A method and device for predicting spare parts procurement based on machine learning

Abstract

The present application belongs to the field of computer technology, and specifically relates to a method and device for predicting the purchase of spare parts based on machine learning. The method includes: obtaining an inspection and maintenance work ticket within a preset time period; the inspection and maintenance work ticket includes information on the corresponding inspection and maintenance; when there is a replacement of spare parts in the corresponding inspection and maintenance process, the inspection and maintenance work ticket includes the model and consumption of the spare parts; semantically matching the inspection and maintenance work ticket with a pre-obtained spare parts function library; the spare parts function library includes the model of spare parts; when the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than a first preset threshold, the demand for the corresponding spare parts is determined according to the consumption in the inspection and maintenance work ticket. The present application establishes a spare parts procurement prediction model in combination with the inspection and maintenance work ticket, which can greatly improve the efficiency and turnover rate of spare parts management and rationalize spare parts inventory.

Description

Technical Field

[0001] The present application belongs to the field of computer technology, and specifically relates to a method and device for predicting the purchase of spare parts based on machine learning. Background Art

[0002] Effective management of spare parts is an indispensable part of enterprise operation, which is directly related to the production efficiency, cost control and market competitiveness of enterprises. With the expansion of enterprise scale and the complexity of production equipment, the management of spare parts has become more and more important. Scientific and reasonable management of spare parts can ensure that enterprises can obtain the required spare parts anytime and anywhere, ensure the normal production, avoid excessive inventory backlog, reduce capital occupancy rate and improve capital utilization efficiency.

[0003] However, there are many problems in the management of spare parts in various enterprises, mainly manifested in unreasonable inventory, reduced production efficiency, capital loss and other issues caused by unreasonable forecast of spare parts procurement demand. Summary of the invention

[0004] The purpose of this application is to provide a spare parts procurement forecasting method and device based on machine learning, so as to solve the problem of unreasonable inventory caused by unreasonable forecast of spare parts procurement demand in the management of spare parts in the prior art.

[0005] Technical solution to achieve the purpose of this application:

[0006] A first aspect of an embodiment of the present application provides a spare parts procurement prediction method based on machine learning, the method comprising:

[0007] Obtaining a maintenance work ticket within a preset time period; the maintenance work ticket includes information about the corresponding maintenance; when spare parts are replaced during the corresponding maintenance process, the maintenance work ticket includes the model and consumption of the spare parts;

[0008] Semantically matching the inspection and maintenance work ticket with a pre-obtained spare parts function library; the spare parts function library includes spare parts models;

[0009] When the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than a first preset threshold, the demand for the corresponding spare parts is determined according to the consumption in the inspection and maintenance work ticket.

[0010] Optionally, the determining the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket also includes:

[0011] Obtain the service life and the last replacement date of the corresponding spare parts;

[0012] Determine whether the corresponding spare parts need to be replaced according to the service life and the last replacement date of the corresponding spare parts;

[0013] If yes, then obtaining the single consumption in the equipment-related spare parts information library;

[0014] Determining the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket specifically includes:

[0015] The demand for corresponding spare parts is determined based on the single consumption and the consumption in the inspection and maintenance work ticket.

[0016] Optionally, judging whether the corresponding spare parts need to be replaced according to the service life and the most recent replacement date of the corresponding spare parts further includes:

[0017] Obtaining the failure frequency of the corresponding spare parts;

[0018] When the failure frequency of the corresponding spare part is higher than a second preset threshold, the service life of the corresponding spare part is updated according to the failure frequency.

[0019] Optionally, the method further includes:

[0020] When the purchase quantity of spare parts is insufficient or there is an inventory backlog, the first preset threshold is adjusted.

[0021] Optionally, the step of obtaining the inspection and maintenance work ticket within a preset time period also includes:

[0022] Obtaining a predicted procurement cycle for the corresponding spare parts;

[0023] According to the predicted procurement cycle and the next inspection and maintenance start time, the acquisition of the inspection and maintenance work ticket within the preset time period is triggered.

[0024] Optionally, triggering the acquisition of the inspection and maintenance work ticket within a preset time period according to the predicted procurement cycle and the next inspection and maintenance start time specifically includes:

[0025] When the time from the next inspection and maintenance start time to the predicted procurement cycle, the acquisition of the inspection and maintenance work ticket within the preset time period is triggered.

[0026] Optionally, obtaining the predicted procurement cycle of the corresponding spare parts specifically includes:

[0027] When the corresponding spare parts are key spare parts, the historical maximum procurement cycle of the corresponding spare parts is used as the predicted procurement cycle;

[0028] When the corresponding spare parts are non-critical spare parts, multiple historical procurement cycles of the corresponding spare parts are obtained; and the predicted procurement cycle is determined according to the multiple historical procurement cycles and the probability density formula of the normal distribution.

[0029] Optionally, the method further includes determining the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket:

[0030] Obtaining the inventory of the corresponding spare parts;

[0031] The purchase quantity of the corresponding spare parts is determined according to the demand and inventory of the corresponding spare parts.

[0032] A second aspect of an embodiment of the present application provides a spare parts procurement prediction device based on machine learning, the device comprising:

[0033] The first acquisition module is used to acquire an inspection and maintenance work ticket within a preset time period; the inspection and maintenance work ticket includes information about the corresponding inspection and maintenance; when there is a replacement of spare parts during the corresponding inspection and maintenance process, the inspection and maintenance work ticket includes the model and consumption of the spare parts;

[0034] A semantic matching module, used for semantically matching the inspection and maintenance work ticket with a pre-obtained spare parts function library; the spare parts function library includes spare parts models;

[0035] The demand determination module is used to determine the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket when the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than a first preset threshold.

[0036] A third aspect of an embodiment of the present application provides a computer-readable storage device having a computer program stored thereon. When the computer program is executed, any one of the spare parts procurement prediction methods based on machine learning provided in the first aspect of an embodiment of the present application is implemented.

[0037] A fourth aspect of an embodiment of the present application provides an electronic device, including a memory and a processor; the memory stores a computer program, and when the processor runs the computer program, it implements any one of the spare parts procurement prediction methods based on machine learning provided in the first aspect of the embodiment of the present application.

[0038] The beneficial technical effects of this application are:

[0039] The embodiment of the present application provides a method and device for predicting the purchase of spare parts based on machine learning, the method comprising: obtaining an inspection and maintenance work ticket within a preset time period; the inspection and maintenance work ticket includes information on the corresponding inspection and maintenance; when spare parts are replaced during the corresponding inspection and maintenance process, the inspection and maintenance work ticket includes the model and consumption of the spare parts; semantically matching the inspection and maintenance work ticket with a pre-obtained spare parts function library; the spare parts function library includes the model of spare parts; when the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than a first preset threshold, determining the demand for the corresponding spare parts according to the consumption in the inspection and maintenance work ticket. The embodiment of the present application establishes a spare parts procurement prediction model in combination with the inspection and maintenance work ticket, which can greatly improve the efficiency and turnover rate of spare parts management and rationalize spare parts inventory. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A flowchart of a spare parts procurement prediction method based on machine learning is provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make those skilled in the art better understand the present application, the technical scheme 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 embodiments described below are only a part of the embodiments of the present application, not all. Based on the embodiments recorded in the present application, all other embodiments obtained by those skilled in the art without paying creative work are within the scope of protection of the present application.

[0042] See also Figure 1 , which is a flow chart of a spare parts procurement prediction method based on machine learning provided in an embodiment of the present application.

[0043] The present application provides a method for predicting spare parts procurement based on machine learning, including:

[0044] Step S101: Obtain inspection and maintenance work tickets within a preset time period.

[0045] In an embodiment of the present application, the inspection and maintenance work ticket includes information on the corresponding inspection and maintenance; when spare parts are replaced during the corresponding inspection and maintenance process, the inspection and maintenance work ticket includes the model and consumption of the spare parts.

[0046] In specific implementation, the inspection and maintenance work ticket includes multiple fields related to the work content, such as "work object", "work task name", "work content and requirements", "related experience feedback", "spare parts model", "consumption quantity", etc. The preset time period can be within an inspection and maintenance cycle or an overhaul cycle, which is not limited here.

[0047] Step S102: semantically matching the inspection and maintenance work ticket with the pre-obtained spare parts function library.

[0048] In the embodiment of the present application, the spare parts function library includes the model of spare parts. In actual application, the metadata in the spare parts function library mainly includes information such as name, material code, model, current inventory, main functional field, key performance of failure, historical shortest service life, normal service life, manufacturer, historical purchase contract number, etc.

[0049] In specific implementation, the BERT similarity matching algorithm can be used to semantically match the inspection and maintenance work ticket with the spare parts function library, and the matching degree of the two can be determined based on the matching of the fields.

[0050] Step S103: When the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than a first preset threshold, the demand for corresponding spare parts is determined according to the consumption in the inspection and maintenance work ticket.

[0051] It should be noted that since both the inspection and maintenance work ticket and the spare parts function library include multiple fields, when the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than the first preset threshold, it can be considered that the inspection and maintenance work ticket has replaced the spare parts, which means that the work object equipment of the current inspection and maintenance work ticket is likely to need to replace the spare parts. According to the consumption in the inspection and maintenance work ticket, the demand for the corresponding spare parts can be determined. As an example, the first preset threshold can be 60%, which can be set according to the actual situation during the specific implementation, and is not limited here.

[0052] The embodiment of the present application establishes a spare parts procurement forecasting model in conjunction with the inspection and maintenance work ticket, which can greatly improve the efficiency and turnover rate of spare parts management and rationalize the spare parts inventory.

[0053] In one case, in addition to the need for replacement of spare parts for inspection and maintenance work, the expiration of the conventional service life will also generate the need for spare parts. Therefore, in some possible practice methods of the embodiment of the present application, step S103 may also include:

[0054] Obtain the service life and the last replacement date of the corresponding spare parts;

[0055] Determine whether the corresponding spare parts need to be replaced according to the service life and the last replacement date of the corresponding spare parts;

[0056] If so, obtain the single consumption in the equipment-related spare parts information library.

[0057] In the embodiment of the present application, the metadata of the equipment-associated spare parts information library may mainly include information such as equipment name, equipment type, associated spare parts name, associated spare parts material code, associated spare parts model, single consumption, last replacement time, etc. When the sum of the last replacement time and the service life is less than or equal to the current date, it is considered that the corresponding spare parts need to be replaced, and the single consumption of the corresponding spare parts is obtained from the equipment-associated spare parts information library.

[0058] Then, step S103 may specifically include:

[0059] The demand for corresponding spare parts is determined based on the single consumption and the consumption in the inspection and maintenance work ticket.

[0060] As an example, the single consumption of the corresponding spare parts and the consumption in the corresponding inspection and maintenance work ticket can be used as the demand for the corresponding spare parts.

[0061] In one example, step S103 may also include:

[0062] Obtaining the failure frequency of the corresponding spare parts;

[0063] When the failure frequency of the corresponding spare part is higher than a second preset threshold, the service life of the corresponding spare part is updated according to the failure frequency.

[0064] In one case, the failure frequency of the corresponding spare parts is analyzed. When the failure frequency is abnormal, it is considered that the spare parts need to be replaced, the service life of such spare parts is refreshed, and the demand for the corresponding spare parts is updated. After the inspection and maintenance work ticket is completed, the abnormal trend recognition algorithm can be dynamically tuned according to the actual consumption of spare parts. In specific implementation, the second preset threshold can be specifically set according to the actual situation, and is not limited here.

[0065] In some possible practice methods of the embodiments of the present application, the method may further include:

[0066] When the purchase quantity of spare parts is insufficient or there is an inventory backlog, the first preset threshold is adjusted.

[0067] For example, when the purchase quantity of spare parts is insufficient, the first preset threshold can be appropriately lowered; when the spare parts inventory is overstocked, the first preset threshold can be appropriately increased so that the corresponding spare parts demand quantity is more in line with the actual situation, thereby improving the accuracy of procurement forecasts.

[0068] In some possible implementations of the embodiments of the present application, step S101 may also include:

[0069] Obtaining a predicted procurement cycle for the corresponding spare parts;

[0070] According to the predicted procurement cycle and the next inspection and maintenance start time, the acquisition of the inspection and maintenance work ticket within the preset time period is triggered.

[0071] In an embodiment of the present application, the demand for spare parts is determined before the next inspection and maintenance starts, taking into account the corresponding predicted procurement cycle to ensure that there are sufficient spare parts before the next inspection and maintenance starts.

[0072] In one example, triggering the acquisition of an inspection and maintenance work ticket within a preset time period according to the predicted procurement cycle and the next inspection and maintenance start time may specifically include:

[0073] When the time from the next inspection and maintenance start time to the predicted procurement cycle, the acquisition of the inspection and maintenance work ticket within the preset time period is triggered.

[0074] In another example, obtaining the predicted procurement cycle of the corresponding spare parts may specifically include:

[0075] When the corresponding spare parts are key spare parts, the historical maximum procurement cycle of the corresponding spare parts is used as the predicted procurement cycle;

[0076] When the corresponding spare parts are non-critical spare parts, multiple historical procurement cycles of the corresponding spare parts are obtained; and the predicted procurement cycle is determined according to the multiple historical procurement cycles and the probability density formula of the normal distribution.

[0077] In an embodiment of the present application, for critical spare parts, the historical maximum procurement cycle is used as the predicted procurement cycle to ensure that the number of critical spare parts meets the demand during inspection and maintenance to ensure the operating stability of critical equipment; for non-critical spare parts, the predicted procurement cycle is determined based on multiple historical procurement cycles and the probability density formula of the normal distribution, which can ensure that spare parts are purchased on time and in quantity in more than 95% of scenarios.

[0078] In one example, P = P μ +2σ, where P is the preset purchase cycle, P μ is the average value of multiple historical procurement cycles; σ is the standard deviation of multiple historical procurement cycles.

[0079] In some possible implementations of the embodiment of the present application, step S103 may further include:

[0080] Obtaining the inventory of the corresponding spare parts;

[0081] The purchase quantity of the corresponding spare parts is determined according to the demand and inventory of the corresponding spare parts.

[0082] The embodiment of the present application provides a spare parts procurement prediction method based on machine learning, which takes into account the service life of spare parts and must be replaced when they expire; according to the field information such as "work object", "work task name", "work content and requirements" of the work ticket, when the spare parts associated with the equipment have not been replaced recently, it is judged whether the cause of the failure is related to the spare parts, and the number of spare parts required for future inspection and maintenance work is predicted. An additional compensation mechanism is added. Even if the spare parts have been replaced recently, if the frequency of similar equipment failures after replacement is abnormal, the spare parts need to be updated; the purchase quantity of each type of spare parts is determined in combination with the current inventory. By establishing a spare parts procurement prediction model in the form of full-scene coverage, the efficiency and turnover rate of spare parts management can be greatly improved, and the spare parts inventory can be rationalized. The embodiment of the present application establishes a spare parts procurement prediction model in combination with the inspection and maintenance work ticket, which can greatly improve the efficiency and turnover rate of spare parts management and rationalize the spare parts inventory.

[0083] Based on the above-mentioned embodiment, a spare parts procurement prediction method based on machine learning is provided, and the embodiment of the present application also provides a spare parts procurement prediction device based on machine learning.

[0084] The embodiment of the present application provides a spare parts procurement prediction device based on machine learning, comprising:

[0085] The first acquisition module is used to acquire an inspection and maintenance work ticket within a preset time period; the inspection and maintenance work ticket includes information about the corresponding inspection and maintenance; when there is a replacement of spare parts during the corresponding inspection and maintenance process, the inspection and maintenance work ticket includes the model and consumption of the spare parts;

[0086] A semantic matching module, used for semantically matching the inspection and maintenance work ticket with a pre-obtained spare parts function library; the spare parts function library includes spare parts models;

[0087] The demand determination module is used to determine the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket when the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than a first preset threshold.

[0088] In some possible implementations of the embodiments of the present application, the device may further include:

[0089] A second acquisition module is used to acquire the service life and the last replacement date of the corresponding spare parts;

[0090] A first judgment module, used to judge whether the corresponding spare part needs to be replaced according to the service life and the last replacement date of the corresponding spare part;

[0091] A third acquisition module is used to acquire a single consumption amount in the equipment-associated spare parts information library when the first judgment module determines that the corresponding spare parts need to be replaced;

[0092] Demand determination module, specifically used for:

[0093] The demand for corresponding spare parts is determined based on the single consumption and the consumption in the inspection and maintenance work ticket.

[0094] In some possible implementations of the embodiments of the present application, the device may further include:

[0095] A fourth acquisition module, used to acquire the failure frequency of the corresponding spare part;

[0096] The life updating module is used to update the service life of the corresponding spare part according to the failure frequency when the failure frequency of the corresponding spare part is higher than a second preset threshold.

[0097] In one example, the device may further include:

[0098] The threshold adjustment module is used to adjust the first preset threshold when the purchase quantity of spare parts is insufficient or the inventory is overstocked.

[0099] In some possible implementations of the embodiments of the present application, the device may further include:

[0100] A cycle acquisition module, used to acquire the predicted procurement cycle of the corresponding spare parts;

[0101] The execution trigger module is used to trigger the acquisition of the inspection and maintenance work ticket within the preset time period according to the predicted procurement cycle and the start time of the next inspection and maintenance.

[0102] In one example, the trigger module is executed to:

[0103] When the time from the next inspection and maintenance start time to the predicted procurement cycle, the acquisition of the inspection and maintenance work ticket within the preset time period is triggered.

[0104] In another example, the period acquisition module is specifically used to:

[0105] When the corresponding spare parts are key spare parts, the historical maximum procurement cycle of the corresponding spare parts is used as the predicted procurement cycle;

[0106] When the corresponding spare parts are non-critical spare parts, multiple historical procurement cycles of the corresponding spare parts are obtained; and the predicted procurement cycle is determined according to the multiple historical procurement cycles and the probability density formula of the normal distribution.

[0107] In some possible implementations of the embodiments of the present application, the device may further include:

[0108] A fifth acquisition module, used to acquire the inventory of the corresponding spare parts;

[0109] The purchase determination module is used to determine the purchase quantity of the corresponding spare parts according to the demand quantity and purchase quantity of the corresponding spare parts.

[0110] The embodiment of the present application establishes a spare parts procurement forecasting model in conjunction with the inspection and maintenance work ticket, which can greatly improve the efficiency and turnover rate of spare parts management and rationalize the spare parts inventory.

[0111] Based on the above-mentioned embodiments, a spare parts procurement forecasting method and device based on machine learning are provided. The embodiments of the present application also provide a computer-readable storage device having a computer program stored thereon. When the computer program is executed, any one of the spare parts procurement forecasting methods based on machine learning provided in the above-mentioned embodiments is implemented.

[0112] Based on the above-mentioned embodiments, a spare parts procurement forecasting method and device based on machine learning are provided, the embodiments of the present application also provide an electronic device, including a memory and a processor; the memory stores a computer program, and when the processor runs the computer program, any one of the spare parts procurement forecasting methods based on machine learning provided in the above-mentioned embodiments is implemented.

[0113] The present application is described in detail above in conjunction with the accompanying drawings and embodiments, but the present application is not limited to the above embodiments, and various changes can be made within the knowledge of ordinary technicians in the field without departing from the purpose of the present application. Any content not described in detail in the present application can adopt the existing technology.

Claims

1. A spare parts procurement prediction method based on machine learning, characterized in that: The method comprises: Obtaining a maintenance work ticket within a preset time period; the maintenance work ticket includes information about the corresponding maintenance; when spare parts are replaced during the corresponding maintenance process, the maintenance work ticket includes the model and consumption of the spare parts; Semantically matching the inspection and maintenance work ticket with a pre-obtained spare parts function library; the spare parts function library includes spare parts models; When the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than a first preset threshold, the demand for the corresponding spare parts is determined according to the consumption in the inspection and maintenance work ticket.

2. The spare parts procurement prediction method based on machine learning according to claim 1 is characterized in that: The step of determining the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket also includes: Obtain the service life and the last replacement date of the corresponding spare parts; Determine whether the corresponding spare parts need to be replaced according to the service life and the last replacement date of the corresponding spare parts; If yes, then obtain the single consumption in the equipment-related spare parts information database; Determining the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket specifically includes: The demand for corresponding spare parts is determined based on the single consumption and the consumption in the inspection and maintenance work ticket.

3. The spare parts procurement prediction method based on machine learning according to claim 2 is characterized in that: The determining whether the corresponding spare parts need to be replaced according to the service life and the last replacement date of the corresponding spare parts also includes: Obtaining the failure frequency of the corresponding spare parts; When the failure frequency of the corresponding spare part is higher than a second preset threshold, the service life of the corresponding spare part is updated according to the failure frequency.

4. The spare parts procurement prediction method based on machine learning according to any one of claims 1 to 3, characterized in that: The method further comprises: When the purchase quantity of spare parts is insufficient or there is an inventory backlog, the first preset threshold is adjusted.

5. The spare parts procurement prediction method based on machine learning according to any one of claims 1 to 3, characterized in that: The step of obtaining the inspection and maintenance work ticket within the preset time period also includes: Obtaining a predicted procurement cycle for the corresponding spare parts; According to the predicted procurement cycle and the next inspection and maintenance start time, the acquisition of the inspection and maintenance work ticket within the preset time period is triggered.

6. The spare parts procurement prediction method based on machine learning according to claim 5 is characterized in that: The triggering of obtaining the inspection and maintenance work ticket within the preset time period according to the predicted procurement cycle and the next inspection and maintenance start time specifically includes: When the time from the next inspection and maintenance start time to the predicted procurement cycle, the acquisition of the inspection and maintenance work ticket within the preset time period is triggered.

7. The spare parts procurement prediction method based on machine learning according to claim 5 is characterized in that: The obtaining of the predicted procurement cycle of the corresponding spare parts specifically includes: When the corresponding spare parts are key spare parts, the historical maximum procurement cycle of the corresponding spare parts is used as the predicted procurement cycle; When the corresponding spare parts are non-critical spare parts, multiple historical procurement cycles of the corresponding spare parts are obtained; and the predicted procurement cycle is determined according to the multiple historical procurement cycles and the probability density formula of the normal distribution.

8. The spare parts procurement prediction method based on machine learning according to any one of claims 1 to 3, characterized in that: The step of determining the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket further includes: Obtaining the inventory of the corresponding spare parts; The purchase quantity of the corresponding spare parts is determined according to the demand and inventory of the corresponding spare parts.

9. A spare parts procurement prediction device based on machine learning, characterized in that: The device comprises: The first acquisition module is used to acquire an inspection and maintenance work ticket within a preset time period; the inspection and maintenance work ticket includes information about the corresponding inspection and maintenance; when there is a replacement of spare parts during the corresponding inspection and maintenance process, the inspection and maintenance work ticket includes the model and consumption of the spare parts; A semantic matching module, used for semantically matching the inspection and maintenance work ticket with a pre-obtained spare parts function library; the spare parts function library includes spare parts models; The demand determination module is used to determine the demand for corresponding spare parts according to the consumption in the inspection and maintenance work ticket when the matching degree between the inspection and maintenance work ticket and the spare parts function library is greater than a first preset threshold.

10. A computer-readable storage device, characterized in that: A computer program is stored thereon, and when the computer program is executed, the spare parts procurement prediction method based on machine learning as described in any one of claims 1 to 8 is implemented.

11. An electronic device, characterized in that: It includes a memory and a processor; the memory stores a computer program, and when the processor runs the computer program, it implements the spare parts procurement prediction method based on machine learning as described in any one of claims 1 to 8.

Images