Automobile supply chain monitoring method and system based on Internet of Things

Abstract

The present invention relates to the field of supply chain monitoring technology, and in particular to an automobile supply chain monitoring method and system based on the Internet of Things. The method comprises the steps of: obtaining the supply importance of a type of parts through the number of supplier levels of a type of parts, the total sales volume of all first-level suppliers of this type of parts, the inverse of the number of first-level suppliers, and the average unit price of orders of all first-level suppliers; converting the supply importance of the parts into a relative probability distribution to obtain the importance of the parts; taking the daily inventory of each type of parts and the sales volume of suppliers at all levels as a sample of this type of parts, dividing the importance of the parts into the sampling probability of each sample in this type of parts, and using the sampling probability of each sample in the isolation forest algorithm to construct an isolation forest to achieve automobile supply chain monitoring, thereby effectively improving the accuracy of automobile supply chain monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain monitoring, and in particular to an automobile supply chain monitoring method and system based on the Internet of Things. Background Art

[0002] The automotive supply chain is a collection of a series of links and processes from raw material procurement, parts manufacturing, vehicle assembly to sales and after-sales service. Its role is to ensure efficient and smooth automobile production and sales to meet market demand. Through the real-time collection and transmission of relevant data from all links of the automotive supply chain through the Internet of Things, the security monitoring of the automotive supply chain can be achieved, and the collaborative efficiency and overall operation level of the supply chain can be improved.

[0003] In order to achieve safety monitoring of the automobile supply chain, the prior art provides a variety of monitoring methods for the automobile supply chain. For example, a patent application document with publication number CN115496366A discloses a supply chain security assessment optimization method and system. The application queries the manufacturers that produce each component in the automobile; constructs a supply chain for the manufacturers to produce the automobile; calculates the importance of each manufacturer relative to the production of the automobile based on the supply chain to construct a model; monitors the time it takes for the manufacturer to produce the components; if the time exceeds the rated value set for the manufacturer, queries other manufacturers that have a correlation with the current manufacturer in the model; and performs security alarm operations on the production of components by other manufacturers.

[0004] The above existing technologies focus on monitoring the production status of parts manufacturers, but cannot fully understand the overall operation of the supply chain. Lack of transparency in the supply chain may result in companies being unable to accurately grasp key information such as inventory levels and parts status, thereby affecting the efficiency and response speed of the automotive supply chain.

[0005] Based on this, how to accurately obtain automobile supply chain monitoring results is an urgent problem to be solved by technical personnel in this field. Summary of the invention

[0006] In order to solve the technical problem of how to accurately obtain automobile supply chain monitoring results, the present invention provides an automobile supply chain monitoring method and system based on the Internet of Things.

[0007] In a first aspect, the present invention provides an automobile supply chain monitoring method based on the Internet of Things, which adopts the following technical solutions:

[0008] The automobile supply chain monitoring method based on the Internet of Things includes the following steps:

[0009] The inventory, number of supplier levels, number of first-tier suppliers, and order price and total sales of various types of automotive parts are obtained; the ratio of the number of supplier levels of a type of parts to the total sales of all first-tier suppliers of this type of parts is recorded as the first feature of this type of parts; the reciprocal of the number of first-tier suppliers of this type of parts, the average order price of all first-tier suppliers and the sum of the first feature are normalized to obtain the supply importance of this type of parts; the supply importance of parts is converted into relative probability distribution to obtain the importance of the parts; the daily inventory of each type of parts and the sales volume of suppliers at all levels are taken as a sample of this type of parts, the importance of the parts is evenly divided into the sampling probability of each sample in this type of parts, and the sampling probability of each sample is used in the isolation forest algorithm to construct an isolation forest to achieve automotive supply chain monitoring.

[0010] When obtaining the monitoring results of the automobile supply chain, the present invention processes the characteristics of various parts in the automobile supply chain through the isolation forest algorithm, and can accurately obtain the risk characteristics in each automobile supply chain. In this process, the present invention takes into account that different categories of parts have different degrees of influence on the stability of the automobile supply chain, and errors may occur when dividing based on the isolation forest algorithm; based on this, the present invention obtains the importance of various parts by analyzing the suppliers and sales of various parts, and uses the importance of various parts to set the sampling probability when dividing based on the isolation forest algorithm, so that important parts can get more attention and more detailed division in the isolation tree division process, accurately obtain the abnormal score of each sample, and effectively improve the accuracy of automobile supply chain monitoring.

[0011] According to the automobile supply chain monitoring method based on the Internet of Things provided by the present invention, the method of obtaining the inventory quantity of various automobile parts, the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of each first-tier supplier also includes: obtaining automobile parts supply data within a preset time period, and pre-processing the automobile parts supply data to obtain the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of various automobile parts.

[0012] The present invention takes into account the large differences in characteristics of the inventory of various types of automobile parts, the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of each first-tier supplier. Direct data processing may cause the isolation forest algorithm to tend to features with higher data values. Therefore, preprocessing is used to reduce the impact of different features on the isolation forest algorithm, thereby preparing for subsequent data processing.

[0013] According to the automobile supply chain monitoring method based on the Internet of Things provided by the present invention, the obtaining of the importance of the component further includes: obtaining the order completion rate, the defective rate of each order and the total number of orders of the first-tier suppliers of various types of components;

[0014] ;

[0015] is the supply stability of the i-th type of parts, is the number of first-tier suppliers of the i-th type of parts, , For the i-th type of parts The order completion rate and total number of orders of first-tier suppliers, , They are respectively Tier 1 Supplier The defective rate and unit price of each order, is the average unit price of orders from all first-tier suppliers of the i-th type of parts, is the linear normalization function, is an absolute value symbol; determines the importance of parts and components, and the importance of parts and components is positively correlated with the supply stability of such parts and components.

[0016] The present invention takes into account that the product quality and price fluctuation of parts suppliers will affect the stability of the automobile supply chain. Therefore, when determining the importance of parts, the supply stability of parts can also be obtained, and the importance of parts can be accurately obtained by combining the supply stability and supply importance of parts.

[0017] According to the automobile supply chain monitoring method based on the Internet of Things provided by the present invention, the importance of the parts satisfies the relationship:

[0018] ;

[0019] is the importance of the i-th type of parts, is the supply importance of the i-th type of parts, is the supply stability of the i-th type of parts, is the mapping function.

[0020] The present invention maps the importance and value of each component to 1 by using a mapping function, so that the sampling probability of each component can be accurately obtained based on this later.

[0021] According to the automobile supply chain monitoring method based on the Internet of Things provided by the present invention, the sampling probability of each sample is used to construct an isolation forest in the isolation forest algorithm, including: presetting the number of samples for constructing an isolation tree, selecting samples as segmentation points based on the sampling probability of each sample to construct an isolation tree, and obtaining an isolation forest for the automobile supply chain.

[0022] According to the automobile supply chain monitoring method based on the Internet of Things provided by the present invention, the sampling probability of each sample is used in the isolation forest algorithm to construct an isolation forest to achieve automobile supply chain monitoring, including: obtaining the average path length of each sample in the isolation tree in the isolation forest of the automobile supply chain;

[0023] ;

[0024] is the abnormal score of the sample, is the average path length of the sample, is the average path length of all samples in the isolated tree; if the anomaly score of a sample is greater than the preset threshold, the monitoring result of the sample is abnormal; otherwise, the monitoring result of the sample is normal.

[0025] According to the automobile supply chain monitoring method based on the Internet of Things provided by the present invention, the automobile supply chain monitoring is implemented, and then further includes: in response to the monitoring result of the sample being abnormal, issuing a prompt externally.

[0026] The present invention takes into account that abnormality in samples of the automobile supply chain may affect the normal operation of the automobile supply chain, and therefore reminds the staff to handle it in a timely manner by sending out prompts.

[0027] In a second aspect, the present invention provides an automobile supply chain monitoring system based on the Internet of Things, which adopts the following technical solutions:

[0028] The automobile supply chain monitoring system based on the Internet of Things includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the automobile supply chain monitoring method based on the Internet of Things is implemented.

[0029] By adopting the above technical solution, the above-mentioned automobile supply chain monitoring method based on the Internet of Things is generated into a computer program and stored in a memory to be loaded and executed by a processor, so that a terminal device is made according to the memory and the processor for easy use.

[0030] The present invention has the following technical effects:

[0031] Based on the above technical solution, the automobile supply chain monitoring method and system based on the Internet of Things provided by the present invention can accurately obtain the risk characteristics of each automobile supply chain by processing the characteristics of various parts in the automobile supply chain through the isolation forest algorithm when obtaining the monitoring results of the automobile supply chain. In this process, the present invention takes into account that different categories of parts have different degrees of influence on the stability of the automobile supply chain, and errors may occur when dividing based on the isolation forest algorithm; based on this, the present invention obtains the importance of various parts by analyzing the suppliers and sales of various parts, and uses the importance of various parts to set the sampling probability when dividing based on the isolation forest algorithm, so that important parts can get more attention and more detailed division in the isolation tree division process, accurately obtain the abnormal score of each sample, and effectively improve the accuracy of automobile supply chain monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0033] Figure 1 A schematic diagram of a process flow in an automobile supply chain monitoring method based on the Internet of Things provided by an embodiment of the present invention;

[0034] Figure 2 A schematic diagram of an automotive parts supply chain structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0036] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0037] The automotive supply chain is a collection of a series of links and processes from raw material procurement, parts manufacturing, vehicle assembly, sales and after-sales service. Its role is to ensure efficient and smooth automobile production and sales to meet market demand. Through the real-time collection and transmission of relevant data from all links of the automotive supply chain through the Internet of Things, the security monitoring of the automotive supply chain can be achieved, and the collaborative efficiency and overall operation level of the supply chain can be improved.

[0038] The existing automotive supply chain monitoring methods focus on monitoring the production of parts manufacturers, but cannot fully understand the overall operation of the supply chain. Lack of transparency in the supply chain may result in companies being unable to accurately grasp key information such as inventory levels and parts status, thereby affecting the efficiency and response speed of the automotive supply chain.

[0039] As an unsupervised learning method, the isolation forest algorithm can isolate abnormal data points by constructing isolation trees by segmenting features without the need for labeled data. Therefore, it can be used for monitoring the automotive supply chain.

[0040] Based on this, an embodiment of the present invention discloses an automobile supply chain monitoring method based on the Internet of Things. The method processes the inventory and sales volume of each component in the automobile supply chain through an isolation forest algorithm, thereby accurately identifying abnormal risk data in the automobile supply chain.

[0041] For details, please see Figure 1 As shown, Figure 1 A schematic diagram of a flow chart of an automobile supply chain monitoring method based on the Internet of Things provided in an embodiment of the present invention, the method specifically includes the following steps.

[0042] S1: Obtain the inventory of various automotive parts, the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of each first-tier supplier.

[0043] It should be noted that when the sales and inventory of various auto parts remain within a relatively stable range, the state of the auto supply chain will be relatively stable. Abnormal sales and inventory reflect supply chain disruptions and improper inventory management, respectively. In this case, the auto supply chain will be at risk.

[0044] Based on this, the embodiment of the present invention processes the inventory and sales volume of each component in the automobile supply chain through the isolation forest algorithm, so as to identify the possible risks in the automobile supply chain.

[0045] By way of example, in an embodiment of the present invention, the inventory quantity of various types of automobile parts, the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of each first-tier supplier are obtained, which also includes: obtaining automobile parts supply data within a preset time period, and pre-processing the automobile parts supply data to obtain the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of various types of automobile parts.

[0046] Among them, the preset period can be set to 180 days. The preset period can be set according to actual needs, and the embodiment of the present invention does not make further restrictions here.

[0047] Specifically, the automobile parts supply data is collected within a preset period based on a preset collection frequency.

[0048] Among them, the preset collection frequency can be set to once a day, and the collection frequency can be set specifically according to actual needs.

[0049] If the collection frequency is once a day and the preset period is 180 days, 180 sets of automobile parts supply data can be collected.

[0050] It is understandable that when the sales and inventory characteristics of various automobile parts are quite different, the isolation forest algorithm may tend to favor features with larger values ​​during random segmentation, thus affecting the stability and generalization ability of the model. Therefore, in order to reduce the impact of excessive feature differences, different features need to be linearly normalized.

[0051] For example, the preprocessing method of automobile parts supply data can be linear normalization processing, missing data interpolation, etc., which can be specifically set according to actual needs, and the embodiment of the present invention does not impose too many restrictions here.

[0052] It should be noted that the automotive parts supply chain is a hierarchical structure, which can be found in Figure 2 , Figure 2 A schematic diagram of an automobile parts supply chain structure provided by an embodiment of the present invention. Supplier 1, Supplier 2 and Supplier 3 are first-tier suppliers of Part 1, Supplier 11 and Supplier 12 are second-tier suppliers under Supplier 1, Supplier 31 is a second-tier supplier under Supplier 3, and Supplier 121 and Supplier 122 are third-tier suppliers under Supplier 12.

[0053] In the automotive parts supply data, higher-cost parts have a greater impact on profits; the first-tier suppliers of parts are the suppliers that directly deliver parts. Therefore, when the number of first-tier suppliers is relatively single, the supply risk will be highly concentrated, and the impact on the supply chain will be higher when there is a risk interruption; precision parts have higher technical and production requirements, and usually require multiple levels of suppliers to collaborate to complete production and supply. In addition, precision parts are highly customized, the market demand is relatively small, and the total order volume will also be relatively small. Therefore, parts with a smaller total order volume and a higher degree of supplier singleness will suffer greater damage to the supply chain when risks arise.

[0054] It can be seen that when processing the sales and inventory of various automobile parts based on the isolation forest algorithm, different types of parts have different importance to the stability of the supply chain. The isolation forest algorithm is an anomaly detection method based on random segmentation, which does not consider the correlation and importance differences between samples. Therefore, if the isolation forest algorithm is used directly for segmentation, the importance differences of parts in the supply chain may be ignored, resulting in the accuracy and efficiency of risk identification being affected.

[0055] Based on this, the embodiment of the present invention can obtain the importance of each type of components, and set the corresponding random sampling probability for the data samples corresponding to this type of components according to the importance of each type of components, so that important components can receive more attention and more detailed division in the isolated tree division process, that is, execute the following steps.

[0056] S2: The ratio of the number of supplier levels of a type of parts to the total sales volume of all first-tier suppliers of this type of parts is recorded as the first feature of this type of parts; the reciprocal of the number of first-tier suppliers of this type of parts, the average unit price of orders from all first-tier suppliers and the sum of the first feature are normalized to obtain the supply importance of this type of parts.

[0057] It should be noted that the higher the cost of a component, the more single the supplier, the more supplier levels there are and the smaller the order volume, the higher the supply importance of the component. Based on this, the supply importance of each type of component can be obtained.

[0058] For example, in the embodiment of the present invention, the supply importance of a component is determined, and the following relationship can be specifically referred to:

[0059] ;

[0060] is the supply importance of the i-th type of parts, is the number of first-tier suppliers of the i-th type of parts, is the average unit price of orders from all first-tier suppliers of the i-th type of parts, is the number of supplier levels of the i-th type of parts, For the i-th type of parts Total orders from first-tier suppliers, For the i-th type of parts Tier 1 Supplier The number of parts per order, is a linear normalization function.

[0061] In the above formula, It indicates the degree of supplier singleness of Category I parts. The larger the value is, the fewer the number of first-tier suppliers of Category I parts is. When there is a risk in the supplier, the greater the impact on Category I parts is. Therefore, the supply importance of Category I parts is greater.

[0062] Among them, if the number of first-tier suppliers changes every day during the collection period, the average number of first-tier suppliers of the i-th type of parts during the collection period can be used as the number of first-tier suppliers of the i-th type of parts.

[0063] The higher the average unit price of orders from all first-tier suppliers of Category i components, the higher the profit of Category i components, and the greater the supply importance of Category i components.

[0064] represents the total sales volume of all first-tier suppliers of the i-th type of parts, It represents the first characteristic of the i-th type of parts. The larger the value is, the larger the number of supplier levels of the i-th type of parts is and the lower the total sales volume of all first-level suppliers is. The higher the possibility that the i-th type of parts is a precision part, the greater the supply importance of the corresponding i-th type of parts will be.

[0065] After obtaining the supply importance of each type of parts based on the above steps, the importance of each type of parts can be obtained based on the supply importance of each type of parts, that is, continue to perform the following steps.

[0066] S3: Convert the supply importance of the component into a relative probability distribution to obtain the importance of the component.

[0067] For example, in the embodiment of the present invention, the importance of a component is determined, and specifically, the following relationship can be referred to:

[0068] ;

[0069] is the importance of the i-th type of parts, is the supply importance of the i-th type of parts, is the mapping function.

[0070] In the above formula, the mapping function is used to convert the supply importance of the parts into a relative probability distribution, using The function maps the supply importance of various parts and components, and finally converts the importance of various parts and components into a probability value whose sum is 1.

[0071] It is understandable that the higher the supply importance of a component, the greater the impact on the automotive supply chain when the supply of the component is abnormal, and the corresponding importance of the component is also higher. When performing isolated tree segmentation, a higher random sampling probability needs to be set. Therefore, a mapping function is used to convert the supply importance of components. The final importance of each type of component is the random sampling probability and value of the sample of this type of component.

[0072] S4: The daily inventory of each type of parts and the sales volume of suppliers at all levels are taken as a sample of this type of parts, and the importance of the parts is evenly divided into the sampling probability of each sample in this type of parts.

[0073] Among them, the characteristics corresponding to the samples are the daily inventory of parts and the sales volume of suppliers at all levels.

[0074] For example, in an embodiment of the present invention, the daily inventory of each type of parts and the sales volume of suppliers at all levels are taken as a sample of that type of parts, and the importance of the parts is evenly divided into the sampling probability of each sample in that type of parts, including: obtaining the number of samples, and taking the ratio of the importance of a type of parts to the number of samples as the sampling probability of each sample in that type of parts.

[0075] For example, if the sampling frequency of Part 1 is once a day and the preset period is 180 days, 180 samples of Part 1 can be collected, and each sample corresponds to a set of inventory and sales. Then the sampling probability of each sample in Part 1 is ,in, is the importance of component 1.

[0076] After obtaining the sampling probability of each sample in each type of parts based on the above steps, the sampling probability of each sample can be used in the isolation forest algorithm to construct an isolation forest, thereby accurately identifying abnormal risk data in the automotive supply chain.

[0077] S5: Use the sampling probability of each sample in the isolation forest algorithm to construct an isolation forest to achieve automotive supply chain monitoring.

[0078] By way of example, in an embodiment of the present invention, the sampling probability of each sample is used in the isolation forest algorithm to construct an isolation forest, including: presetting the number of samples for constructing an isolation tree, selecting samples as split points based on the sampling probability of each sample to construct an isolation tree, and obtaining an isolation forest for the automobile supply chain.

[0079] Among them, the number of samples for constructing an isolated tree can be set to 300, and the number of isolated trees in the isolation forest can be set to 100; the number of samples for constructing an isolated tree and the number of isolated trees can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0080] Specifically, the sampling probability of each sample is used to construct an isolation forest in the isolation forest algorithm, including: initializing the maximum height of the tree, the number of samples for constructing an isolated tree, and the number of isolated trees in the isolation forest; obtaining the features corresponding to each sample, randomly selecting samples from all samples according to the sampling probability of each sample to form a training data set, and using each sample in the training data set as the root node of the isolation tree; for each root node, randomly selecting a feature to split the training data set into two parts to obtain a left child node and a right child node, and recursively performing the same operation on each child node until a preset stop condition is met; repeating the sampling process and the splitting process, and finally generating multiple isolated trees to form an isolation forest.

[0081] The preset stop condition may be set to the tree reaching a maximum height, which may be specifically set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0082] For example, in the embodiment of the present invention, the abnormal score of the sample is determined, and the specific relationship can be as follows:

[0083] ;

[0084] is the abnormal score of the sample, is the average path length of the sample, is the average path length of all samples in the isolated tree.

[0085] By way of example, in an embodiment of the present invention, the sampling probability of each sample is used in the isolation forest algorithm to construct an isolation forest to achieve automobile supply chain monitoring, including: obtaining the average path length of each sample in the isolated tree in the isolation forest of the automobile supply chain; calculating the anomaly score of each sample, if the anomaly score of the sample is greater than a preset threshold, the monitoring result of the sample is abnormal; otherwise, the monitoring result of the sample is normal.

[0086] The preset threshold may be set to 0.8; the preset threshold may be set specifically according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0087] It is understandable that if the monitoring results of samples in the automotive supply chain are abnormal, it means that there is a risk in the characteristics corresponding to the sample, which may affect the normal operation of the supply chain. Therefore, the abnormal monitoring results can be prompted so that the staff can deal with them in time.

[0088] For example, in an embodiment of the present invention, automobile supply chain monitoring is implemented, and then the method further includes: in response to the monitoring result of the sample being abnormal, issuing a prompt externally.

[0089] The method of issuing the prompt externally may be a sound prompt or a signal light prompt, which may be specifically set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0090] It can be seen that in the embodiment of the present invention, when determining the monitoring results of the automobile supply chain, the inventory of various types of automobile parts, the number of supplier levels, the number of first-level suppliers, and the order unit price and total sales volume of each first-level supplier can be obtained; the ratio of the number of supplier levels of a type of parts to the total sales volume of all first-level suppliers of this type of parts is recorded as the first feature of this type of parts; the reciprocal of the number of first-level suppliers of this type of parts, the average order unit price of all first-level suppliers and the sum of the first feature are normalized to obtain the supply importance of this type of parts; the supply importance of the parts is converted into a relative probability distribution to obtain the importance of the parts; the daily inventory of each type of parts and the sales volume of suppliers at all levels are taken as a sample of this type of parts, the importance of the parts is evenly divided into the sampling probability of each sample in this type of parts, and the sampling probability of each sample is used in the isolation forest algorithm to construct an isolation forest to achieve automobile supply chain monitoring.

[0091] In this way, the embodiment of the present invention processes the characteristics of various parts in the automobile supply chain through the isolation forest algorithm, and can accurately obtain the risk characteristics in each automobile supply chain. In this process, the embodiment of the present invention takes into account that different categories of parts have different degrees of influence on the stability of the automobile supply chain, and errors may occur when dividing based on the isolation forest algorithm; based on this, the embodiment of the present invention obtains the importance of various parts by analyzing the suppliers and sales of various parts, and uses the importance of various parts to set the sampling probability when dividing based on the isolation forest algorithm, so that important parts can get more attention and more detailed division in the isolation tree division process, accurately obtain the abnormal score of each sample, and effectively improve the accuracy of automobile supply chain monitoring.

[0092] Based on the above embodiment, in the above step S3, the supply importance of the component is converted into a relative probability distribution. When the importance of the component is obtained, the importance of the component can also be obtained through the supply importance and supply stability of the component. For details, please refer to the following steps.

[0093] It should be noted that when analyzing the importance of parts and components, in addition to the supply importance of parts and components, the supply stability of parts and components can also be considered. If the order completion rate of all parts and components suppliers is high, the defective rate is low and the price fluctuation is low, it means that the supplier stability of such parts and components is relatively high. The automotive supply chain relies on a sound and stable supply relationship, and the chain reaction between different links is relatively large. If there is a risk in the sound supply relationship, it will be difficult to remedy it in a short period of time, which will affect the supply chain.

[0094] Based on this, the embodiment of the present invention can also obtain the order completion rate, defective rate and price fluctuation degree of all suppliers of each component in the supply chain, and obtain their supply stability to identify whether the supply relationship of each component is sound.

[0095] By way of example, in an embodiment of the present invention, determining the importance of parts and components also includes: obtaining the order completion rate, defective rate of each order and total number of orders of first-tier suppliers of each type of parts and components; calculating the supply stability of each type of parts and components to determine the importance of the parts and components, and the importance of the parts and components is positively correlated with the supply stability of this type of parts and components.

[0096] For example, when obtaining the order completion rate of a first-tier supplier of parts, the ratio of the number of orders of the first-tier supplier of this type of parts to the number of orders of the first-tier suppliers of all types of parts can be recorded as the order completion rate of this type of parts.

[0097] For example, when obtaining the order defective rate of a first-tier supplier of parts, the ratio of the number of defective products in the order to the total number of parts in the order can be recorded as the order defective rate.

[0098] For example, in the embodiment of the present invention, the supply stability of parts is calculated, and the following relationship can be specifically referred to:

[0099] ;

[0100] is the supply stability of the i-th type of parts, is the number of first-tier suppliers of the i-th type of parts, For the i-th type of parts The order completion rate of first-tier suppliers, For the i-th type of parts Total orders from first-tier suppliers, For the i-th type of parts Tier 1 Supplier The defective rate of orders, For the i-th type of parts Tier 1 Supplier The unit price of an order, is the average unit price of orders from all first-tier suppliers of the i-th type of parts, is the linear normalization function, is the absolute value symbol.

[0101] Among them, get the i-th type of parts Tier 1 Supplier When calculating the unit price of an order, if the unit prices of the i-th type of parts in the order are different, the average unit price of the i-th type of parts in the order can be used as the unit price of the order.

[0102] In the above formula, Indicates the i-th type of parts Tier 1 Supplier The price fluctuation of the order. The larger the value, the The higher the price volatility of a first-tier supplier, the lower the corresponding stability will be.

[0103] Indicates the i-th type of parts The change index of the first-tier supplier is larger, indicating that the i-th type of parts is The higher the defective rate of the first-tier supplier and the higher the price volatility, the higher the The lower the reliability of a tier-one supplier, the higher the likelihood of variability.

[0104] The lower the volatility index of all first-tier suppliers of Category I components, the higher the stability of all first-tier suppliers of Category I components, the more sound the corresponding supply relationship of Category I components, and the higher the supply stability.

[0105] Based on the above steps, the supply stability of various parts can be obtained. If the supply importance of parts is higher and the supply stability is higher, it means that when risks occur in this type of parts, the impact on the automotive supply chain will be greater and it will be more difficult to restore the supply relationship in time. Therefore, the importance of the parts is also higher. Based on this, the importance of various parts can be accurately obtained.

[0106] For example, in the embodiment of the present invention, the importance of a component is determined, and specifically, the following relationship can be referred to:

[0107] ;

[0108] is the importance of the i-th type of parts, is the supply importance of the i-th type of parts, is the supply stability of the i-th type of parts, is the mapping function.

[0109] In the above formula, use The function maps the product of the supply importance and supply stability of each type of parts, and finally converts the importance of each type of parts into a probability value whose sum is 1.

[0110] After the importance of each component is obtained based on the above steps, the sampling probability can be set for each sample in the isolation forest algorithm, that is, steps S4 to S5 in the above embodiment are executed, so as to accurately obtain the automobile supply chain monitoring results.

[0111] In this way, the embodiment of the present invention can accurately obtain the supply stability of various parts by analyzing the order completion rate, defective rate and price fluctuation of all parts suppliers. When determining the importance of parts, the supply importance and supply stability of parts are comprehensively considered, and it can accurately determine whether the supply relationship of parts is sound, which effectively improves the accuracy of identifying abnormal data in the automotive supply chain.

[0112] An embodiment of the present invention also discloses an automobile supply chain monitoring system based on the Internet of Things, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an automobile supply chain monitoring method based on the Internet of Things provided by the present invention is implemented.

[0113] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0114] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM, a dynamic random access memory DRAM, a static random access memory SRAM, an enhanced dynamic random access memory EDRAM, a high bandwidth memory HBM, a hybrid memory cube HMC, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0115] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0116] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. The automobile supply chain monitoring method based on the Internet of Things is characterized by: include: Obtain the inventory of various automotive parts, the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of each first-tier supplier; The ratio of the number of supplier levels of a type of parts to the total sales volume of all first-tier suppliers of this type of parts is recorded as the first feature of this type of parts; the reciprocal of the number of first-tier suppliers of this type of parts, the average unit price of orders from all first-tier suppliers and the sum of the first feature are normalized to obtain the supply importance of this type of parts; The supply importance of parts is converted into relative probability distribution to obtain the importance of the parts. The daily inventory of each type of parts and the sales volume of suppliers at all levels are taken as a sample of this type of parts. The importance of parts is evenly divided into the sampling probability of each sample in this type of parts. The sampling probability of each sample is used in the isolation forest algorithm to construct an isolation forest to achieve automotive supply chain monitoring.

2. The automobile supply chain monitoring method based on the Internet of Things according to claim 1 is characterized in that: The method of obtaining the inventory of various automotive parts, the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of each first-tier supplier also includes: The automobile parts supply data within a preset time period is obtained, and after pre-processing the automobile parts supply data, the number of supplier levels, the number of first-tier suppliers, and the order unit price and total sales volume of various types of automobile parts are obtained.

3. The automobile supply chain monitoring method based on the Internet of Things according to claim 1 is characterized in that: The obtaining of the importance of the component further includes: Obtain the order completion rate, defective rate of each order and total number of orders of first-tier suppliers of various parts; ； is the supply stability of the i-th type of parts, is the number of first-tier suppliers of the i-th type of parts, , For the i-th type of parts The order completion rate and total number of orders of first-tier suppliers, , They are respectively Tier 1 Supplier The defective rate and unit price of each order, is the average unit price of orders from all first-tier suppliers of the i-th type of parts, is the linear normalization function, is an absolute value symbol; determines the importance of parts and components, and the importance of parts and components is positively correlated with the supply stability of such parts and components.

4. The method for monitoring automobile supply chain based on the Internet of Things according to claim 3 is characterized in that: The importance of the components satisfies the relationship: ； is the importance of the i-th type of parts, is the supply importance of the i-th type of parts, is the supply stability of the i-th type of parts, is the mapping function.

5. The automobile supply chain monitoring method based on the Internet of Things according to claim 1 is characterized in that: The method of constructing an isolation forest using the sampling probability of each sample in the isolation forest algorithm includes: The number of samples for building an isolation tree is preset, and samples are selected as split points based on the sampling probability of each sample to build an isolation tree, thus obtaining an isolation forest for the automotive supply chain.

6. The method for monitoring automobile supply chain based on the Internet of Things according to claim 5, characterized in that: The isolation forest algorithm uses the sampling probability of each sample to construct an isolation forest to achieve automobile supply chain monitoring, including: Obtain the average path length of each sample in the isolation tree in the isolation forest of the automotive supply chain; ； is the abnormal score of the sample, is the average path length of the sample, is the mean path length of all samples in the isolated tree; If the anomaly score of a sample is greater than the preset threshold, the monitoring result of the sample is abnormal; otherwise, the monitoring result of the sample is normal.

7. The method for monitoring automobile supply chain based on the Internet of Things according to claim 6, characterized in that: The implementation of automobile supply chain monitoring further includes: In response to abnormal monitoring results of the sample, a prompt is issued.

8. The automobile supply chain monitoring system based on the Internet of Things is characterized by: 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 automobile supply chain monitoring method based on the Internet of Things according to any one of claims 1 to 7 is implemented.

Images