Spare part putting method and system driven by spare part portrait

Abstract

The invention provides a spare part portrait-driven spare part delivery method and system, and the method comprises the following steps: obtaining a spare part index of a delivery project through a data integration service module, the spare part index comprising basic data and consumption behavior data; portraying the spare parts through a portraying integration module, and quantifying scores of the spare part indexes to obtain importance labels of the spare parts; and according to the spare part indexes and the spare part importance labels, determining the putting quantity of spare parts so as to obtain a spare part putting list sliced according to time. The spare parts can be evaluated more objectively and dynamically in real time, and hysteresis and deviation caused by expert experience are reduced, so that the accuracy of putting the spare parts is enhanced, and the economic loss risk caused by deviation is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of spare parts management for construction machinery, and in particular to a spare parts delivery method and system driven by spare parts portraits. Background Art

[0002] Demand management of construction machinery spare parts is an important aspect of improving aftermarket service capabilities, especially in the field of large-scale high-end construction machinery. With the characteristics of high value of complete equipment, high reliability requirements, high value of spare parts, large inventory and capital occupation, it is particularly critical to determine the early release demand list of spare parts for the entire life cycle, especially for new projects and new models of complete equipment. In the prior art, the key prediction algorithm of spare parts is generally studied through massive historical data, and the spare parts preparation list is realized through dynamic inventory adjustment. For example, the Chinese invention patent with publication number CN111798069A discloses a method and system for managing spare parts for complex equipment based on machine learning. According to the experience of experts in the field, a set of screening criteria for key spare parts is defined, and then spare parts prediction is performed using machine learning. This traditional experience weighting method cannot maintain an objective evaluation of materials, especially the lag and deviation of expert experience in the changing market and new models, lacks flexibility in operating complete equipment, and has high cost of continuous maintenance. At the same time, directly taking the predicted demand value as the final result will reduce customer experience and even cause certain economic losses due to the deviation of certain key components. Summary of the invention

[0003] In order to solve the technical problems existing in the prior art, the first aspect of the present application provides a spare parts delivery method driven by spare parts portrait, the method comprising the following steps: Step S1: obtaining spare parts indicators of a project through a data integration service module, wherein the spare parts indicators include basic data and consumption behavior data; Step S2: Profiling the spare parts through the portrait integration module, quantifying the scores of the spare parts indicators to obtain the importance labels of the spare parts; Step S3: Determine the number of spare parts to be delivered according to the spare parts index and the spare parts importance label obtained in step S2, so as to obtain a spare parts delivery list sliced ​​by time.

[0004] Furthermore, the basic data includes the number of complete equipment of the target vehicle model, the number of installed spare parts of the complete equipment and the number of R&D suggestions.

[0005] Furthermore, the consumption behavior data includes historical replacement records and replacement rates of spare parts.

[0006] Furthermore, the step S1 also includes determining the loss probability of the spare parts according to the consumption behavior data.

[0007] Furthermore, the step S2 includes detecting the integrity of the spare parts indicators; performing interval processing on the spare parts indicators to quantify the scores of each spare parts indicator; obtaining a spare parts comprehensive score by weighted summing up the scores of different spare parts indicators; and obtaining a spare parts importance label by dividing the importance of the spare parts according to the spare parts comprehensive score.

[0008] Furthermore, the step S3 comprises: When the missing rate of spare parts importance labels is less than the preset threshold and the number of complete equipment is greater than the preset threshold, the number of spare parts delivered is determined by the number of complete equipment, the number of installed spare parts for the complete equipment and the probability of loss; When the missing rate of spare parts importance labels is less than the preset threshold and the number of complete equipment is less than or equal to the preset threshold, the number of spare parts delivered is determined by the number of complete equipment, the number of installed spare parts for the complete equipment and the replacement rate; When the missing rate of spare parts importance labels is greater than its preset threshold and the number of complete equipment is less than or equal to its preset threshold, the number of spare parts to be put into use is determined by the number of R&D recommendations and the number of complete equipment.

[0009] Furthermore, the step S3 also includes: assigning a spare part business weight coefficient according to the spare part importance label obtained in step 2, and determining a final spare part delivery quantity according to the business weight coefficient and the spare parts delivery quantity.

[0010] Furthermore, the consumption behavior data also includes a minimum delivery cycle for spare parts, and the warranty period of the entire equipment is evenly divided into delivery time slices through the minimum delivery cycle, and the spare parts delivery lists for different time slices are determined through the spare parts indicators under the time slices.

[0011] The second aspect of the present application provides a spare parts delivery system driven by spare parts portrait, including: Data integration service module, used to obtain spare parts indicators according to the preset collection frequency; A portrait integration module, used for obtaining a spare part importance label by quantifying the score of the spare part indicator; An intelligent analysis module, used to calculate the number of spare parts delivered according to the spare parts index and the spare parts importance label, and determine the spare parts delivery list for different time slices; The report export module is used to output visual spare parts delivery reports.

[0012] Furthermore, the data integration service module includes a customer relationship management system, an enterprise resource planning system and a spare parts management system which are interconnected.

[0013] Compared with the prior art, the beneficial effects of this application are: 1. The spare parts delivery method driven by spare parts portrait and the portrait integration module in the system of the present application, by processing spare parts indicators into intervals, quantifies the scores of various spare parts indicators. It is possible to evaluate spare parts more objectively and in real time, reduce the lag and deviation caused by expert experience, thereby enhancing the accuracy of spare parts delivery and reducing the risk of economic losses caused by deviations.

[0014] 2. The spare parts delivery method and system driven by spare parts portrait of the present application divides the warranty period of the whole equipment evenly into delivery time slices through the minimum delivery cycle, and determines the spare parts delivery list of different time slices through the spare parts indicators under the time slices. In this way, the number of periodic replacement parts can be estimated and preparations for spare parts delivery can be made. It can improve spare parts turnover and reduce the pressure of spare parts inventory at the beginning of the period. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of the spare parts delivery method driven by spare parts portrait of the present application; Figure 2 It is a flow chart of the spare parts delivery method driven by spare parts portrait of the present application. DETAILED DESCRIPTION

[0016] In order to facilitate the understanding of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0017] The present application provides a spare parts delivery system driven by spare parts portrait, including: The data integration service module includes the interconnected customer relationship management system (CRM), enterprise resource planning system (ERP) and spare parts management system (PMS), which are used to obtain the spare parts indicators of the whole equipment related to research and development, production and use according to the preset collection frequency; organize the collected data through the whole equipment code, spare parts code and timestamp, remind the operator to process the missing data, process the multi-source heterogeneous data and explore the data quality to ensure the timeliness and accuracy of the data.

[0018] The portrait integration module is used to measure and integrate the basic data collected by the data integration service module, and calculate and store the actual consumption behavior data generated by the whole equipment. By calculating and classifying the spare parts indicators, quantifying and scoring the spare parts indicators, and integrating the content of the spare parts indicators with the comprehensive scoring of subjective and objective factors, the final portraits of different spare parts in the same model of whole equipment are obtained, and the spare parts importance labels and the integrity of the spare parts importance labels are obtained.

[0019] The intelligent analysis module uses a spare parts delivery strategy that combines expert experience with machine learning algorithms to calculate the number of spare parts delivered by time slices based on the needs of the delivery project and the spare parts importance labels and integrity of spare parts obtained from spare parts portraits. At the same time, it records the spare parts business weight coefficients and algorithm model parameters given by expert experience for continuous updates and iterations.

[0020] Report export module: visualize the delivery plan and output spare parts delivery reports sliced ​​by time.

[0021] refer to Figure 1 As shown, the present application also provides a spare parts delivery method driven by spare parts portrait, including: Step S1, obtain the spare parts index of the launch project through the data integration service module, and the spare parts index includes basic data and consumption behavior data. Specifically, through the whole equipment code, spare parts code and timestamp, the data in the customer relationship management system (CRM), enterprise resource planning system (ERP) and spare parts management system (PMS) are integrated and processed to form a cross-system multi-data source record for the target spare parts in the same model of whole equipment. The basic data and consumption behavior data are obtained in real time from the data integration service module according to the preset collection frequency, wherein the basic data includes: the number of whole equipment of the launch project, the number of spare parts installed for the whole equipment, and the number of R&D recommendations.

[0022] In some embodiments, the information table of the launch project is obtained from the demand department to obtain information about the launch project, including the product model, complete equipment code, spare parts code and timestamp of the launch area. The spare parts operation data such as the market price, procurement cost, and current inventory of spare parts are obtained from the spare parts management system (PMS) through an interface. The spare parts information of the product model in the equipment R&D and design process is obtained from the enterprise resource planning system (ERP) in the form of a report, including the system to which the spare parts belong, the type of spare parts, the number of spare parts installed for the complete equipment (that is, the number of spare parts assembled / single unit), the design life of the spare parts, and the number of spare parts R&D recommendations. The consumption behavior data of the spare parts corresponding to the complete equipment of the product model is obtained from the customer relationship management system (CRM) through an interface. Then, the data quality of the data obtained from each system is evaluated, and duplicate values ​​are eliminated field by field. By calculating the proportion of vacant values ​​in each field to all fields, if the vacancy ratio exceeds a certain threshold, the data source needs to be filled. Optionally, the cleaned data is provided in the form of a text file or a Restful API (system interface).

[0023] When the sample size of the newly launched complete equipment is insufficient, the probability distribution form is fitted according to the consumption behavior data of spare parts, and the distribution form is tested by the Ko1mogorov-Smirnov test method. The test distribution includes uniform distribution, normal distribution, half-normal distribution and Laplace distribution. The consumption behavior data of spare parts calculated by the parameters of the test distribution include the first replacement time t1 of the target spare parts on the target model, the average first replacement time t2 on different complete equipment of the target model, the average interval replacement time t3 on the target model, the replacement rate k1, the proportion of complete equipment replaced k2 and the minimum release cycle T1. Among them, the replacement time is counted according to the GPS IoT working hours of the corresponding complete equipment. In addition, k1=n1 / n2, where k1 is the replacement rate, n1 is the number of replacement parts, and n2 is the number of complete equipment. k2=n3 / n2, where k2 is the proportion of complete equipment replaced, n3 is the number of complete equipment replaced, and n2 is the number of complete equipment.

[0024] Furthermore, the minimum delivery cycle T1 is determined according to the location of the delivery project and the logistics time, and the spare parts delivery time slice is obtained through the minimum delivery cycle T1. Specifically, the minimum delivery cycle T1 of the whole equipment is used as the interval interval, and the warranty period of the whole equipment is evenly divided into delivery time slices through the minimum delivery cycle T1, and the spare parts indicators under different delivery time slices are calculated. The spare parts delivery list of different time slices is determined through the spare parts indicators under the time slices. In this way, the number of periodic replacement parts is estimated and preparations for spare parts delivery are made. It can improve spare parts turnover and reduce the pressure of spare parts inventory at the beginning of the period.

[0025] Step S2: Use the image integration module to profile the spare parts, and obtain the importance label of the spare parts by quantifying the scores of the spare parts indicators; for details, refer to Figure 2 As shown, step S2 includes: Step S21, perform data quality verification on the above-mentioned spare parts indicators to obtain the vacancy rate of the spare parts indicators. According to whether the system data will change over time, the spare parts indicators are divided into static indicators and dynamic indicators. Among them, the basic data of spare parts belongs to the static indicators directly converted from data that does not change over time, and the consumption behavior data belongs to the dynamic indicators directly converted from data that changes over time. Detect the integrity of each spare parts indicator. For the data missing from the static indicators, supplement the missing values; for the data missing from the dynamic indicators, update the indicator information by regularly refreshing the interface data to ensure the timeliness and completeness of the spare parts indicators. Count each spare parts indicator and calculate the vacancy rate of each spare parts indicator, that is, the ratio of the number of all vacant indicators of the spare part to the number of all indicators of the spare part. If the vacancy rate of the spare parts indicator exceeds the preset value, supplement the corresponding data. In this embodiment, the target spare parts label vacancy rate is greater than 40%, marking the spare part as an incomplete information part that requires additional information.

[0026] Step S22, the spare parts index is processed into intervals, and the scores of each spare parts index are quantified. Optionally, the equal-width binning method is used to assign corresponding scores to different intervals, and the experts flexibly configure the number and range of bins according to the business scenario requirements. For example, according to the equal-width interval of 0.1, the spare parts replacement rate k1 is processed into intervals: [0-0.1), [0.1-0.2), [0.2-0.3), [0.3-0.4), [0.4,0.5), [0.5,0.6), [0.6,0.7), [0.7,0.8) [0.8,0.9) [0.9,1]. The replacement rates k1 of different spare parts are arranged to obtain the scores of the corresponding intervals of the spare parts replacement rates k1. Furthermore, the final scores of each spare parts label are determined by the comprehensive weighting method of subjective weighting plus entropy weighting method.

[0027] Step S23, obtain the comprehensive score of the spare parts by weighted summing the scores of different spare parts indicators. Perform min-max normalization on the comprehensive score of the spare parts, that is, the weighted comprehensive score of the spare parts = (comprehensive score of a spare part - min (comprehensive score of all spare parts)) / (max (comprehensive score of all spare parts) - min (comprehensive score of all spare parts)).

[0028] Step S24: divide the spare parts importance labels obtained according to the weighted comprehensive scores of the spare parts, divide the weighted comprehensive scores of the spare parts into 4 equal parts according to the same interval, and correspond to the four labels of important spare parts, relatively important spare parts, generally important spare parts and unimportant spare parts according to the scores. Store the obtained intervalized spare parts index scores and the final spare parts importance labels in the database. And calculate the missing rate of spare parts importance labels, that is, the ratio of the number of missing spare parts indicators used to calculate spare parts importance labels to the number of spare parts indicators used to calculate spare parts importance labels.

[0029] Step S3: Determine the number of spare parts to be delivered according to the spare parts index and the spare parts importance label obtained in step S2, so as to obtain a spare parts delivery list sliced ​​by time.

[0030] Step S31: Based on the completeness of the spare parts importance label and the total number of complete equipment in the project, the selection basis of the spare parts delivery strategy is as follows: Step S311, when the missing rate of the importance label of the target spare part is less than 20% and the total number of complete equipment is greater than 30; it indicates that the complete equipment of this model has accumulated a certain amount of operating data and the spare parts index integrity is high. According to the spare parts replacement record of the complete equipment, a data set is constructed, and machine learning is used to complete the number of spare parts put in n and the loss probability P in different time periods, and the survival analysis model of LightGBM plus Poisson regression is used for prediction. Among them, the features involved are the first replacement time t1 of the target spare part on the target model, the average first replacement time t2 on different complete equipment of the target model, the average interval replacement time t3 on the target model, the replacement rate k1 and the proportion of complete equipment replaced k2. The decision variable is, the loss function is the continuous ranking probability score CRPS statistic, and the loss probability P in different time periods obtained by the algorithm. Obtain the number of spare parts put in n, n=p*n2*n4. Among them, n is the number of spare parts put in, p is the loss probability, n2 is the number of complete equipment, and n4 is the number of spare parts installed for the complete equipment.

[0031] Step S312: When the missing rate of the importance label of the target spare part is less than 20% and the number of complete equipment is less than or equal to 30. This means that although the complete equipment of this model has certain operating data, the number is small. The number of spare parts put into use is determined by the number of complete equipment, the number of spare parts installed for the complete equipment, and the replacement rate, that is, n=k1*n2*n4. Among them, n is the number of spare parts put into use, k1 is the replacement rate, n2 is the number of complete equipment, and n4 is the number of spare parts installed for the complete equipment.

[0032] Further, step S312 also includes step S3121, assigning a spare part business weight coefficient according to the spare part importance label obtained in step 2, and determining the final spare part delivery quantity according to the business weight coefficient and the spare part delivery quantity. In this embodiment, the spare part business weight coefficient of an important spare part is 2, the spare part business weight coefficient of a relatively important spare part is 1.5, the spare part business weight coefficient of a generally important spare part is 1, and the spare part business weight coefficient of an unimportant spare part is 0.5.

[0033] Step S313: When the missing rate of the target spare parts importance label is greater than 20% and the number of complete equipment is less than or equal to 30, the number of spare parts put into use is determined by the number of R&D recommendations and the number of complete equipment, i.e. n=n2*n5, where n is the number of spare parts put into use and n2 is the number of complete equipment.

[0034] Further, step S313 also includes step S3131, according to the spare parts importance label obtained in step 2, assigning business weights to different spare parts importances to adjust the number of R&D recommendations to obtain the final number of releases. In this embodiment, the spare parts business weight coefficient of important spare parts is 0.8, the spare parts business weight coefficient of more important spare parts is 0.6, the spare parts business weight coefficient of generally important spare parts is 0.4, and the spare parts business weight coefficient of unimportant spare parts is 0.2.

[0035] Step S32, calculate the number of different spare parts on the whole equipment in turn, and form a spare parts delivery list. This embodiment can make a more objective and real-time dynamic evaluation of spare parts, reduce the lag and deviation caused by expert experience, thereby enhancing the accuracy of spare parts delivery and reducing the risk of economic losses caused by deviations.

[0036] Step S4: visualize the deployment plan and output a spare parts deployment report sliced ​​by time.

[0037] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited by the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present application. Without departing from the spirit and scope of the present application, the present application may have various changes and improvements, which fall within the scope of the present application to be protected. The scope of protection claimed in the present application is defined by the attached claims and their equivalents.

Claims

1. A spare parts delivery method driven by spare parts portrait, It is characterized in that The method comprises the following steps: Step S1: obtaining spare parts indicators of a project through a data integration service module, wherein the spare parts indicators include basic data and consumption behavior data; Step S2: Profiling the spare parts through the portrait integration module, quantifying the scores of the spare parts indicators to obtain the importance labels of the spare parts; Step S3: Determine the number of spare parts to be delivered according to the spare parts index and the spare parts importance label obtained in step S2, so as to obtain a spare parts delivery list sliced ​​by time.

2. The spare parts delivery method driven by spare parts portrait according to claim 1, It is characterized in that The basic data includes the number of complete equipment of the target vehicle model, the number of spare parts installed for the complete equipment and the number of R&D recommendations.

3. The spare parts delivery method driven by spare parts portrait according to claim 2, It is characterized in that The consumption behavior data includes the historical replacement records and replacement rates of spare parts.

4. The spare parts delivery method driven by spare parts portrait according to claim 3, It is characterized in that The step S1 also includes determining the loss probability of the spare parts according to the consumption behavior data.

5. The spare parts delivery method driven by spare parts portrait according to claim 1, It is characterized in that The step S2 includes detecting the integrity of the spare parts index; performing interval processing on the spare parts index and quantifying the scores of each spare parts index; and obtaining a comprehensive score of the spare parts by weighted summing the scores of different spare parts indexes; Spare part importance labels and missing rates of spare part importance labels obtained by dividing the importance of spare parts according to the spare part comprehensive scores.

6. The spare parts delivery method driven by spare parts portrait according to claim 4, It is characterized in that The step S3 comprises: When the missing rate of spare parts importance labels is less than the preset threshold and the number of complete equipment is greater than the preset threshold, the number of spare parts delivered is determined by the number of complete equipment, the number of installed spare parts for the complete equipment and the probability of loss; When the missing rate of spare parts importance labels is less than the preset threshold and the number of complete equipment is less than or equal to the preset threshold, the number of spare parts delivered is determined by the number of complete equipment, the number of installed spare parts for the complete equipment and the replacement rate; When the missing rate of spare parts importance labels is greater than its preset threshold and the number of complete equipment is less than or equal to its preset threshold, the number of spare parts to be put into use is determined by the number of R&D recommendations and the number of complete equipment.

7. The spare parts delivery method driven by spare parts portrait according to claim 6, It is characterized in that The step S3 also includes: assigning a spare part business weight coefficient according to the spare part importance label obtained in step 2, and determining a final spare part delivery quantity according to the business weight coefficient and the spare parts delivery quantity.

8. The spare parts delivery method driven by spare parts portrait according to claim 6, It is characterized in that The consumption behavior data also includes the minimum delivery cycle of spare parts. The warranty period of the entire equipment is evenly divided into delivery time slices through the minimum delivery cycle, and the spare parts delivery lists of different time slices are determined through the spare parts indicators under the time slices.

9. A spare parts delivery system driven by spare parts portrait, It is characterized in that include: Data integration service module, used to obtain spare parts indicators according to the preset collection frequency; A portrait integration module, used for obtaining a spare part importance label by quantifying the score of the spare part indicator; An intelligent analysis module, used to calculate the number of spare parts delivered according to the spare parts index and the spare parts importance label, and determine the spare parts delivery list for different time slices; The report export module is used to output visual spare parts delivery reports.

10. The spare parts delivery system driven by spare parts portrait according to claim 9, It is characterized in that The data integration service module includes a customer relationship management system, an enterprise resource planning system and a spare parts management system which are interconnected.

Images