Dynamic multi-dimensional supplier portrait construction method and system

Abstract

The invention relates to the technical field of data processing, in particular to a dynamic multi-dimensional supplier portrait construction method and system. The method comprises the steps of obtaining an initial delivery quality value of a target supplier at a target moment; obtaining a similarity degree value of the target supplier and the reference supplier on the initial delivery quality value; taking the similarity values corresponding to the reference suppliers as weights, and performing weighted summation on the delivery quality feature values of different reference suppliers of the target supplier to obtain an adjustment coefficient; and multiplying the initial delivery quality value of the target supplier at the target moment by the adjustment coefficient to obtain a delivery evaluation value of the target supplier at the target moment, so as to determine a construction result of the portrait of the target supplier according to the delivery evaluation values of the target supplier at different moments. Through the technical scheme, the portrait of the supplier can be constructed, so that the supplier can be managed.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for constructing a dynamic multi-dimensional supplier portrait. Background Art

[0002] Suppliers refer to external organizations or individuals that provide products, services, raw materials, parts or other resources to enterprises. Depending on the services or products they provide, suppliers can be divided into raw material suppliers, parts suppliers, finished product suppliers, service providers and technology suppliers.

[0003] Supplier management refers to the systematic management activities that an enterprise conducts on its suppliers. Supplier management includes supplier selection, evaluation, development, collaboration and monitoring. Effective supplier management can ensure that an enterprise obtains high-quality products or services, ensure the stability and efficiency of its own supply chain, and thus reduce the production and operation costs of the enterprise.

[0004] By building supplier portraits, companies can have a more comprehensive understanding of suppliers' supply capabilities, supply quality, credibility, and risk resistance, allowing them to make more informed choices; accurate supplier portraits help companies predict potential problems that may exist in the supply chain and take appropriate handling or preventive measures in advance.

[0005] At the same time, through in-depth analysis of suppliers, enterprises can identify potential risk factors. For example, enterprises can find out whether suppliers have financial instability and delivery quality issues, etc., which enables enterprises to adaptively select suppliers with better supply capabilities, thereby improving the company's ability to resist risks. Therefore, it is necessary to construct a portrait of suppliers. Summary of the invention

[0006] In order to realize the construction of supplier portraits, the present application provides a dynamic multi-dimensional supplier portrait construction method and system.

[0007] According to the first aspect of an embodiment of the present application, a dynamic multi-dimensional supplier portrait construction method is provided, including: obtaining an initial delivery quality value of a target supplier at a target time; the initial delivery quality value is used to evaluate the delivery quality of the target supplier, and the initial delivery quality value is determined based on the delivery volume growth rate, product qualification rate, and delivery punctuality rate of the target supplier; obtaining a similarity value between the target supplier and a reference supplier in the initial delivery quality value, wherein the reference supplier is other suppliers of the same supply type as the target supplier; using the similarity value corresponding to the reference supplier as a weight, and performing weighted summation on the delivery quality characteristic values ​​of different reference suppliers of the target supplier to obtain an adjustment coefficient; the delivery quality characteristic value is equal to the difference in delivery volume growth rate between the reference supplier and the target supplier; multiplying the initial delivery quality value of the target supplier at the target time by the adjustment coefficient to obtain a delivery evaluation value of the target supplier at the target time, so as to determine the construction result of the portrait of the target supplier based on the delivery evaluation value of the target supplier at different times.

[0008] In this way, by obtaining the initial delivery quality value of the target supplier at the target time and considering the initial delivery quality values ​​of other suppliers with the same supply type as the target supplier, the adjustment coefficient for adjusting the initial delivery quality value of the target supplier can be obtained. The delivery quality of the target supplier can be evaluated more objectively by using the delivery evaluation value obtained by multiplying the adjustment coefficient by the initial delivery quality value. Therefore, a more accurate portrait of the target supplier can be constructed.

[0009] Optionally, the initial delivery quality value is determined in the following manner: normalize the delivery volume growth rate, product qualification rate and delivery punctuality rate respectively, and multiply the normalized delivery volume growth rate, the normalized product qualification rate and the normalized delivery punctuality rate to obtain the initial delivery quality value.

[0010] In this way, the initial delivery quality value obtained can comprehensively reflect the performance of the three key indicators of delivery volume growth rate, product qualification rate and delivery punctuality rate, so as to provide a more comprehensive assessment of the supplier's supply situation at the current moment.

[0011] Optionally, the similarity between the target supplier and the reference supplier in terms of initial delivery quality values ​​is determined by: , where D is the similarity between the target supplier and the reference supplier in terms of initial delivery quality, and R is the number of historical moments. is the initial delivery quality value of the target supplier at the rth historical moment, is the initial delivery quality value of the reference supplier at the rth historical moment.

[0012] In this way, the consistency of the initial delivery quality values ​​of the two suppliers can be compared, and the similarity value can better reflect the similarity between the reference supplier and the target supplier in terms of delivery quality.

[0013] Optionally, the similarity value corresponding to the reference supplier is used as a weight, and the delivery quality characteristic values ​​of different reference suppliers of the target supplier are weighted and summed to obtain an adjustment coefficient, including: , where S is the adjustment coefficient, norm is the normalization function, and H is the number of reference suppliers of the target supplier. is the similarity value between the target supplier and the hth reference supplier, are the delivery quality characteristic values ​​of the target supplier and the hth reference supplier.

[0014] In this way, the similarity value corresponding to the reference supplier is used as the weight, and the obtained adjustment coefficient can be better used to adjust the initial delivery quality value of the target supplier.

[0015] Optionally, the result of constructing a portrait of the target supplier is determined based on the delivery evaluation values ​​of the target supplier at different moments, including: determining an abnormal moment from the multiple moments with respect to the delivery evaluation values ​​of the target supplier at multiple moments before the current moment; the difference in delivery evaluation value between the abnormal moment and other moments in the multiple moments is greater than a preset difference threshold; and determining the result of constructing the portrait of the target supplier based on the delivery evaluation value of the target supplier at at least one abnormal moment.

[0016] In this way, at least one abnormal moment of the delivery evaluation value of the enterprise's target supplier is determined, and a portrait of the target supplier is constructed based on the delivery evaluation value of the target supplier at at least one abnormal moment, which helps the enterprise to manage suppliers based on the portrait of the target supplier.

[0017] Optionally, determining the abnormal moment from multiple moments includes: using an anomaly detection algorithm to determine an abnormal delivery evaluation value from the delivery evaluation values ​​of the target supplier at multiple moments before the current moment, and taking the moment where the abnormal delivery evaluation value is located as the abnormal moment; the difference between the abnormal delivery evaluation value and other delivery evaluation values ​​in the delivery evaluation values ​​corresponding to the multiple moments is greater than a preset difference threshold; the anomaly detection algorithm is any one of a box plot algorithm, an interquartile range method, and an isolation forest algorithm.

[0018] Optionally, based on the delivery evaluation value of the target supplier at at least one abnormal moment, the result of constructing the portrait of the target supplier is determined, including: obtaining information on factors affecting delivery quality of the target supplier at the abnormal moment, wherein the information on factors affecting delivery quality is the degree of influence of factors affecting delivery quality on delivery quality; based on the information on factors affecting delivery quality of the target supplier at at least one abnormal moment, determining a portrait label corresponding to the target supplier to complete the construction of the portrait of the target supplier.

[0019] Optionally, the method further includes: sending a prompt message to the target supplier when there is an abnormal moment in multiple moments before the current moment of the target supplier; the prompt message is used to prompt the target supplier that there is an abnormality in the delivery quality at the abnormal moment.

[0020] This will make it easier for target suppliers to improve their delivery quality and help ensure their delivery quality.

[0021] Optionally, the method further includes: determining the priority for displaying different suppliers based on the portraits constructed for different suppliers, so as to display different suppliers according to the priority for displaying different suppliers.

[0022] According to the second aspect of an embodiment of the present application, a dynamic multi-dimensional supplier portrait construction system is provided, including: a processor and a memory, the memory storing computer program instructions, and the computer program instructions, when executed by the processor, implement the steps of the dynamic multi-dimensional supplier portrait construction method provided in the first aspect of the present application.

[0023] The technical solution provided by the embodiments of the present application may include the following beneficial effects: obtaining the initial delivery quality value of the target supplier at the target time, and taking the similarity value corresponding to the reference supplier as the weight, performing weighted summation on the delivery quality characteristic values ​​of different reference suppliers of the target supplier to obtain the adjustment coefficient, and the delivery quality characteristic value is equal to the difference in delivery volume growth rate between the reference supplier and the target supplier. The obtained adjustment coefficient can take into account other suppliers of the same supply type, and the delivery evaluation value obtained by multiplying the adjustment coefficient by the initial delivery quality value can more objectively evaluate the delivery quality of the target supplier. Therefore, a more accurate portrait can be constructed for the target supplier so that the enterprise can better manage its suppliers.

[0024] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1is a flow chart of a method for constructing a dynamic multi-dimensional supplier portrait according to an exemplary embodiment; Figure 2 It is a structural diagram of a dynamic multi-dimensional supplier portrait construction system according to an exemplary embodiment. DETAILED DESCRIPTION

[0026] First, the application scenario of the embodiment of the present application is briefly introduced. In the application scenario of the present application, the enterprise can determine the supply quality of the supplier through the historical transaction records with the same supplier. However, it is difficult to accurately evaluate the supply quality of the supplier only through the historical transaction records with the same supplier. Therefore, it is difficult to obtain a more accurate portrait of the supplier, making it difficult to effectively manage the supplier.

[0027] In view of the above technical problems, the present application provides a method and system for constructing a dynamic multi-dimensional supplier portrait. Figure 1 is a flowchart of a method for constructing a dynamic multi-dimensional supplier portrait according to an exemplary embodiment. Figure 1 As shown, the method includes the following steps.

[0028] In step S101, the initial delivery quality value of the target supplier at the target time is obtained.

[0029] The initial delivery quality value is used to evaluate the delivery quality of the target supplier. The initial delivery quality value is determined based on the target supplier's delivery volume growth rate, product qualification rate, and delivery on-time rate.

[0030] The target moment can be the current moment and any one of multiple historical moments before the current moment; different moments correspond to different moments when the target supplier delivers physical products or services to the enterprise.

[0031] The target supplier can be any one of the company's multiple suppliers; if the target supplier's delivery volume growth rate indicates that the target supplier's delivery volume is increasing, it at least means that the company's demand for physical products or services from the target supplier is increasing, and the target supplier's delivery quality is more in line with the company's actual needs.

[0032] On the contrary, if the target supplier's delivery volume growth rate indicates a decline in the target supplier's delivery volume, it means that the company's demand for the target supplier's products or services has declined, and the target supplier's delivery quality may not meet the company's actual needs, or the company may have reduced its demand for the target supplier's products or services due to the impact of the market environment.

[0033] The product qualification rate can reflect the quality of the physical products or services provided by the target supplier; the higher the product qualification rate of the target supplier, the better the quality of the physical products or services provided by the target supplier; on the contrary, the lower the product qualification rate of the target supplier, the worse the quality of the physical products or services provided by the target supplier.

[0034] The on-time delivery rate can reflect the efficiency of the target supplier in delivering physical products or services; the higher the on-time delivery rate of the target supplier at the target time, the higher the efficiency of the target supplier in delivering physical products or services; on the contrary, the lower the on-time delivery rate of the target supplier at the target time, the lower the efficiency of the target supplier in delivering physical products or services.

[0035] The initial delivery quality value determined by the target supplier's delivery volume growth rate, product qualification rate and delivery punctuality rate can be combined with the target supplier's delivery volume growth rate, product qualification rate and delivery punctuality rate, so as to better characterize the delivery quality of the target supplier at the target time through the initial delivery quality value.

[0036] In one embodiment, the initial delivery quality value is determined by normalizing the delivery volume growth rate, product qualification rate, and delivery on-time rate, respectively, and multiplying the normalized delivery volume growth rate, the normalized product qualification rate, and the normalized delivery on-time rate to obtain the initial delivery quality value.

[0037] By multiplying the normalized delivery volume growth rate, product qualification rate and delivery on-time rate, the initial delivery quality value comprehensively reflects the performance of the three key indicators of delivery volume growth rate, product qualification rate and delivery on-time rate, and can provide a more comprehensive assessment of the supplier's supply situation at the current moment.

[0038] By normalizing the delivery volume growth rate, product qualification rate and delivery on-time rate respectively, the comparability of the three can be improved, and the influence of different dimensions or value ranges on the calculation process can be avoided.

[0039] Normalization processing can normalize the variables to be normalized to the range of 0 to 1. For example, the normalization of the delivery volume growth rate, product qualification rate and delivery on-time rate can be achieved through minimum-maximum standardization, logarithmic transformation, inverse tangent function and sigmoid function.

[0040] In step S102, the similarity value between the target supplier and the reference supplier in terms of initial delivery quality value is obtained.

[0041] The reference supplier is another supplier of the same supply type as the target supplier. Suppliers of the same supply type may refer to suppliers that provide the same type of physical products or services; for example, if the target supplier is a supplier that provides logistics and transportation services to the enterprise, the reference supplier may be another supplier that provides logistics and transportation services to the enterprise. The logistics and transportation services provided by the target supplier and the reference supplier may complement each other to ensure the stability of the enterprise's logistics and transportation services.

[0042] The similarity value between the target supplier and the reference supplier in the initial delivery quality value is used to characterize the similarity between the target supplier's delivery quality and the reference supplier's delivery quality; the greater the similarity value between the target supplier and the reference supplier in the initial delivery quality value, the more similar the delivery performance of the target supplier and the reference supplier at historical moments.

[0043] If it is determined that the target supplier and the reference supplier have similar delivery performance at historical moments based on the similarity values ​​of their initial delivery quality values, then the change in the initial delivery quality value of the reference supplier can provide a reference for the reason for the change in the delivery quality value of the target supplier.

[0044] For example, if the reference supplier and the target supplier have similar performance in the initial delivery quality at a historical moment, it means that the delivery situation of the reference supplier and the target supplier is relatively similar. Since the reference supplier and the target supplier are suppliers of the same supply type, the reference supplier and the target supplier usually face the same or similar road congestion conditions, corporate demand or environmental factors and other influencing factors. Therefore, if the initial delivery quality values ​​of the reference supplier and the target supplier at the current moment both decline, then this decline in the initial delivery quality value of the target supplier is more likely to be caused by road congestion, decreased corporate demand or environmental factors.

[0045] For another example, if the reference supplier and the target supplier have similar performance in initial delivery quality at historical moments, and the initial delivery quality value of the target supplier at the current moment decreases, while the initial delivery quality value of the reference supplier at the current moment increases or remains unchanged, then it means that the target supplier's road congestion, the company's demand or environmental factors and other reasons may not have changed significantly to cause a decline in delivery quality, and the target supplier itself has a higher probability of actually experiencing a decline in delivery quality. Therefore, considering the initial delivery quality value of the reference supplier with the same supply type as the target supplier will help to more objectively evaluate the supply situation of the target supplier.

[0046] In one embodiment, the similarity between the target supplier and the reference supplier in terms of initial delivery quality values ​​is determined in the following manner: , where D is the similarity between the target supplier and the reference supplier in terms of initial delivery quality, and R is the number of historical moments. is the initial delivery quality value of the target supplier at the rth historical moment, is the initial delivery quality value of the reference supplier at the rth historical moment.

[0047] In the calculation formula of the similarity value, the units of the numerator and the denominator are the same, so that the obtained similarity value can avoid the influence of the dimension and the influence of different value ranges of different variables, so as to compare the difference in the initial delivery quality value between the target supplier and the reference supplier. The initial delivery quality value of the reference supplier can be determined by referring to the process of obtaining the initial delivery quality value of the target supplier.

[0048] In multiple historical moments before the current moment, the more consistent the initial delivery quality values ​​of the reference supplier and the target supplier at the same historical moment, the larger the similarity value is and the closer it is to 1.

[0049] On the contrary, in multiple historical moments before the current moment, the more inconsistent the initial delivery quality values ​​of the reference supplier and the target supplier at the same historical moment are, the smaller the similarity value is and the closer it is to 0.

[0050] In this way, based on the initial delivery quality values ​​of the reference supplier and the target supplier at the same historical moment, the consistency of the initial delivery quality values ​​of the two suppliers can be compared, and the obtained similarity value can better characterize the similarity between the reference supplier and the target supplier in delivery quality.

[0051] Since the external environmental factors faced by suppliers of the same supply type are usually similar, for example, a decrease in supply demand from an enterprise leads to a decrease in the delivery volume growth rate of different suppliers, or insufficient supply capacity of raw products from a manufacturing enterprise leads to a decrease in the on-time delivery rate of different suppliers. Therefore, the similarity value between the reference supplier and the target supplier can characterize the reference value of the reference supplier's delivery quality relative to the target supplier, so as to more objectively evaluate the delivery quality of the target supplier.

[0052] In step S103, the similarity value corresponding to the reference supplier is used as a weight, and the delivery quality characteristic values ​​of different reference suppliers of the target supplier are weighted and summed to obtain the adjustment coefficient.

[0053] The delivery quality characteristic value is equal to the difference in delivery volume growth rate between the reference supplier and the target supplier; the delivery quality characteristic value may be equal to the difference in delivery volume growth rate between the target supplier and the reference supplier.

[0054] The larger the delivery quality characteristic value of the target supplier relative to the reference supplier, the better the delivery quality of the target supplier at the current moment. Since the reference supplier is a supplier of the same supply type as the target supplier, the delivery quality characteristic value can reflect the relative quality between the target supplier's delivery quality and other suppliers of the same supply type, so as to more objectively evaluate the supply quality of the company's target supplier.

[0055] Since the similarity value can characterize the similarity between the reference supplier and the target supplier in terms of delivery quality at historical moments, the similarity between the reference supplier and the target supplier in terms of delivery quality at historical moments can reflect the reference value of the reference supplier's delivery quality at the current moment to the target supplier's delivery quality at the current moment.

[0056] The similarity value can reflect the reference value of the reference supplier relative to the target supplier at the current moment. By using the similarity value corresponding to the reference supplier as the weight, a higher weight can be determined for the reference supplier with a higher similarity value, thereby increasing the contribution of the initial delivery quality value of the reference supplier with a higher reference value to the weighted sum, thereby obtaining a more accurate evaluation result of the delivery quality of the target supplier at the current moment.

[0057] In one embodiment, the similarity value corresponding to the reference supplier is used as a weight, and the delivery quality characteristic values ​​of different reference suppliers of the target supplier are weighted and summed to obtain the adjustment coefficient, including: , where S is the adjustment coefficient, norm is the normalization function, and H is the number of reference suppliers of the target supplier. is the similarity value between the target supplier and the hth reference supplier, are the delivery quality characteristic values ​​of the target supplier and the hth reference supplier.

[0058] The normalization function is used to normalize the variable to be normalized to the range of 0 to 1. For example, the normalization processing function can be minimum-maximum standardization, logarithmic transformation, inverse tangent function, sigmoid function, etc.

[0059] The number of reference suppliers of the target supplier may be less than or equal to the number of all suppliers who provide products or services to the same enterprise and whose supply type is the same as that of the target supplier; the larger the delivery quality characteristic value between the target supplier and the reference supplier, the larger the initial delivery quality value of the target supplier at the current moment is relative to the initial delivery quality values ​​of other suppliers with the same supply type, and the greater the probability that the target supplier's delivery quality at the current moment is actually better.

[0060] Alternatively, the larger the delivery quality characteristic value between the target supplier and the reference supplier, if the initial delivery quality value of the target supplier at the current moment is affected by the decline in enterprise demand or environmental factors, it means that under the influence of the external environment, the target supplier still shows a larger initial delivery quality value relative to other suppliers, and the target supplier has better delivery quality at the current moment, which can increase the obtained adjustment coefficient so that after adjusting the initial delivery quality values ​​of all suppliers, a larger delivery evaluation value can be determined for the target supplier than other suppliers.

[0061] In this way, the similarity value corresponding to the reference supplier is used as the weight, and the delivery quality characteristic values ​​of different reference suppliers of the target supplier are weighted and summed to obtain the adjustment coefficient. The obtained adjustment coefficient can be used to obtain a more accurate evaluation result of the delivery quality of the target supplier.

[0062] In step S104, the initial delivery quality value of the target supplier at the target time is multiplied by the adjustment coefficient to obtain the delivery evaluation value of the target supplier at the target time, so as to determine the construction result of the target supplier's portrait according to the delivery evaluation value of the target supplier at different times.

[0063] Since the adjustment coefficient is obtained by weighted summing up the delivery quality characteristic values ​​of different reference suppliers of the target supplier, the adjustment coefficient takes into account the similarity between the delivery quality of the target supplier and the reference supplier at historical moments, and the adjustment coefficient takes into account the difference in the initial delivery quality values ​​between the target supplier and the reference supplier at the current moment, therefore, compared with the unadjusted initial delivery quality value, the delivery evaluation value obtained after adjusting the initial delivery quality value using the adjustment coefficient can more objectively or accurately reflect the delivery quality of the target supplier.

[0064] For example, the unadjusted initial delivery quality value only takes into account factors such as the delivery volume growth rate, product qualification rate, and delivery punctuality rate. When the company's demand for products or services decreases, the target supplier's delivery volume growth rate will show a decrease. Or, when sudden environmental factors such as heavy snow or rain that cannot be attributed to the target supplier cause logistics to be blocked, the target supplier's delivery punctuality rate may be affected, causing the initial delivery quality value to show a decrease.

[0065] Therefore, when the demand for products or services decreases, or the on-time delivery rate is affected by other factors other than the actual delivery capacity of the target supplier, the unadjusted initial delivery quality value is difficult to reflect the actual delivery quality of the target supplier itself.

[0066] When the demand for products or services by an enterprise decreases, in addition to the delivery volume growth rate of the target supplier, the delivery growth rate of other suppliers with the same supply type as the target supply chain may also decrease accordingly.

[0067] Alternatively, the target supplier's on-time delivery rate is affected by other factors other than the target supplier's actual delivery capability, and the reference supplier's on-time delivery rate may also be affected by other factors other than the reference supplier's actual delivery capability, causing the reference supplier's initial delivery quality value to decline.

[0068] Since the adjustment coefficient can reflect the difference between the initial delivery quality value of the target supplier at the target time and the delivery quality value of the reference supplier at the target time, the delivery evaluation value obtained after adjusting the initial delivery quality by the adjustment coefficient can better reflect the target supplier's own delivery capability at the target time, and avoid the impact of other factors other than the supplier's actual delivery capability, such as changes in demand for products or services, on the evaluation results of delivery quality.

[0069] In one embodiment, the result of constructing a portrait of the target supplier is determined based on the delivery evaluation values ​​of the target supplier at different moments, including: determining an abnormal moment from the multiple moments with respect to the delivery evaluation values ​​of the target supplier at multiple moments before the current moment; the difference in delivery evaluation value between the abnormal moment and other moments in the multiple moments is greater than a preset difference threshold; and determining the result of constructing the portrait of the target supplier based on the delivery evaluation value of the target supplier at at least one abnormal moment.

[0070] Since the delivery evaluation value at the current moment can better represent the actual delivery capability of the target supplier at the current moment, in order to achieve a more comprehensive evaluation of the target supplier, the moment when the delivery evaluation value of the target supplier is abnormal can be determined, so that the enterprise can analyze under what circumstances the delivery quality of the target supplier will be abnormal; the enterprise can make adaptive selection of suppliers for the delivery of physical products or services when the delivery quality of the target supplier is abnormal or the delivery quality is about to be abnormal.

[0071] If the difference in delivery evaluation value between a certain moment and other moments is greater than the preset difference threshold, it means that the target supplier's supply capacity at the moment when the difference in delivery evaluation value between other moments is greater than the preset difference threshold is highly isolated. In order to facilitate the enterprise to analyze the supply capacity of the target supplier at different moments, the moment when the difference in delivery evaluation value between other moments among multiple moments is greater than the preset difference threshold can be regarded as an abnormal moment.

[0072] Based on the delivery evaluation value of the target supplier at at least one abnormal moment, the result of constructing the portrait of the target supplier can be determined. For example, the average period at which the abnormal moments of the target supplier occur can be determined, or the supply capacity of the target supplier at the abnormal moments can be determined, so that the enterprise can respond to possible abnormalities in the supply capacity of the target supplier in a timely manner according to the constructed portrait.

[0073] In this way, by determining at least one abnormal moment of the delivery evaluation value of the target supplier of the enterprise, and determining the result of constructing a portrait of the target supplier based on the delivery evaluation value of the target supplier at at least one abnormal moment, the enterprise can avoid the impact of possible abnormalities in the supply capacity of the target supply chain on the enterprise's physical products or services based on the portrait of the target supplier.

[0074] In one embodiment, determining an abnormal moment from multiple moments includes: using an anomaly detection algorithm to determine an abnormal delivery evaluation value from the delivery evaluation values ​​of the target supplier at multiple moments before the current moment, and taking the moment where the abnormal delivery evaluation value is located as the abnormal moment; the difference between the abnormal delivery evaluation value and other delivery evaluation values ​​in the delivery evaluation values ​​corresponding to the multiple moments is greater than a preset difference threshold; the anomaly detection algorithm is any one of a box plot algorithm, an interquartile range method, and an isolation forest algorithm.

[0075] The preset difference threshold can be determined based on the maximum value of the difference between the delivery evaluation values ​​at different times; the preset difference threshold can be equal to 0.7 to 0.8 of the maximum value of the difference between the delivery evaluation values ​​at different times. For example, when the maximum value of the difference between different delivery evaluation values ​​is 0.7, the preset difference threshold can be between 0.49 and 0.56.

[0076] In one embodiment, the result of constructing a portrait of the target supplier is determined based on the delivery evaluation value of the target supplier at at least one abnormal moment, including: obtaining information on factors affecting delivery quality of the target supplier at the abnormal moment, wherein the information on factors affecting delivery quality is the degree of influence of factors affecting delivery quality on delivery quality; determining a portrait label corresponding to the target supplier based on the information on factors affecting delivery quality of the target supplier at at least one abnormal moment, so as to complete the construction of the portrait of the target supplier.

[0077] The factors affecting the delivery quality of the target supplier at abnormal times may be a capital turnover cycle, a raw material supply cycle, a logistics transportation cycle, etc.; the information on factors affecting delivery quality may refer to the degree of impact of the factors affecting delivery quality.

[0078] Based on the information on factors affecting delivery quality of the target supplier at at least one abnormal moment, the portrait label corresponding to the target supplier can be determined to complete the construction of the portrait of the target supplier; the portrait label can be, for example, "insufficient supply capacity in the early month of each month", "low delivery efficiency but low delivery costs" and "high delivery efficiency but high delivery costs".

[0079] After completing the construction of portraits of all suppliers, the enterprise can adaptively select or plan suppliers at the current moment or at a future moment based on the portraits of all suppliers, thereby reducing the enterprise's production and operation costs or increasing the enterprise's profit margin while ensuring the normal production and operation of the enterprise.

[0080] In one embodiment, a prompt message may also be sent to the target supplier when there is an abnormal moment in multiple moments before the current moment of the target supplier; the prompt message is used to prompt the target supplier that there is an abnormality in the delivery quality at the abnormal moment.

[0081] In the case that a target supplier has abnormal moments in multiple moments before the current moment, a prompt message can be sent to the target supplier to remind the target supplier, so that the target supplier can avoid possible abnormalities in supply capacity, ensure the target supplier's supply of physical products or services to the enterprise, and maintain the stability of the supply and demand relationship between the supplier and the enterprise.

[0082] For example, the prompt message may include "The delivery efficiency this time is low, please improve the delivery efficiency in subsequent delivery batches", or the prompt message may include "The delivery qualification rate this time is low, but still higher than the minimum allowed qualification rate, please improve the delivery qualification rate in subsequent delivery batches", so as to facilitate the target supplier to improve the delivery quality.

[0083] In a possible implementation, when the target supplier has abnormal moments in multiple moments before the current moment, the frequency of random inspections of physical products or services from the target supplier may be increased.

[0084] In this way, it is possible to strengthen the supervision of the physical products or services from the target supplier when there are abnormalities in the delivery quality of the target supplier.

[0085] In one embodiment, the priority for displaying different suppliers may also be determined based on the portraits constructed for different suppliers, so that different suppliers may be displayed according to the priority for displaying different suppliers.

[0086] Since different suppliers differ in delivery quality and delivery costs, the portrait constructed for the supplier may at least include the supplier's delivery quality and delivery costs; based on the portraits constructed for different suppliers, the enterprise can set corresponding weight coefficients for delivery quality and delivery costs according to actual needs; based on the weight coefficients corresponding to the delivery quality and delivery costs, as well as the supplier's delivery quality and delivery costs, the priority for displaying the supplier can be determined.

[0087] For example, if the corresponding weight coefficients of delivery quality and delivery cost are set to 0.6 and 0.4 respectively, then the priority value = normalized delivery evaluation value × 0.6 + normalized delivery cost × 0.4. All suppliers are displayed in order of priority value from large to small, which makes it easier to select suppliers from multiple companies that can take into account both delivery quality and delivery costs.

[0088] In this way, different suppliers are displayed according to their display priorities, which can facilitate the management or planning of suppliers by users of the enterprise.

[0089] Figure 2 1 is a schematic diagram of a dynamic multi-dimensional supplier portrait construction system 1000 according to an exemplary embodiment. Figure 2 The dynamic multi-dimensional supplier portrait construction system 1000 includes: a processor 1100 and a memory 1200, wherein the memory 1200 stores computer program instructions, and when the computer program instructions are executed by the processor 1100, all or part of the steps of the dynamic multi-dimensional supplier portrait construction method in this application are implemented.

[0090] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary technical means in the art that are not disclosed in the present application, and the specification and embodiments are only considered as exemplary.

[0091] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A dynamic multi-dimensional supplier portrait construction method, characterized in that: include: Obtain the initial delivery quality value of the target supplier at the target time; The initial delivery quality value is used to evaluate the delivery quality of the target supplier. The initial delivery quality value is determined based on the target supplier's delivery volume growth rate, product qualification rate, and delivery on-time rate. Obtaining a similarity value between the target supplier and the reference supplier in terms of initial delivery quality values, wherein the reference supplier is another supplier of the same supply type as the target supplier; The similarity value corresponding to the reference supplier is used as the weight, and the delivery quality characteristic values ​​of different reference suppliers of the target supplier are weighted and summed to obtain the adjustment coefficient; the delivery quality characteristic value is equal to the difference in the growth rate of delivery volume between the reference supplier and the target supplier; The initial delivery quality value of the target supplier at the target time is multiplied by the adjustment coefficient to obtain the delivery evaluation value of the target supplier at the target time, so as to determine the result of constructing the portrait of the target supplier according to the delivery evaluation value of the target supplier at different times.

2. The method for constructing a dynamic multi-dimensional supplier portrait according to claim 1, characterized in that: The initial delivery quality value is determined as follows: The delivery volume growth rate, product qualification rate and delivery on-time rate are normalized respectively, and the normalized delivery volume growth rate, normalized product qualification rate and normalized delivery on-time rate are multiplied together to obtain the initial delivery quality value.

3. The method for constructing a dynamic multi-dimensional supplier portrait according to claim 1, characterized in that: The similarity between the target supplier and the reference supplier in terms of initial delivery quality values ​​is determined in the following way: , where D is the similarity between the target supplier and the reference supplier in terms of initial delivery quality, and R is the number of historical moments. is the initial delivery quality value of the target supplier at the rth historical moment, is the initial delivery quality value of the reference supplier at the rth historical moment.

4. The method for constructing a dynamic multi-dimensional supplier portrait according to claim 1, characterized in that: The similarity value corresponding to the reference supplier is used as the weight, and the delivery quality characteristic values ​​of different reference suppliers of the target supplier are weighted and summed to obtain the adjustment coefficient, including: , where S is the adjustment coefficient, norm is the normalization function, and H is the number of reference suppliers of the target supplier. is the similarity value between the target supplier and the hth reference supplier, are the delivery quality characteristic values ​​of the target supplier and the hth reference supplier.

5. The method for constructing a dynamic multi-dimensional supplier portrait according to claim 1, characterized in that: According to the delivery evaluation values ​​of the target supplier at different times, the construction results of the target supplier portrait are determined, including: For the delivery evaluation values ​​of the target supplier at multiple moments before the current moment, an abnormal moment is determined from the multiple moments; the difference in delivery evaluation value between the abnormal moment and other moments in the multiple moments is greater than a preset difference threshold; The result of constructing a profile of the target supplier is determined based on the delivery evaluation value of the target supplier at at least one abnormal moment.

6. The method for constructing a dynamic multi-dimensional supplier portrait according to claim 5, characterized in that: Identify abnormal moments from multiple moments, including: By using an anomaly detection algorithm, an abnormal delivery evaluation value is determined from the delivery evaluation values ​​of the target supplier at multiple moments before the current moment, and the moment at which the abnormal delivery evaluation value is located is taken as the abnormal moment; the difference between the abnormal delivery evaluation value and other delivery evaluation values ​​in the corresponding multiple moments is greater than a preset difference threshold; the anomaly detection algorithm is any one of the box plot algorithm, interquartile range method and isolation forest algorithm.

7. The method for constructing a dynamic multi-dimensional supplier portrait according to claim 5, characterized in that: According to the delivery evaluation value of the target supplier at at least one abnormal moment, the result of building the profile of the target supplier is determined, including: Acquiring information on factors influencing delivery quality of the target supplier at an abnormal moment, wherein the information on factors influencing delivery quality is the degree of influence of factors influencing delivery quality on delivery quality; Based on the information of factors affecting the delivery quality of the target supplier at at least one abnormal moment, the portrait label corresponding to the target supplier is determined to complete the construction of the portrait of the target supplier.

8. The method for constructing a dynamic multi-dimensional supplier portrait according to claim 5, characterized in that: The method further comprises: In the case that the target supplier has an abnormal moment in multiple moments before the current moment, a prompt message is sent to the target supplier; the prompt message is used to prompt the target supplier that there is an abnormality in the delivery quality at the abnormal moment.

9. The method for constructing a dynamic multi-dimensional supplier portrait according to claim 1, characterized in that: The method further comprises: Based on the portraits constructed for different suppliers, the priority for displaying different suppliers is determined, and different suppliers are displayed according to the priority for displaying different suppliers.

10. A dynamic multi-dimensional supplier portrait construction system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the dynamic multi-dimensional supplier portrait construction method according to any one of claims 1-9 is implemented.

Images