027] Obtain all the local response degrees;

[0028] Obtain all the multi-time scales;

[0029] Determine the confluence eigenvalue of the storage confluence decision domain according to all the local response degrees, all the multi-time scales and the confluence coefficient.

[0030] In some embodiments, the historical warehousing dataset includes the location of each semi-empty warehouse, the remaining inventory capacity of each semi-empty warehouse, and the timestamp of storing various fresh products in each semi-empty warehouse.

[0031] In a second aspect, the present application provides an artificial intelligence-based supply chain management system, including:

[0032] An acquisition module, configured to acquire a historical warehousing dataset of a target fresh product supply chain platform;

[0033] A processing module, configured to determine multiple warehousing data nodes according to the historical warehousing dataset, and convert all the warehousing data nodes into a warehousing data node sequence;

[0034] The processing module is further configured to determine the local response degree of each warehousing data node in the warehousing data node sequence;

[0035] The processing module is further configured to determine a warehousing confluence decision domain through all the local response degrees, and further determine the confluence eigenvalue of the warehousing confluence decision domain;

[0036] The processing module is further configured to determine an abnormal monitoring quantification value of the target fresh product supply chain platform according to the confluence eigenvalue and the local response degrees of all warehousing data nodes;

[0037] An execution module, configured to perform monitoring and early warning on the target fresh product supply chain platform through the abnormal monitoring quantification value.

[0038] In a third aspect, the present application provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-mentioned artificial intelligence-based supply chain management method.

[0039] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned artificial intelligence-based supply chain management method is implemented.

[0040] The technical solutions provided by the disclosed embodiments of the present application have the following beneficial effects:

[0041] In the supply chain management method and system based on artificial intelligence provided by this application, first, a sequence of warehousing data nodes for measuring the busyness of semi-empty warehouses is determined according to the historical warehousing data set. The sequence of warehousing data nodes can characterize the semi-empty warehouses where warehousing confluence occurs. Secondly, by determining the local responsiveness of each warehousing data node in the sequence of warehousing data nodes, the local responsiveness reflects the response speed of the semi-empty warehouses corresponding to different warehousing data nodes when warehousing confluence may occur. Through the local responsiveness, the abnormal conditions of the semi-empty warehouses can be initially judged. Then, a warehousing confluence decision domain that defines the range of abnormal semi-empty warehouses is determined through all the local responsiveness, which can improve the monitoring efficiency of abnormal semi-empty warehouses. Furthermore, the confluence eigenvalue of the warehousing confluence decision domain is determined. Through the confluence eigenvalue, the degree to which abnormal semi-empty warehouses in the warehousing confluence decision domain can interact with each other to enhance warehousing confluence can be measured. Finally, the abnormal monitoring quantification of the target fresh food supply chain platform is determined according to the confluence eigenvalue and the local responsiveness of all warehousing data nodes. Thus, the target fresh food supply chain platform is monitored and warned through the abnormal monitoring quantification, that is, the warehousing confluence warning of the semi-empty warehouses is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is an exemplary flowchart of a supply chain management method based on artificial intelligence shown in some embodiments of this application;

[0043] Figure 2 is a schematic flowchart of determining the local responsiveness of each warehousing data node in the sequence of warehousing data nodes in some embodiments of this application;

[0044] Figure 3 is a schematic flowchart of determining the confluence eigenvalue of the warehousing confluence decision domain in some embodiments of this application;

[0045] Figure 4 is a block diagram of the structure of a supply chain management system based on artificial intelligence in some embodiments of this application;

[0046] Figure 5 is a schematic diagram of the structure of a computer device for implementing a supply chain management method based on artificial intelligence shown in some embodiments of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The core of this application is to obtain the historical warehousing dataset of the target fresh food supply chain platform, then determine multiple warehousing data nodes based on the historical warehousing dataset, convert all the warehousing data nodes into a warehousing data node sequence, thereby determine the local response degree of each warehousing data node in the warehousing data node sequence, determine the warehousing convergence decision domain through all the local response degrees, further determine the convergence eigenvalue of the warehousing convergence decision domain, and then determine the abnormal monitoring quantification of the target fresh food supply chain platform according to the convergence eigenvalue and the local response degrees of all warehousing data nodes, so as to monitor and warn the target fresh food supply chain platform through the abnormal monitoring quantification, that is, realize the warehousing convergence warning for the semi-empty warehouses.

[0048] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners. Refer to Figure 1 , this figure is an exemplary flowchart of a supply chain management method based on artificial intelligence shown in some embodiments of this application. The supply chain management method 100 based on artificial intelligence mainly includes the following steps:

[0049] In step 101, obtain the historical warehousing dataset of the target fresh food supply chain platform.

[0050] Specifically, when implemented, obtain the historical warehousing dataset from the database of the target fresh food supply chain platform. The historical warehousing dataset specifically includes the location of each semi-empty warehouse, the remaining inventory capacity of each semi-empty warehouse, and the timestamp when various fresh products are stored in each semi-empty warehouse. It should be noted that the remaining inventory capacity in this application represents the remaining available capacity of the semi-empty warehouse.

[0051] In step 102, determine multiple warehousing data nodes based on the historical warehousing dataset, and convert all the warehousing data nodes into a warehousing data node sequence.

[0052] Specifically, when implemented, determine multiple warehousing data nodes based on the historical warehousing dataset, that is: create multiple warehousing data nodes, and store the information of the location of each semi-empty warehouse, the remaining inventory capacity of each semi-empty warehouse, and the timestamp when various fresh products are stored in each semi-empty warehouse in the historical warehousing data into the corresponding warehousing data nodes, so as to obtain multiple different warehousing data nodes. It should be noted that the warehousing data node in this application represents a memory block for storing historical warehousing data.

[0053] In some embodiments, converting all the warehousing data nodes into a warehousing data node sequence can be implemented by the following steps:

[0054] Determine the multi-time scale of each warehousing data node;

[0055] Determine the sequence of warehousing data nodes according to the multi-time scales of all warehousing data nodes.

[0056] Among them, in some embodiments, the multi-time scale of each warehousing data node can be determined by the following formula:

[0057]

[0058] Among them, represents the multi-time scale of the th warehousing data node, represents the type of fresh products stored in the semi-empty state warehouse corresponding to the th warehousing data node, represents the time stamp for storing the rd type of fresh product, represents the time stamp for storing the th type of fresh product, represents the base of the natural logarithm.

[0059] It should be noted that the multi-time scale mentioned in this application represents a parameter for measuring the busy degree of the semi-empty state warehouse corresponding to the warehousing data node. The larger the multi-time scale, the busier the semi-empty state warehouse corresponding to the warehousing data node. In addition, the sequence of warehousing data nodes can be used to measure the busy degrees of different semi-empty state warehouses in the fresh product supply chain platform, and the sequence of warehousing data nodes can characterize the semi-empty state warehouses where warehousing confluence occurs.

[0060] Specifically, when implemented, determine the sequence of warehousing data nodes according to the multi-time scales of all warehousing data nodes, that is: sort all warehousing data nodes in descending order according to the multi-time scales of the warehousing data nodes, and use the sequence obtained from the descending order as the transitional sequence of warehousing data nodes. Then, sort the transitional sequence of warehousing data nodes in ascending order according to the remaining inventory capacity of the warehousing data nodes, and use the sequence obtained from the ascending order as the sequence of warehousing data nodes.

[0061] In step 103, determine the local response degree of each warehousing data node in the sequence of warehousing data nodes.

[0062] In some embodiments, as shown in Figure 2 which is a schematic flowchart for determining the local response degree of each warehousing data node in the sequence of warehousing data nodes in some embodiments of this application. In this embodiment, the local response degree of each warehousing data node in the sequence of warehousing data nodes can be implemented by the following steps:

[0063] First, in step 1031, determine the set of warehousing far neighbor distances according to the sequence of warehousing data nodes;

[0064] Secondly, in step 1032, determine the far neighbor equilibrium coefficient of the storage far neighbor distance set;

[0065] Finally, in step 1033, determine the local response degree of each storage data node in the storage data node sequence according to the storage far neighbor distance set and the far neighbor equilibrium coefficient.

[0066] In specific implementation, determine the storage far neighbor distance set according to the storage data node sequence, that is: select the first storage data node in the storage data node sequence, calculate the distance between the semi-empty state warehouse corresponding to the first storage data node and the semi-empty state warehouses corresponding to the adjacent storage data nodes, and take the obtained value as the adjacent factor of the first storage data node. If all the adjacent factors are greater than the distance between the semi-empty state warehouse corresponding to the first storage data node and the semi-empty state warehouses corresponding to the remaining storage data nodes, then take the adjacent factor as the storage far neighbor distance of the first storage data node. If the adjacent factor is less than or equal to the distance between the semi-empty state warehouse corresponding to the first storage data node and any one of the semi-empty state warehouses corresponding to the remaining storage data nodes, then take the maximum distance among the distances between the semi-empty state warehouse corresponding to the first storage data node and the semi-empty state warehouses corresponding to the remaining storage data nodes as the storage far neighbor distance of the first storage data node. Repeat the above steps to sequentially determine the storage far neighbor distances of the remaining storage data nodes in the storage data node sequence, and form a set of all the storage far neighbor distances of the storage data nodes as the storage far neighbor distance set.

[0067] It should be noted that in this application, for the first storage data node and the last storage data node in the storage data node sequence, their adjacent storage data nodes are the second storage data node and the penultimate storage data node respectively. In addition, for the storage data nodes other than the first storage data node and the last storage data node, their adjacent storage data nodes are the two storage data nodes adjacent to the left and right in the storage data node sequence.

[0068] Among them, in specific implementation, determine the far neighbor equilibrium coefficient of the storage far neighbor distance set, that is: calculate the average value of the storage far neighbor distances of all the storage data nodes in the storage far neighbor distance set, and take the obtained value as the far neighbor equilibrium coefficient of the storage far neighbor distance set. It should be noted that in this application, the far neighbor equilibrium coefficient represents a parameter for measuring the trend of the storage far neighbor distances of the storage data nodes in the storage far neighbor distance set concentrated in the high numerical range. The larger the far neighbor equilibrium coefficient, the higher the trend of the storage far neighbor distances of the storage data nodes in the storage far neighbor distance set concentrated in the high numerical range.

[0069] Among them, in some embodiments, determining the local response degree of each warehousing data node in the warehousing data node sequence according to the warehousing far - neighbor distance set and the far - neighbor equilibrium coefficient can be implemented by the following steps:

[0070] Obtain the distance between the semi - vacant state warehouse corresponding to the th warehousing data node in the warehousing data node sequence and the semi - vacant state warehouse corresponding to the th warehousing data node; ;

[0071] Obtain the far - neighbor equilibrium coefficient of the warehousing far - neighbor distance set ;

[0072] Obtain the warehousing far - neighbor distance of the th warehousing data node in the warehousing far - neighbor distance set ;

[0073] Obtain the adjacent factor of the th warehousing data node in the warehousing far - neighbor distance set ;

[0074] According to the distance between the semi - vacant state warehouse corresponding to the th warehousing data node in the warehousing data node sequence and the semi - vacant state warehouse corresponding to the th warehousing data node, the far - neighbor equilibrium coefficient of the warehousing far - neighbor distance set , the warehousing far - neighbor distance of the th warehousing data node in the warehousing far - neighbor distance set and the adjacent factor of the th warehousing data node in the warehousing far - neighbor distance set determine the local response degree of each warehousing data node in the warehousing data node sequence. Among them, the local response degree of the th warehousing data node in the warehousing data node sequence can be determined by the following formula: where,

[0075]

[0076] where, represents the local response degree of the th warehousing data node in the warehousing data node sequence, represents the total number of warehousing data nodes in the warehousing far - neighbor distance set, represents the total number of warehousing data nodes in the warehousing data node sequence.

[0077] It should be noted that the local responsiveness described in this application reflects the response speed of the semi-empty warehouse corresponding to the warehousing data node to the possible warehousing confluence. The greater the local responsiveness, the faster the response speed of the semi-empty warehouse corresponding to the warehousing data node to the possible warehousing confluence. Through the local responsiveness, the abnormal conditions of the semi-empty warehouse can be preliminarily judged.

[0078] In step 104, the warehousing confluence decision domain is determined through all the local responsiveness, and then the confluence eigenvalue of the warehousing confluence decision domain is determined.

[0079] In some embodiments, determining the warehousing confluence decision domain through all the local responsiveness can be implemented by the following steps:

[0080] Perform flattening on all the local responsiveness to obtain a warehousing measure flattening sequence;

[0081] Determine the warehousing confluence decision domain according to the warehousing measure flattening sequence.

[0082] Among them, in some embodiments, performing flattening on all the local responsiveness to obtain a warehousing measure flattening sequence can be implemented by the following steps:

[0083] Determine the scaling and flattening eigenvalue of each local responsiveness;

[0084] Determine the warehousing measure flattening sequence according to all the scaling and flattening eigenvalues.

[0085] Specifically, when implementing, determining the warehousing measure flattening sequence according to all the scaling and flattening eigenvalues, that is: perform a descending order sorting on all the scaling and flattening eigenvalues, and the sequence obtained after the descending order sorting is used as the warehousing measure flattening sequence.

[0086] Among them, in some embodiments, determining the scaling and flattening eigenvalue of each local responsiveness can be implemented by the following steps:

[0087] Select a local responsiveness;

[0088] Perform scaling on this local responsiveness to obtain the measure scaling value of this local responsiveness;

[0089] Perform flattening on the warehousing far-neighbor distance of the warehousing data node corresponding to this local responsiveness through the measure scaling value of this local responsiveness to obtain the scaling and flattening eigenvalue of this local responsiveness.

[0090] Repeat the above steps to obtain the scaling and flattening eigenvalues of the remaining local responsiveness.

[0091] In specific implementation, take the natural logarithm of the largest local response degree among all local response degrees, and use the obtained value as the maximum logarithmic response measure. Take the natural logarithm of the smallest local response degree among all local response degrees, and use the obtained value as the minimum logarithmic response measure. Take the natural logarithm of this local response degree, and use the obtained value as the logarithmic response measure of this local response degree. Divide the logarithmic response measure by the difference between the maximum logarithmic response measure and the minimum logarithmic response measure, and use the obtained quotient as the measure scaling value of this local response degree. For example, the measure scaling value of this local response degree , where represents this local response degree, represents the maximum logarithmic response measure, represents the minimum logarithmic response measure; Normalize the storage far-neighbor distance of the storage data node corresponding to this local response degree, and multiply the obtained value by the measure scaling value of this local response degree. Use the obtained product as the scaling and suppressing eigenvalue of this local response degree. For example, the scaling and suppressing eigenvalue of this local response degree , where, represents the measure scaling value of this local response degree, represents the storage far-neighbor distance of the storage data node corresponding to this local response degree, represents the maximum storage far-neighbor distance of the storage data nodes in the storage far-neighbor distance set, represents the minimum storage far-neighbor distance of the storage data nodes in the storage far-neighbor distance set. It should be noted that in this application, one scaling and suppressing eigenvalue corresponds to a semi-empty warehouse, that is: one scaling and suppressing eigenvalue corresponds to one local response degree, one local response degree corresponds to one storage data node, and one storage data node corresponds to one semi-empty warehouse.

[0092] It should be noted that in this application, the suppression can avoid the situation that the storage measure suppression sequence is only affected by the storage far-neighbor distance or the local response degree due to the uneven distribution of the storage far-neighbor distance and the local response degree.

[0093] Among them, in some embodiments, determining the storage confluence decision domain according to the storage measure suppression sequence can be implemented by the following steps:

[0094] Determine the suppression feature confluence curve according to the storage measure suppression sequence;

[0095] Determine the storage confluence decision domain through the suppression feature confluence curve.

[0096] In specific implementation, a flattening feature confluence curve is determined according to the storage measurement flattening sequence, that is: in the order of the storage measurement flattening sequence, the first scaled flattening feature value is selected, and this scaled flattening feature value is used as the first ordinate value, the number of types of fresh products in the semi-empty warehouse corresponding to this scaled flattening feature value is used as the first abscissa value, the coordinate formed by the first ordinate value and the first abscissa value is used as the feature point of the first scaled flattening feature value, the second scaled flattening feature value is selected, the second ordinate value and the second abscissa value corresponding to the second scaled flattening feature value are determined, and the coordinate formed by the second ordinate value and the second abscissa value is used as the feature point of the second scaled flattening feature value. The above steps are repeated to successively determine the feature points of the remaining scaled flattening feature values. All the feature points of the scaled flattening feature values are mapped onto a two-dimensional plane. The feature point of the first scaled flattening feature value is used as the starting point, and the feature point of the last scaled flattening feature value is used as the ending point. The ending point and the starting point are connected to form a flattening feature confluence curve, and the expression of this flattening feature confluence curve is obtained from the ordinate values and abscissa values of the starting point and the ending point.

[0097] Among them, a storage confluence decision domain is determined according to the flattening feature confluence curve, that is: the distances from all the feature points of the scaled flattening feature values to the flattening feature confluence curve are obtained, the scaled flattening feature value corresponding to the feature point with the largest distance is selected, and the positions of the semi-empty warehouses corresponding to this scaled flattening feature value, the position of the semi-empty warehouse corresponding to the first scaled flattening feature value, and the position of the semi-empty warehouse corresponding to the last scaled flattening feature value are connected pairwise to form a triangular region, and this triangular region is used as the storage confluence decision domain.

[0098] It should be noted that in this application, the storage confluence decision domain represents the region composed of semi-empty warehouses with a relatively high probability of anomalies. The range of abnormal semi-empty warehouses can be delineated through the storage confluence decision domain, and the monitoring efficiency of abnormal semi-empty warehouses can be improved.

[0099] In some embodiments, as shown in Figure 3 This figure is a schematic flowchart of the process for determining the confluence feature values of the storage confluence decision domain in some embodiments of this application. The confluence feature values of the storage confluence decision domain in this embodiment can be implemented by the following steps:

[0100] In step 1041, the confluence coefficient of the storage confluence decision domain is determined;

[0101] Secondly, in step 1042, all local response degrees are obtained;

[0102] Then, in step 1043, all multi-time scales are obtained;

[0103] Finally, in step 1044, the confluence eigenvalue of the warehousing confluence decision domain is determined by all local response degrees, all multi-time scales, and the confluence coefficient.

[0104] Specifically, when implemented, the confluence coefficient of the warehousing confluence decision domain is determined as follows: the average value of the remaining inventory capacities of all semi-empty warehouses in the warehousing confluence decision domain is obtained, and the obtained average value is transformed by an exponential negative power function with as the base, and the transformed value is used as the confluence coefficient of the warehousing confluence decision domain. For example, the confluence coefficient of the warehousing confluence decision domain represents the average value of the remaining inventory capacities of all semi-empty warehouses in the warehousing confluence decision domain, represents the base of the natural logarithm,

[0105] It should be noted that the confluence coefficient is a parameter that measures the abnormality degree of all semi-empty warehouses in the warehousing confluence decision domain. The smaller the confluence coefficient, the higher the abnormality degree of all semi-empty warehouses in the warehousing confluence decision domain.

[0106] Among them, in some embodiments, determining the confluence eigenvalue of the warehousing confluence decision domain by all local response degrees, all multi-time scales, and the confluence coefficient can be implemented by the following steps:

[0107] Obtain the confluence coefficient of the warehousing confluence decision domain ;

[0108] Obtain the local response degree corresponding to the th semi-empty warehouse in the warehousing confluence decision domain ;

[0109] Obtain the multi-time scale corresponding to the th semi-empty warehouse in the warehousing confluence decision domain ;

[0110] According to the confluence coefficient of the warehousing confluence decision domain, the local response degree corresponding to the th semi-empty warehouse in the warehousing confluence decision domain, and the multi-time scale corresponding to the th semi-empty warehouse in the warehousing confluence decision domain, determine the confluence eigenvalue of the warehousing confluence decision domain, where the confluence eigenvalue can be determined by the following formula:

[0111]

[0112] Among them, represents the confluence eigenvalue of the warehousing confluence decision domain represents the maximum local response degree corresponding to the semi-empty warehouses in the warehousing confluence decision domain represents the minimum local response degree corresponding to the semi-empty warehouses in the warehousing confluence decision domain represents the maximum multi-time scale of the warehousing data nodes corresponding to the semi-empty warehouses in the warehousing confluence decision domain the minimum multi-time scale of the warehousing data nodes corresponding to the semi-empty warehouses in the warehousing confluence decision domain represents the base of the natural logarithm represents the total number of semi-empty warehouses in the warehousing confluence decision domain

[0113] It should be noted that the confluence eigenvalue in this application is determined according to the local response degree and the multi-time scale. Through the confluence eigenvalue, it can be judged to what extent the abnormal semi-empty warehouses in the warehousing confluence decision domain can interact with each other to enhance the warehousing confluence. The larger the confluence eigenvalue, the greater the extent to which the abnormal semi-empty warehouses in the warehousing confluence decision domain can interact with each other to enhance the warehousing confluence

[0114] In step 105, an abnormal monitoring quantification of the target fresh food supply chain platform is determined according to the confluence eigenvalue and the local response degrees of all warehousing data nodes

[0115] In some embodiments, determining the abnormal monitoring quantification of the target fresh food supply chain platform according to the confluence eigenvalue and the local response degrees of all warehousing data nodes can be implemented by the following steps

[0116] Obtain the local response degree of the th warehousing data node in the warehousing data node sequence ;

[0117] Obtain the confluence eigenvalue of the warehousing confluence decision domain ;

[0118] According to the local response degree of the th warehousing data node in the warehousing data node sequence and the confluence eigenvalue of the warehousing confluence decision domain determine the abnormal monitoring quantification of the target fresh food supply chain platform, where the abnormal monitoring quantification can be determined by the following formula

[0119]

[0120] where represents the abnormal monitoring quantification of the target fresh food supply chain platform Indicates the total number of warehousing data nodes in the warehousing data node sequence.

[0121] It should be noted that the abnormal monitoring quantification described in this application can be used to judge the impact degree of the warehousing confluence generated by the semi-empty warehouses in the warehousing confluence decision domain on the entire fresh food supply chain platform. The larger the abnormal monitoring quantification, the greater the impact degree of the warehousing confluence generated by the semi-empty warehouses in the warehousing confluence decision domain on the entire fresh food supply chain platform.

[0122] In step 106, the target fresh food supply chain platform is monitored and warned through the abnormal monitoring quantification.

[0123] Specifically, when the abnormal monitoring quantification is less than or equal to the preset monitoring threshold, all the semi-empty warehouses in the warehousing confluence decision domain are determined not to have a warehousing confluence. When the abnormal monitoring quantification is greater than the preset monitoring threshold, all the semi-empty warehouses in the warehousing confluence decision domain are determined to have a warehousing confluence, and a warning is sent to the target fresh food supply chain platform at the same time.

[0124] It should be noted that in this application, the monitoring threshold can be set according to the specific requirements of the target fresh food supply chain platform. For example, if it is required that the target fresh food supply chain platform has a high sensitivity to the warehousing confluence of semi-empty warehouses, the monitoring threshold can be set within a lower range of abnormal monitoring quantification.

[0125] In addition, on the other hand of this application, in some embodiments, this application provides a supply chain management system based on artificial intelligence. Refer to Figure 4 , this figure is a schematic diagram of the exemplary hardware and / or software of the supply chain management system based on artificial intelligence shown in some embodiments of this application. The supply chain management system 400 based on artificial intelligence includes: an acquisition module 401, a processing module 402, and an execution module 403, which are described as follows:

[0126] Acquisition module 401, in this application, the acquisition module 401 is mainly used to acquire the historical warehousing data set of the target fresh food supply chain platform;

[0127] Processing module 402, in this application, the processing module 402 is used to determine a plurality of warehousing data nodes according to the historical warehousing data set and convert all the warehousing data nodes into a warehousing data node sequence;

[0128] Specifically, in this application, the processing module 402 is further used to determine the local response degree of each warehousing data node in the warehousing data node sequence;

[0129] In addition, the processing module 402 described in the present application is further configured to determine a warehousing convergence decision domain based on all local response degrees, and then determine the convergence eigenvalue of the warehousing convergence decision domain;

[0130] In addition, the processing module 402 is further configured to determine the abnormal monitoring quantification of the target fresh food supply chain platform according to the convergence eigenvalue and the local response degrees of all warehousing data nodes;

[0131] The execution module 403, in the present application, the execution module 403 is mainly configured to perform monitoring and early warning on the target fresh food supply chain platform through the abnormal monitoring quantification.

[0132] In addition, the present application also provides a computer device, the computer device includes a memory and a processor, the memory stores code, and the processor is configured to obtain the code and execute the above-mentioned artificial intelligence-based supply chain management method.

[0133] In some embodiments, refer to Figure 5 , this figure is a schematic structural diagram of a computer device applying the artificial intelligence-based supply chain management method according to some embodiments of the present application. The artificial intelligence-based supply chain management method in the above embodiments can be implemented by Figure 5 the computer device shown, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0134] The processor 501 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more for controlling the execution of the artificial intelligence-based supply chain management method in the present application.

[0135] The communication bus 502 may include a path for transmitting information between the above components.

[0136] The memory 503 can be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or it can also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but not limited to this. The memory 503 can exist independently and be connected to the processor 501 through the communication bus 502. The memory 503 can also be integrated with the processor 501.

[0137] Among them, the memory 503 is used to store the program code for executing the solution of this application, and is controlled by the processor 501 to execute. The processor 501 is used to execute the program code stored in the memory 503. The program code can include one or more software modules. The determination of the supply chain aggregation area in the above embodiments can be implemented through one or more software modules in the program code of the processor 501 and the memory 503.

[0138] The communication interface 504 uses any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0139] In a specific implementation, as an embodiment, the computer device can include multiple processors, and each of these processors can be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, the processor can refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).

[0140] The above computer device can be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device can be a desktop computer, a laptop computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of the present application do not limit the type of the computer device.

[0141] In addition, the present application also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above-mentioned artificial intelligence-based supply chain management method is implemented.

[0142] In summary, in the artificial intelligence-based supply chain management method and system disclosed in the embodiments of the present application, by obtaining the historical warehousing data set of the target fresh food supply chain platform, and then determining a plurality of warehousing data nodes according to the historical warehousing data set, converting all the warehousing data nodes into a warehousing data node sequence, thereby determining the local response degree of each warehousing data node in the warehousing data node sequence, determining the warehousing confluence decision domain through all the local response degrees, further determining the confluence characteristic value of the warehousing confluence decision domain, and then determining the abnormal monitoring quantification of the target fresh food supply chain platform according to the confluence characteristic value and the local response degrees of all the warehousing data nodes, so as to monitor and warn the target fresh food supply chain platform through the abnormal monitoring quantification, that is, the warehousing confluence warning for the half-empty warehouse is realized.

[0143] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0144] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A supply chain management method based on artificial intelligence, characterized in that, It includes the following steps: Obtain the historical warehousing data set of the semi-empty warehouses of the target fresh food supply chain platform; Determine multiple warehousing data nodes based on the historical warehousing data set, and convert all the warehousing data nodes into a warehousing data node sequence; including: Determine the multi-time scales of each warehousing data node; determine the warehousing data node sequence according to the multi-time scales of all the warehousing data nodes; The multi-time scales are determined by the following formula: Among them, represents the multi-time scale of the th warehousing data node, represents the type of fresh products stored in the semi-empty state warehouse corresponding to the th warehousing data node, represents the timestamp for storing the th type of fresh product, represents the timestamp for storing the th type of fresh product, represents the base of the natural logarithm; Determine the local response degree of each warehousing data node in the warehousing data node sequence; including: determine the warehousing far-neighbor distance set according to the warehousing data node sequence; determine the far-neighbor equilibrium coefficient of the warehousing far-neighbor distance set; wherein, calculate the average value of the warehousing far-neighbor distances of all the warehousing data nodes in the warehousing far-neighbor distance set, and take the obtained value as the far-neighbor equilibrium coefficient of the warehousing far-neighbor distance set; determine the local response degree of each warehousing data node in the warehousing data node sequence according to the warehousing far-neighbor distance set and the far-neighbor equilibrium coefficient; Determine the warehousing confluence decision domain through all the local response degrees, and further determine the confluence characteristic value of the warehousing confluence decision domain; including: determine the confluence coefficient of the warehousing confluence decision domain; obtain the local response degrees of all the warehousing confluence decision domains; obtain the multi-time scales of all the warehousing confluence decision domains; determine the confluence characteristic value of the warehousing confluence decision domain through the local response degrees of all the warehousing confluence decision domains, the multi-time scales in all the warehousing confluence decision domains, and the confluence coefficient; Among them, the average value of the remaining inventory capacities of all semi-empty warehouses in the warehousing convergence decision domain is obtained, and an exponential negative power function with [this value] as the base is used to transform the obtained average value, and the transformed value is used as the convergence coefficient of the warehousing convergence decision domain; The warehousing confluence decision domain represents the area composed of semi-empty warehouses with a relatively high probability of abnormality; determine the abnormal monitoring quantification of the target fresh food supply chain platform according to the confluence characteristic value and the local response degrees of all the warehousing data nodes; Monitor and give early warnings to the target fresh food supply chain platform through the abnormal monitoring quantification.

2. The method according to claim 1, wherein Determining the warehousing confluence decision domain through all the local response degrees specifically includes: Suppress all the local response degrees to obtain a warehousing measurement suppression sequence; Determine the warehousing confluence decision domain according to the warehousing measurement suppression sequence.

3. The method according to claim 2, characterized in that Suppressing all the local response degrees to obtain a warehousing measurement suppression sequence specifically includes: Determine the scaling and suppression characteristic value of each local response degree; Determine the warehousing measurement suppression sequence according to all the scaling and suppression characteristic values.

4. The method according to claim 3, wherein Determining the scaling and suppression characteristic value of each local response degree specifically includes: Select a local response degree; Scale the selected local response degree to obtain the measurement scaling value of the selected local response degree; Suppress the warehousing far-neighbor distance of the warehousing data node corresponding to the selected local response degree through the measurement scaling value of the selected local response degree to obtain the scaling and suppression characteristic value of the selected local response degree; Repeat the above steps to obtain the scaling and suppression characteristic values of the remaining local response degrees.

5. The method according to claim 1, characterized in that The historical warehousing data set includes the location of each semi-empty warehouse, the remaining inventory capacity of each semi-empty warehouse, and the time stamps of storing various fresh products in each semi-empty warehouse.

6. A supply chain management system based on artificial intelligence, characterized in that, Including: An acquisition module, which acquires the historical warehousing data set of the semi-empty warehouses of the target fresh food supply chain platform; A processing module determines the multi-time scales of each warehousing data node; Determine a warehousing data node sequence according to the multi-time scales of all warehousing data nodes; The multi-time scales are determined by the following formula: Among them, represents the multi-time scale of the th warehousing data node, represents the type of fresh products stored in the semi-empty state warehouse corresponding to the th warehousing data node, represents the time stamp for storing the th type of fresh product, represents the time stamp for storing the th type of fresh product, represents the base of the natural logarithm; Determine the local response degree of each warehousing data node in the warehousing data node sequence; including: determining a warehousing far-neighbor distance set according to the warehousing data node sequence; determining a far-neighbor equilibrium coefficient of the warehousing far-neighbor distance set; wherein, taking the mean value of the warehousing far-neighbor distances of all warehousing data nodes in the warehousing far-neighbor distance set as the far-neighbor equilibrium coefficient of the warehousing far-neighbor distance set; determining the local response degree of each warehousing data node in the warehousing data node sequence according to the warehousing far-neighbor distance set and the far-neighbor equilibrium coefficient; Determine a warehousing confluence decision domain through all local response degrees, and further determine the confluence characteristic value of the warehousing confluence decision domain; including: determining a confluence coefficient of the warehousing confluence decision domain; obtaining the local response degrees of all the warehousing confluence decision domains; obtaining the multi-time scales of all warehousing confluence decision domains; determining the confluence characteristic value of the warehousing confluence decision domain through the local response degrees of all warehousing confluence decision domains, the multi-time scales in all warehousing confluence decision domains, and the confluence coefficient; Among them, the average value of the remaining inventory capacities of all semi-empty warehouses in the warehousing convergence decision domain is obtained, and the obtained average value is converted by an exponential negative power function with [this value] as the base, and the converted value is used as the convergence coefficient of the warehousing convergence decision domain. The warehousing confluence decision domain represents an area composed of semi-empty warehouses with a relatively high probability of abnormality; determine the abnormal monitoring quantification of the target fresh food supply chain platform according to the confluence characteristic value and the local response degrees of all warehousing data nodes; An execution module is used to monitor and give early warnings to the target fresh food supply chain platform through the abnormal monitoring quantification.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the artificial intelligence-based supply chain management method described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the artificial intelligence-based supply chain management method described in any one of claims 1 to 5.

Images