An automobile parts management system based on big data

Abstract

The present application relates to the field of data processing technology, and specifically discloses an automobile parts management system based on big data, which mainly includes a defect detection module, a parts evaluation module and an inventory management module; the defect detection module is used to detect old parts to be recycled, and screen out non-recyclable old parts from the parts to be recycled, so as to obtain the remaining old parts as the parts to be evaluated; the parts evaluation module is used to obtain the repair information of the parts to be evaluated, and obtain the recycling status information of the parts to be evaluated based on the repair information. The present application detects and evaluates the degree of damage of the old parts to be recycled and ultimately guides the inventory management module to customize reasonable procurement and inventory strategies, thereby ensuring the stability of the supply chain, avoiding inventory shortages or surpluses, greatly reducing inventory backlogs and capital occupation, and improving procurement efficiency and cost control to achieve higher profits and efficiency, thereby improving corporate competitiveness.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an automobile parts management system based on big data. Background Art

[0002] The management system of auto parts mainly includes auto parts inventory management functions, such as recording the inventory quantity, location and status of auto parts, processing order requests and replenishment management. In addition, the system also provides inventory counting and management functions, supports regular inventory counting and generates simple reports and analyzes data, which can meet inventory management needs, help companies understand inventory status, manage ordering processes, and conduct data analysis.

[0003] At present, automobile parts management systems usually include statistics on purchased quantities and sales demand, but often ignore the maintenance status of old parts and their impact on recycling demand. As a result, the data of old parts that have been repaired are not fully utilized, and it is difficult to accurately predict the demand for recycling parts, which can easily cause inventory backlogs or parts shortages, resulting in waste of resources and low recycling efficiency. Summary of the invention

[0004] The present application provides an automobile parts management system based on big data, which solves the technical problem that the automobile parts management system in the prior art ignores the maintenance status of old parts and their impact on recycling demand, resulting in waste of resources and low recycling efficiency. It can detect and evaluate the degree of damage of old parts to be recycled and ultimately guide the inventory management module to customize reasonable procurement and inventory strategies, which can effectively avoid inventory shortages or surpluses, greatly reduce inventory backlogs and capital occupation, and improve procurement efficiency and cost control.

[0005] The present application provides an automobile parts management system based on big data, which includes a defect detection module, a parts evaluation module and an inventory management module; the defect detection module is used to detect old parts to be recycled, and screen out non-recyclable old parts from the parts to be recycled, so as to obtain the remaining old parts as parts to be evaluated; the parts evaluation module is used to obtain the repair information of the parts to be evaluated, and obtain the recycling status information of the parts to be evaluated based on the repair information, and the recycling status information includes a recyclable status and a non-recyclable status; the inventory management module is used to customize procurement and inventory strategies according to the recycling status information of all parts to be evaluated.

[0006] Inspect and evaluate the degree of damage of the old parts to be recycled, quickly and intelligently screen out non-recyclable parts, evaluate the old parts to be recycled to obtain their repair information, and use this to accurately determine the recycling status of the parts, thereby guiding the inventory management module to customize reasonable procurement and inventory strategies, ensure the stability of the supply chain, avoid inventory shortages or surpluses, greatly reduce inventory backlogs and capital occupation, improve procurement efficiency and cost control, so as to achieve higher profits and efficiency, thereby enhancing corporate competitiveness.

[0007] Furthermore, detecting the old parts to be recycled specifically includes: obtaining an appearance image of the part to be evaluated; inputting the appearance image into a trained deep learning model to obtain the degree of damage of the auto part; and making a judgment that the auto part is non-recyclable based on the degree of damage of the auto part.

[0008] Furthermore, obtaining the recycling status information of the accessory to be evaluated specifically includes: obtaining the predicted sales demand of the accessory to be evaluated in a future specified time period; and the purchased quantity P; obtain the data information of the recovered old parts of the same type as the parts to be evaluated, the data information includes the number of the same old parts that have been repaired F, the number of the same old parts that have not been repaired and the corresponding repair success rate ; Calculate the recycling demand r of the accessory to be evaluated, r is: , where SS represents the safety stock of the parts to be evaluated, represents the same type of old parts that have not been repaired yet. represents the repair success rate of the i-th unrepaired old parts of the same type, and m represents the total number of unrepaired old parts of the same type; determines whether the recycling demand r is greater than the preset demand threshold , if the recycling demand r is greater than the demand threshold , then the recycling status information of the accessory to be evaluated is determined to be a recyclable state, otherwise the recycling status information of the accessory to be evaluated is determined to be a non-recyclable state.

[0009] Furthermore, the maintenance information includes a predicted maintenance success rate. ; When determining the recycling status information of the accessory to be evaluated, determine whether the recycling demand r is greater than the preset demand threshold and determining the predicted maintenance success rate. Is it greater than the preset success rate threshold? , for: ; In the formula, k represents the amount of recovery demand r used to adjust the repair success rate If the recycling demand r is greater than the demand threshold And the predicted repair success rate Greater than the preset success rate threshold , then the recycling status information of the accessory to be evaluated is determined to be a recyclable state, otherwise the recycling status information of the accessory to be evaluated is determined to be a non-recyclable state.

[0010] Furthermore, the maintenance information also includes the predicted remaining service life of the component to be evaluated. If the remaining service life is less than the preset life threshold , it is determined that the recycling status information of the accessory to be evaluated is in a non-recyclable state.

[0011] Furthermore, the customized purchasing and inventory strategy includes the calculation of the reorder point ROP, which is: ; In the formula, t represents the average time interval between two batches of recycled old parts. Represents the average demand for recycled used parts within the time interval t between every two batches of recycled used parts.

[0012] Furthermore, the cost H of holding the old parts inventory per unit time t is obtained; the customized procurement and inventory strategy includes calculating the order quantity Q, which is: ; In the formula, Q represents the most economical order quantity of used parts in one order, and S represents the total cost required to recycle used parts, which includes the maintenance cost.

[0013] Further, ; In the formula, Represents the recovery cost of the i-th used part that has been repaired within unit time t.

[0014] In the second aspect, the present application provides an automobile parts management method based on big data, which is applied to an automobile parts management system based on big data in the first aspect, and includes: detecting old parts to be recycled, and screening out non-recyclable old parts from the parts to be recycled, so as to obtain the remaining old parts as parts to be evaluated; obtaining the repair information of the parts to be evaluated, and obtaining the recycling status information of the parts to be evaluated based on the repair information, and the recycling status information includes a recyclable status and a non-recyclable status; customizing procurement and inventory strategies according to all the parts to be evaluated in a recyclable state and the inventory of the parts to be evaluated obtained in advance.

[0015] In a third aspect, the present application provides an automobile parts management device based on big data, which includes: a memory and a processor, the memory is used to store a computer program, and the processor is used to implement the steps of the automobile parts management method based on big data of the second aspect when executing the computer program.

[0016] The technical solution provided by this application has at least the following technical effects or advantages:

[0017] This application detects and evaluates the degree of damage of old parts to be recycled, quickly and intelligently screens out non-recyclable parts, evaluates the old parts to be recycled and finally guides the inventory management module to customize reasonable procurement and inventory strategies, effectively solving the problem that the current automotive parts management system ignores the repair status of old parts and its impact on recycling needs, thereby failing to fully utilize the data of old parts that have been repaired, easily causing inventory backlogs or parts shortages, leading to waste of resources and low recycling efficiency. It ensures the stability of the supply chain, avoids inventory shortages or surpluses, greatly reduces inventory backlogs and capital occupation, improves procurement efficiency and cost control, and achieves higher profits and efficiency, thereby improving corporate competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a module diagram of the automobile parts management system based on big data in this application;

[0019] Figure 2 Flowchart of purchasing and inventory strategy customized for this application. DETAILED DESCRIPTION

[0020] This application solves the defect that the current automobile parts management system ignores the maintenance status of old parts and their impact on recycling needs, resulting in waste of resources and low recycling efficiency. It can detect and evaluate the degree of damage of old parts to be recycled and ultimately guide the inventory management module to customize reasonable procurement and inventory strategies to avoid inventory shortages or surpluses, greatly reduce inventory backlogs and capital occupation, improve procurement efficiency and cost control, so as to achieve higher profits and efficiency, thereby enhancing corporate competitiveness.

[0021] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0022] Embodiment 1

[0023] like Figure 1-Figure 2 As shown, the present application provides an automobile parts management system based on big data, which includes a defect detection module, a parts evaluation module and an inventory management module.

[0024] The defect detection module is used to detect the old parts to be recycled and screen out the non-recyclable old parts from the parts to be recycled, so as to obtain the remaining old parts as the parts to be evaluated.

[0025] The main function of the module is to detect and screen the old parts to be recycled. Before the detection and screening, it is necessary to collect information on the old parts to be recycled, including the model, production date, appearance condition and other information, and establish an accessories information database. After that, the old parts to be recycled can be tested by using automated equipment to identify the parts that cannot be repaired due to serious defects and damage. Since the determined non-recyclable parts are screened out, the number of a batch of parts to be recycled is reduced, and the number of old parts entering the evaluation and processing process will be reduced, thereby improving the recycling efficiency and quality.

[0026] In this embodiment, detecting the old accessories to be recycled specifically includes the following steps:

[0027] S1. Obtain an appearance image of the accessory to be evaluated.

[0028] Use video equipment or a camera to take photos or videos of the old accessories to be recycled to obtain images of the accessories' appearance. Ensure that the images are clear, at an appropriate angle, and cover all parts and details of the accessories.

[0029] S2. Input the appearance image into the trained deep learning model to obtain the damage degree of the auto parts.

[0030] Establish or select a trained deep learning model that can identify and evaluate the degree of damage of automotive parts. Take the appearance image of the part to be evaluated as input data and process it through the model's image recognition algorithm. The deep learning model will analyze and process the input part image, identify the damage of the part, and output a corresponding damage degree score. The damage degree score can be a numerical value that represents the degree or percentage of the damage degree of the part.

[0031] S3. Determine whether auto parts are non-recyclable based on the degree of damage.

[0032] A threshold or standard is set to determine whether the degree of damage of the accessory reaches the level of non-recyclability. If the damage degree score exceeds the set threshold, the accessory is judged to be non-recyclable; otherwise, the accessory can continue to be recycled or further evaluated.

[0033] When training a deep learning model, first collect a large amount of image data of auto parts, including images of damaged and intact parts, annotate the image data, and mark the degree of damage of each part image. You can use bounding box annotation, semantic segmentation annotation, etc., and then preprocess the image, such as size standardization, color space conversion, image enhancement, etc., to improve the training effect of the model. Choose a suitable deep learning model architecture, such as convolutional neural network, for image recognition and damage assessment. Depending on the specific situation, you can choose an existing classic model structure, such as VGG, ResNet, or design a model structure yourself, use the annotated auto parts image data set, and train the deep learning model to learn the recognition of damage. Deploy the trained deep learning model to the actual system, and you can evaluate the damage degree of the old parts to be recycled and judge whether they are not recyclable.

[0034] During the training process, the model parameters are updated through the back-propagation algorithm so that the model can gradually understand and learn the characteristics of the degree of damage. The trained model is evaluated using the validation set to evaluate the performance of the model on new data. The model parameters are tuned according to the evaluation results, such as learning rate adjustment and regularization, to improve the generalization ability of the model.

[0035] The deep learning model can automatically detect and evaluate the degree of damage of old parts to be recycled, improving the accuracy and efficiency of detection, thereby quickly and intelligently screening out non-recyclable parts, ensuring the quality and availability of parts for subsequent evaluation and inventory management, thereby improving the work efficiency of the overall recycling process and reducing labor costs.

[0036] The accessory evaluation module is used to obtain the repair information of the accessory to be evaluated, and obtain the recycling status information of the accessory to be evaluated based on the repair information, and the recycling status information includes a recyclable state and a non-recyclable state.

[0037] The information to be repaired may include the appearance, damage degree, historical repair record, service life, etc. of the parts. This information can be combined with the information about the parts obtained through manual inspection, damage degree, etc. identified by deep learning models, or other methods. The recyclable status means that the parts to be evaluated will be recycled, and the non-recyclable status means that the parts to be evaluated cannot be recycled this time due to some reasons.

[0038] It should be noted that accessories that can be restored to a usable state through repair and restoration, or that can be reused as second-hand accessories, may also be non-recyclable due to some reasons.

[0039] After determining the recycling status of the parts to be evaluated, the information can be used for subsequent processing and decision-making. Parts in a recyclable state can be sent to the repair center for repair and then put into storage after the repair is completed. By judging the recycling status of the parts to be evaluated, the rational use and management of the parts can be achieved.

[0040] Obtaining the recycling status information of the accessories to be evaluated specifically includes the following steps:

[0041] S21. Obtain the predicted sales demand of the accessories to be evaluated in the specified future time period and the purchased quantity P; obtain the data information of the recovered old parts of the same type as the parts to be evaluated, including the number of the same old parts that have been repaired F, the number of the same old parts that have not been repaired, and the corresponding repair success rate .

[0042] The sales demand for the accessory to be evaluated in the future can be estimated based on market trends, historical sales data and other factors. The purchased quantity P refers to the number of parts to be evaluated that have been purchased or ordered in the current time period. The recycled old parts data information includes the relevant information of the old parts of the same type as the parts to be evaluated that have been recycled and evaluated. Among them, the number of old parts of the same type that have been repaired F refers to the number of parts that have been repaired but not yet put into storage among the recycled old parts of the same type. The number of old parts of the same type that have not been repaired refers to the number of parts of the same type that have not been repaired among the recycled old parts of the same type. The repair success rate The repair success rate refers to the probability that the parts that have not been repaired can be successfully repaired and put into use again. It can be obtained by analyzing historical data or manually inferring it, such as collecting historical repair data of the same old parts to infer the probability of successful repair of the part to be evaluated.

[0043] S22. Calculate the recycling demand r of the accessories to be evaluated, , where SS represents the safety stock of the parts to be evaluated, represents the i-th used part of the same type that has not been repaired yet, represents the repair success rate of the i-th used part of the same type that has not been repaired, and m represents the total number of used parts of the same type that have not been repaired.

[0044] Safety stock is the additional inventory reserved to cope with sudden demand fluctuations, supply chain delays or other risks. The setting of safety stock can take into account factors such as supply chain reliability and delivery cycle. By introducing multiple factors such as sales demand, safety stock, purchased quantity, maintenance progress, etc., the calculation of parts recycling demand is more accurate, thereby reducing inventory errors, helping to reduce waste or excess caused by inaccurate parts estimation, and improving the utilization rate of inventory parts.

[0045] S23. Determine whether the recycling demand r is greater than the preset demand threshold , if the recycling demand r is greater than the demand threshold , then the recycling status information of the accessory to be evaluated is determined to be a recyclable state, otherwise the recycling status information of the accessory to be evaluated is determined to be a non-recyclable state.

[0046] After the recycling demand r is calculated, the recycling demand r and the pre-set demand threshold Compare, if the recycling demand r is greater than the preset demand threshold , then the recycling status information of the to-be-evaluated accessory is determined to be recyclable, indicating that the market demand for the to-be-evaluated accessory is large and it can be recycled and put into storage. If the recycling demand r is less than or equal to the preset demand threshold , then the recycling status information of the accessory to be evaluated is determined to be non-recyclable, which may indicate that the market demand for the accessory to be evaluated is small, and it may not be worth recycling or there is already a sufficient amount of inventory.

[0047] The preset threshold can be set according to the actual situation and business needs. The threshold can usually be determined based on market demand, inventory level, cost-effectiveness and other factors. The setting of the threshold needs to take into account the balance between ensuring inventory turnover and meeting market demand in order to make a reasonable recycling decision. In one solution of this embodiment, the demand threshold It can be 0. If the calculated recycling demand is greater than 0, it means that the accessory to be evaluated still needs to be recycled, and the recycling status information of the accessory to be evaluated is judged to be recyclable. If the calculated recycling demand is a negative number, it means that the accessory to be evaluated already has sufficient inventory and no longer needs to be recycled, so the recycling status information of the accessory to be evaluated is judged to be non-recyclable.

[0048] Based on the maintenance information and sales demand forecast of accessories, the recycling status of accessories can be accurately determined, and then the inventory management module can be guided to customize reasonable procurement and inventory strategies, which greatly reduces inventory backlogs and capital occupation.

[0049] Repair information including predicted repair success rate ; When determining the recycling status information of the accessory to be evaluated, determine whether the recycling demand r is greater than the preset demand threshold While judging the predicted maintenance success rate Is it greater than the preset success rate threshold? .

[0050] When determining the recycling status information of the accessories to be evaluated, in addition to comparing the recycling demand r with the demand threshold In addition, it is necessary to compare the predicted repair success rate The preset success rate threshold If the predicted repair success rate Greater than the preset success rate threshold , it means that the probability of successful repair of the evaluated parts is high, which is conducive to reducing repair costs and thus obtaining sales revenue. These two factors can more comprehensively evaluate the recycling status of the accessories to be evaluated, so as to formulate reasonable maintenance and inventory management strategies and reduce inventory costs and risks.

[0051] for: ; In the formula, k represents the amount of recovery demand r used to adjust the repair success rate A constant that determines the degree of influence.

[0052] This formula describes the repair success rate required to recycle the evaluated parts as the recycling demand r of the evaluated parts decreases. Will gradually increase, that is, the repair success rate It is negatively correlated with the recycling demand r. As the recycling demand r of the parts to be evaluated decreases, the required repair success rate of the parts to be evaluated can be considered. , gradually improve the success rate of maintenance The requirements are met. Since the probability of successful repair of the evaluated parts is high, it is helpful to reduce the repair cost and thus obtain sales revenue.

[0053] If the recycling demand r is greater than the demand threshold And the predicted repair success rate Greater than the preset success rate threshold , then the recycling status information of the accessory to be evaluated is determined to be a recyclable state, otherwise the recycling status information of the accessory to be evaluated is determined to be a non-recyclable state.

[0054] The judgment condition for the recycling status information of the accessories to be evaluated is: the recycling demand r is greater than the demand threshold And the predicted maintenance success rate is greater than the success rate threshold If these two conditions are met at the same time, the recycling status information of the accessory to be evaluated is judged to be recyclable, otherwise it is judged to be non-recyclable. By setting different recycling status judgment conditions, the recycling potential and market demand of accessories can be more accurately evaluated, thereby improving recycling efficiency. If the recycling demand r is greater than the demand threshold And the predicted repair success rate Above the success rate threshold , indicating that the market demand for the evaluated accessory is large, so the recycling quantity should be considered to meet the market demand. If the recycling demand r meets the demand threshold However, the predicted repair success rate Not higher than the success rate threshold , it indicates that the market demand for the evaluated parts is small, and the repair success rate should be considered. To reduce maintenance costs and try to avoid economic losses caused by maintenance damage.

[0055] By considering the recycling demand and the repair success rate, we can better respond to the market demand for accessories. When the market demand for accessories is large, we focus on the recycling quantity to meet the market demand. When the market demand is small, we focus on improving the repair success rate to reduce the repair cost. The recycling status of accessories can be judged according to the recycling demand and the repair success rate in a targeted manner, which helps to optimize resource allocation and management. According to the market demand and repair cost, we can reasonably arrange the recycling and repair plan of accessories, more accurately evaluate the recycling potential and market demand of accessories, improve the recycling efficiency, and effectively avoid the additional costs and losses caused by repair failures.

[0056] In the technical solution of this embodiment, the information to be repaired also includes the predicted remaining service life of the component to be evaluated. If the remaining service life is less than the preset life threshold , then the recycling status information of the accessory to be evaluated is determined to be non-recyclable.

[0057] By analyzing and predicting the historical data, usage, maintenance records, etc. of the accessories, the remaining service life of the accessories can be estimated and a preset life threshold can be set. , when the remaining service life of the accessory is less than this life threshold When the recycling demand exceeds the demand threshold, it is judged as unrecyclable. , the repair success rate is higher than the success rate threshold However, if the remaining service life of the accessory is short, the accessory will still be judged as non-recyclable.

[0058] The remaining service life of accessories can be predicted through machine learning-based methods. The historical data of accessories, including usage, maintenance records, working environment and other related information, can come from sensors, maintenance records, operation logs, etc., and then the collected data is cleaned, processed and feature extracted, including data cleaning, missing value processing, feature selection and other steps to provide effective data for machine learning models. After selecting a suitable model, the selected model is trained using historical data, such as machine learning models, regression models, etc. The trained model is used to predict the remaining service life of the automotive accessories to be evaluated. Based on the results of the model output, the remaining service life of the accessories can be estimated.

[0059] Taking the remaining service life of accessories into consideration can avoid recycling and repairing accessories that are about to be scrapped, thereby avoiding waste of resources, avoiding damage or failure after repair due to the short life of accessories, reducing maintenance costs and risks, and avoiding recycling and repair of accessories that cannot continue to be used for a long time. This helps to improve maintenance efficiency and reduce unnecessary maintenance workload. At the same time, it can save recycling and maintenance costs and improve overall cost-effectiveness.

[0060] The inventory management module is used to customize procurement and inventory strategies based on the recycling status information of all accessories to be evaluated.

[0061] Based on the demand, forecasted sales and inventory status of the recyclable accessories to be evaluated, reasonable procurement and inventory strategies can be formulated to achieve optimal management and utilization of inventory, better respond to market demand, avoid inventory backlogs and shortages, and optimize the utilization of inventory resources. It can also reduce capital occupation costs and improve the recycling efficiency of recyclable accessories to achieve higher profits and efficiency, thereby enhancing corporate competitiveness.

[0062] Customized purchasing and inventory strategies include calculation of the reorder point (ROP), which is: ; In the formula, t represents the average time interval between two batches of recycled old parts. Represents the average demand for recycled used parts within the time interval t between every two batches of recycled used parts.

[0063] By calculating the order point ROP, we can calculate the demand for recycled old parts within the time interval t. And the safety stock SS is used to accurately measure the inventory level. When the inventory drops below the order point ROP, it is a signal to reorder, which helps to avoid inventory shortages and excessive inventory reserves, ensure timely replenishment and supply, and improve the efficiency and responsiveness of the supply chain. By considering the safety stock, you can reduce inventory risks, respond to emergencies or demand fluctuations, and ensure the stability of the supply chain.

[0064] Obtain the cost H of holding used parts inventory per unit time t; the customized procurement and inventory strategy includes calculating the order quantity Q, Q is: ; In the formula, Q represents the most economical order quantity of used parts in one order, and S represents the total cost required to recycle used parts, which includes the maintenance cost.

[0065] ; In the formula, represents the recovery cost of the i-th used part that has been repaired within unit time t, and n represents the number of used parts that have been repaired within unit time t.

[0066] By calculating the order quantity Q by considering the recovery cost of all used parts that have been repaired within unit time t, the accuracy of the order quantity can be improved, and the ordering cost and holding cost can be reduced by avoiding excessive or insufficient ordering, achieving the optimal order quantity, and thus improving profitability.

[0067] Embodiment 2

[0068] The present application provides an automobile parts management method based on big data, which is applied to an automobile parts management system based on big data in Example 1, which includes: detecting old parts to be recycled, and screening out non-recyclable old parts from the parts to be recycled, so as to obtain the remaining old parts as the parts to be evaluated; obtaining the repair information of the parts to be evaluated, and obtaining the recycling status information of the parts to be evaluated based on the repair information, and the recycling status information includes a recyclable state and a non-recyclable state; customizing the procurement and inventory strategy according to all the parts to be evaluated in a recyclable state and the inventory of the parts to be evaluated in advance. The automobile parts management method based on big data has all the advantages of an automobile parts management system based on big data in Example 1, so it will not be repeated.

[0069] Embodiment 3

[0070] The present application provides an automobile parts management device based on big data, which includes: a memory and a processor, the memory is used to store a computer program, and the processor is used to implement the steps of the automobile parts management method based on big data in Example 2 when executing the computer program.

[0071] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0072] Computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide the instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in a box or multiple boxes. Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

Claims

1. An automobile parts management system based on big data, characterized in that: It includes: Defect detection module: It is used to detect the old parts to be recycled and screen out the non-recyclable old parts from the parts to be recycled, so as to obtain the remaining old parts as the parts to be evaluated; Parts evaluation module: used to obtain the repair information of the parts to be evaluated, and obtain the recycling status information of the parts to be evaluated based on the repair information, and the recycling status information includes a recyclable state and a non-recyclable state; Inventory management module: It is used to customize procurement and inventory strategies based on the recycling status information of all accessories to be evaluated; Wherein, obtaining the recycling status information of the accessory to be evaluated specifically includes: Obtain the predicted sales demand of the accessory to be evaluated within a specified future time period and the purchased quantity P; obtain the data information of the recovered old parts of the same type as the parts to be evaluated, the data information includes the number of the same old parts that have been repaired F, the number of the same old parts that have not been repaired and the corresponding repair success rate ; Calculate the recycling demand r of the accessory to be evaluated, where r is: In the formula, SS represents the safety stock of the parts to be evaluated. represents the same type of old parts that have not been repaired yet. represents the repair success rate of the i-th used part of the same type that has not been repaired, and m represents the total number of used parts of the same type that have not been repaired; Determine whether the recycling demand r is greater than the preset demand threshold , if the recycling demand r is greater than the demand threshold , then the recycling status information of the accessory to be evaluated is determined to be a recyclable state, otherwise the recycling status information of the accessory to be evaluated is determined to be a non-recyclable state.

2. The automobile parts management system based on big data as claimed in claim 1, characterized in that: The old parts to be recycled include: Acquire an appearance image of the accessory to be evaluated; Input the appearance image into the trained deep learning model to obtain the damage degree of the auto parts; The auto parts are judged as non-recyclable based on the degree of damage.

3. The automobile parts management system based on big data as claimed in claim 1, characterized in that: The maintenance information includes the predicted maintenance success rate. ; When determining the recycling status information of the accessory to be evaluated, determine whether the recycling demand r is greater than the preset demand threshold and determining the predicted maintenance success rate. Is it greater than the preset success rate threshold? , for: In the formula, k represents the amount of recovery demand r used to adjust the maintenance success rate A constant indicating the degree of influence of If the recycling demand r is greater than the demand threshold And the predicted repair success rate Greater than the preset success rate threshold , then the recycling status information of the accessory to be evaluated is determined to be a recyclable state, otherwise the recycling status information of the accessory to be evaluated is determined to be a non-recyclable state.

4. The automobile parts management system based on big data as claimed in claim 1, characterized in that: The maintenance information also includes the predicted remaining service life of the component to be evaluated. If the remaining service life is less than the preset life threshold , it is determined that the recycling status information of the accessory to be evaluated is in a non-recyclable state.

5. The automobile parts management system based on big data as claimed in claim 1, characterized in that: Customized purchasing and inventory strategies include calculation of the reorder point (ROP), which is: In the formula, t represents the average time interval between two batches of recycled old parts. Represents the average demand for recycled used parts within the time interval t between every two batches of recycled used parts.

6. The automobile parts management system based on big data as claimed in claim 5, characterized in that: Obtain the cost H of holding used parts inventory per unit time t; the customized procurement and inventory strategy includes calculating the order quantity Q, Q is: In the formula, Q represents the most economical order quantity of used parts in one order, and S represents the total cost required to recycle used parts, which includes the maintenance cost.

7. The automobile parts management system based on big data as claimed in claim 6, characterized in that: In the formula, represents the recovery cost of the i-th used part that has been repaired within unit time t, and n represents the number of used parts that have been repaired within unit time t.

Images