Production planning method and apparatus

Abstract

A production planning method according to an embodiment of the present application includes the steps of acquiring production data of products produced in a plurality of production lines, generating derived variables based on the acquired production data of the products, inputting the derived variables to an artificial neural network model, the artificial neural network model learning the inputted derived variables to predict production run rates of the products, and reflecting the predicted production run rates to make optimal production plans.

Description

Technical Field

[0001] Embodiments of the present invention relate to a production planning method and apparatus. Background Technology

[0002] Production planning in a product manufacturing plant consisting of multiple production lines (e.g., a PCB manufacturing plant using SMT equipment) is a crucial element for more effectively managing production across each line. In particular, high consistency between production plans and actual production results allows for more effective management of plant resources to improve production efficiency, and productivity can be enhanced by accurately allocating tasks to multiple production lines.

[0003] Typically, when planning product production, users directly review historical production data to calculate approximate operating rates and formulate production plans. This often results in significant discrepancies between the actual production figures and the actual output. In this case, production line operating rate can be defined as the product output. (Theoretical production cycle time / Actual product production time)

[0004] At this point, the key indicators determining the operating rate are composed of various factors such as time, equipment, work document characteristics, and personnel. Therefore, it is more difficult for users to predict the actual operating rate of the factory or the production volume of products. In other words, because production plans are formulated based on the user's subjective judgment and have relatively low accuracy, many errors in the production plan may occur. Furthermore, due to uneven production among multiple production lines, the overall product production efficiency of the factory may decrease. Summary of the Invention Technical issues

[0005] According to one aspect of the present invention, the main technical problem is to provide a production planning method and apparatus for formulating optimal production plans by more accurate production operation rate prediction.

[0006] However, these problems are exemplary, and the technical problems to be solved by the present invention are not limited thereto. Technical solution

[0007] A production planning method according to an embodiment of the present invention includes the following steps: acquiring production data of products produced in multiple production lines; generating derived variables based on the acquired production data of the products; inputting the derived variables into an artificial neural network model, wherein the artificial neural network model learns the input derived variables to predict the production operating rate of the products; and formulating an optimal production plan based on the predicted production operating rate.

[0008] The production data may include production equipment characteristics, production time characteristics, installation point characteristics, and production cycle time characteristics.

[0009] In the step of acquiring production data of the product from the multiple production lines, the timing of acquiring the production data can be adjusted to match the operating characteristics of the factory.

[0010] The steps for generating derived variables based on the acquired product production data may include the following steps: setting the importance order of the derived variables based on the product production data.

[0011] The steps for generating derived variables based on the acquired product production data may include the following steps: setting the importance weight values ​​of the derived variables based on the product production data.

[0012] A production planning apparatus according to an embodiment of the present invention includes: a data acquisition unit for acquiring production data of products produced in multiple production lines; a data preprocessing unit for generating derived variables based on the acquired production data of the products; an artificial neural network unit for learning the input derived variables to predict the production operating rate of the products; and a production planning unit for formulating a production plan for the products based on the production operating rate predicted by the artificial neural network unit.

[0013] The artificial neural network unit may include: a derived variable learning unit that receives derived variables generated by the data preprocessing unit and uses the input derived variables to perform learning of an artificial neural network model; and an operation rate prediction unit that predicts the production operation rate of the product using the artificial neural network model learned by the derived variable learning unit.

[0014] The artificial neural network model can be a gradient boosting model (GBM).

[0015] The data acquisition unit may include a data acquisition time period adjustment unit, which adjusts the time for acquiring the production data of the product to conform to the operating characteristics of the factory.

[0016] The data preprocessing unit may include a derived variable importance order setting unit, which sets the importance order of derived variables based on the production data of the product.

[0017] The data preprocessing unit may include a derived variable importance weight setting unit, which sets the importance weight values ​​of derived variables based on the production data of the product.

[0018] Other aspects, features, and advantages, besides those described above, will become apparent from the detailed description, claims, and drawings. Invention Effects

[0019] According to an embodiment of the present invention, a production planning method utilizes production data of products produced through multiple production lines and uses an artificial neural network model to more accurately predict the production line operating rate, thereby enabling the formulation of an optimal production plan based on the predicted operating rate.

[0020] The technical effects of the present invention are not limited to those mentioned above, and those skilled in the art can clearly understand other technical effects not mentioned from the description of the claims. Attached Figure Description

[0021] Figure 1 This is a hardware configuration diagram illustrating the internal structure of a production planning device according to an embodiment of the present invention.

[0022] Figure 2 This is a block diagram illustrating the configuration and operation of a production planning device according to an embodiment of the present invention.

[0023] Figure 3 This is a block diagram illustrating the structure and operation of an artificial neural network unit according to an embodiment of the present invention.

[0024] Figure 4 This is a block diagram illustrating the structure and operation of a data acquisition unit according to an embodiment of the present invention.

[0025] Figure 5 This is a block diagram illustrating the structure and operation of a data preprocessing unit according to an embodiment of the present invention.

[0026] Figure 6 This is an exemplary table showing production data, derived variables, importance order, and importance weight values ​​according to an embodiment of the present invention.

[0027] Figure 7 This is a sequence diagram illustrating a production planning method according to an embodiment of the present invention. Detailed Implementation

[0028] This invention can be modified in many ways and can have many embodiments. Specific embodiments will be shown in the accompanying drawings and described in detail in the description of the invention. However, this is not intended to limit the invention to the specific embodiments, and it should be understood that the invention includes all modifications, equivalents, and even substitutions encompassed within the spirit and scope of the invention. In the description of the invention, the same reference numerals are used for the same constituent elements, even in other embodiments shown.

[0029] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. When describing with reference to the accompanying drawings, the same or corresponding constituent elements will be given the same reference numerals, and repeated descriptions thereof will be omitted.

[0030] In the following embodiments, the terms "first," "second," etc., are not limiting in meaning, but are used to distinguish one constituent element from another.

[0031] In the following embodiments, unless the context clearly indicates otherwise, singular expressions also include plural expressions.

[0032] In the following embodiments, terms such as "comprising" or "having" refer to the presence of features or constituent elements described in the specification, and do not preclude the possibility of adding more than one other feature or constituent element.

[0033] In the accompanying drawings, the sizes of the constituent elements may be enlarged or reduced for ease of illustration. For example, the sizes and thicknesses of the various components shown in the drawings are arbitrarily illustrated for ease of explanation, and therefore the size or shape of the present invention is not limited to the size or shape shown.

[0034] Where an embodiment can be implemented in different ways, a particular process sequence may be performed differently than the described sequence. For example, two consecutively described processes may be performed substantially simultaneously, or they may be performed in the reverse order of the described sequence.

[0035] The terminology used in this application is for illustrative purposes only and is not intended to limit the invention. In this application, terms such as "comprising" or "having" are intended to specify the presence of features, numbers, steps, operations, constituent elements, components, or combinations thereof as described in the specification, but should be understood as not precluding the presence or possibility of one or more other features, numbers, steps, operations, constituent elements, components, or combinations thereof.

[0036] The following detailed description of the invention is illustrated with reference to the accompanying drawings, which illustrate specific embodiments through which the invention can be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. While the various embodiments of the invention differ from one another, they should be understood to be mutually exclusive. For example, specific shapes, structures, and characteristics described in this specification may be modified from one embodiment and implemented in another without departing from the spirit and scope of the invention. Furthermore, it should be understood that the position or arrangement of various constituent elements in each embodiment may be modified without departing from the spirit and scope of the invention. Therefore, the following detailed description should not be construed as limiting, and the scope of the invention should be interpreted to include all the scope claimed by the claims and their equivalents. In the drawings, similar reference numerals denote the same or similar constituent elements in several respects. Artificial intelligence (AI) disclosed in this invention can refer to the field of studying artificial intelligence or the methodology capable of creating artificial intelligence, and machine learning, as a field of artificial intelligence technology, can be an algorithm that enables a computing device to understand a specific object or condition through data learning, or a technique for finding patterns in data and classifying them. The machine learning disclosed in this invention can be understood to include the meaning of operational methods for learning artificial intelligence models.

[0037] The following is for reference Figures 1 to 6 The production planning apparatus according to an embodiment of the present invention will be described.

[0038] Figure 1 This is a hardware configuration diagram illustrating the internal structure of a production planning device according to an embodiment of the present invention. Figure 2 This is a block diagram illustrating the configuration and operation of a production planning device according to an embodiment of the present invention. Figure 3 This is a block diagram illustrating the structure and operation of an artificial neural network unit according to an embodiment of the present invention. Figure 4 This is a block diagram illustrating the structure and operation of a data acquisition unit according to an embodiment of the present invention. Figure 5 This is a block diagram illustrating the structure and operation of a data preprocessing unit according to an embodiment of the present invention. Figure 6 This is an exemplary table showing production data, derived variables, importance order, and importance weight values ​​according to an embodiment of the present invention.

[0039] Reference Figures 1 to 6According to an embodiment of the present invention, a production planning apparatus includes: a data acquisition unit 110 for acquiring production data of products produced in multiple production lines; a data preprocessing unit 120 for generating derived variables based on the acquired production data of the products; an artificial neural network unit 130 for learning the input derived variables to predict the production operating rate of the products; and a production planning unit 140 for formulating a production plan for the products based on the production operating rate predicted by the artificial neural network unit.

[0040] Typically, when using multiple production lines to manufacture products, users predict the operating rate and plan production during the production planning phase based on the existing production history of the same product. However, actual production operating rates vary greatly depending on factors such as the skill level of production workers, time, weekdays, rest periods, shift schedules, equipment malfunctions and repair times, and worker errors. Predicting operating rates based on subjective judgment and considering these factors, and then using multiple production lines for production planning, will result in significant discrepancies with actual production outcomes. Predicting production results becomes even more difficult when there is an initial production volume of a product that has never been produced before.

[0041] Furthermore, users typically report low production efficiency when relying on subjective judgment for production planning. This leads to actual production finishing earlier or later than planned, frequently necessitating urgent changes to the production schedule. Moreover, operating on an unplanned production line can increase preparation time and worker fatigue, while inaccurate planning can extend production time and potentially result in unmet delivery deadlines.

[0042] Therefore, according to one embodiment of the present invention, various derived variables can be generated based on pre-produced production data in a production plant, and these derived variables can be learned by inputting them into an artificial neural network model, thereby enabling more accurate prediction of production operating rates. At this point, a production plan is created based on the predicted production operating rate, minimizing the possibility of errors or urgent changes to the production plan through a more accurate plan, thus minimizing the maximum operating time of the entire plant and providing a method to improve the overall productivity of the plant.

[0043] Reference Figure 1In one embodiment of the present invention, the production planning device 100 may include a processor 101, a memory 102, an input / output unit 103, and a communication unit 104. The memory 102, as a computer-readable recording medium, may include permanent mass storage devices such as random access memory (RAM), read-only memory (ROM), and disk drives. Furthermore, the memory 202 may temporarily or permanently store program code and settings for controlling the organ volume measurement device 200, camera images, and object posture data.

[0044] Processor 101 can be configured to perform basic arithmetic, logical, and input / output operations to process computer program instructions. Instructions can be provided to processor 101 via memory 102 or communication unit 104. For example, processor 101 can be configured to execute received instructions based on program code stored in a recording device such as memory 102.

[0045] The communication unit 104 can provide functionality for communicating with an external server via the network 10. This external server could be, for example, a factory server that provides product production data. As an example, requests generated by the processor 101 of the production planning device 100 based on program code stored in a recording device such as memory 102 can be transmitted to the external server via the network 10 under the control of the communication unit 104. Conversely, control signals or instructions, content, files, etc., provided under the control of the processor of the external server can be received by the production planning device 100 via the network 10 and the communication unit 104. For example, control signals or instructions received from the external server via the communication unit 104 can be transmitted to the processor 101 or memory 102, and content or files can be stored in a storage medium that may also be included in the production planning device 100. Furthermore, the communication method of the communication unit 104 is not limited, but the network 10 can be a short-range wireless communication network. For example, the network 10 can be a Bluetooth, Bluetooth Low Energy (BLE), or Wi-Fi communication network.

[0046] The input / output unit 103 can receive user input and display output data. According to one embodiment, the input / output unit 103 can display the predicted operating rate on a display.

[0047] Furthermore, in other embodiments, the production planning device may include a ratio Figure 1The constituent elements are more numerous. However, it is not necessary to explicitly show most of the existing technical constituent elements. For example, the production planning device 100 may include a battery and charging device that supply power to the internal constituent elements of the user terminal, and may be implemented to include at least a portion of the above-mentioned input / output devices, or may also include other constituent elements such as a transceiver, a Global Positioning System (GPS) module, various sensors, a database, etc.

[0048] Reference Figure 2 , Figure 4 In one embodiment of the present invention, the production planning device 210 may include a data acquisition unit 110. The data acquisition unit 110 can acquire production data of products produced on multiple production lines.

[0049] Product production data may include, for example, production equipment data related to production equipment, production time data related to production time, installation point data related to component installation points, and cycle time data related to the cycle time of the produced products. The aforementioned production data may use production data of products pre-produced in the factory. Since the predicted operating rate is inferred by the artificial neural network unit described later using the production data and operating rate of pre-produced products, the artificial neural network model of the artificial neural network unit according to an embodiment of the present invention can perform supervised learning using the production data of pre-produced products.

[0050] Reference Figure 4 According to this embodiment, the data acquisition unit 110 may include a data acquisition time period adjustment unit 111. The data acquisition time period adjustment unit 111 can adjust the time for acquiring product production data to conform to the operating characteristics of the factory.

[0051] Factory productivity fluctuates over time, making it crucial to continuously acquire production data within the factory. For example, new production data can be acquired at set times each week to reflect the latest factory operating rates.

[0052] However, the more production data a factory has, the more accurate the operating rate prediction will be due to the increased data sample size. Conversely, the more production data accumulated over a longer period, the less accurate the operating rate prediction may be. Therefore, it is necessary to adjust and set the data acquisition period to match the production characteristics of the products being acquired by the factory.

[0053] Reference Figure 2 , Figure 5 and Figure 6In one embodiment of the present invention, the production planning device 100 may include a data preprocessing unit 120. The data preprocessing unit 120 may generate derived variables based on the acquired production data of the products.

[0054] At this time, the data preprocessing unit 120 may include a derived variable importance order setting unit 121 that sets the importance order of derived variables based on the product's production data, and a derived variable importance weight value setting unit 122 that sets the importance weight values ​​of derived variables.

[0055] Here, the artificial neural network unit 130 can learn derived variables to predict the production run rate of the product. At this time, the importance order of the derived variables through the derived variable importance order setting unit 121 and the importance weight value of the derived variables through the derived variable importance weight value setting unit 122 can be input into the artificial neural network unit 130 along with the derived variables. The artificial neural network unit 130 can learn the derived variables, the importance order of the derived variables, and the importance weight value of the derived variables to output a predicted value of the product's production run rate.

[0056] Reference Figure 6 Multiple derived variables can be generated for each piece of production data. For example, derived variables for production equipment data can generate derived variables for production equipment, production lines, and the order in which production equipment is arranged. Furthermore, derived variables for production time data can generate derived variables for operating days, holiday days, production time, and production offset time.

[0057] For example, derived variables for mounting point data can be equivalent to the case of producing PCBs using SMT equipment. In this case, the mounting point data can generate derived variables such as the number of mounting points, the total number of mounting points during operation, and the difference between the number of mounting points of the bottleneck device (the device with the highest cycle time in the production line) and the number of mounting points of the current device.

[0058] Derived variables for production cycle time data can be analogous to the production of PCBs using SMT equipment. In this case, production cycle time data could include cycle time, preceding and following cycle times (in a production line with two lanes), bottleneck time, maximum number of PCBs produced per hour based on lane bottleneck time, maximum number of PCBs produced per hour based on line bottleneck time, maximum number of PCBs produced per hour based on cycle time (the time each machine takes to produce one PCB), inherent chip per hour (CPH: Chip Hour, the number of chips that can be installed per hour), the ratio of the inherent CPH of the current machine to the inherent CPH of the bottleneck machine, the difference between the machine cycle time and the bottleneck time, the maximum line cycle time between lines, the bottleneck line cycle time between lines, line CPH, the maximum CPH between lines, the bottleneck installation point sequence, and the relative distance to the bottleneck installation point, etc.

[0059] Here, the maximum channel cycle time between lines is a derived variable generated per line based on the maximum cycle time of the two preceding and succeeding channels in each line. Furthermore, the bottleneck channel cycle time between channels can refer to the channel cycle time of the bottleneck device per channel. Channel CPH can refer to the maximum number of chips that can be installed per device per hour per channel. The maximum CPH between lines can refer to the maximum number of chips that can be installed per device per hour per line. The bottleneck installation point order can refer to the order of the bottleneck devices. For example, if the third device out of four is the bottleneck, then the derived variable for the bottleneck installation point order can be 3. The bottleneck installation point and relative distance can be the relative distance value between the bottleneck device and the device currently performing the statistics.

[0060] Furthermore, referring to Figure 6 It can display the importance order and weight value of each derived variable. This importance order and weight value set by derived variable can be adjusted according to the production characteristics of the factory.

[0061] Reference Figure 2 and Figure 3 The artificial neural network unit 130 may include an artificial neural network model. The artificial neural network unit 130 may include a derived variable learning unit 131 and an operating rate prediction unit 132.

[0062] The derived variable learning unit 131 can receive derived variables generated by the data preprocessing unit 120, and can use the input derived variables to perform learning of an artificial neural network model. The operation rate prediction unit 132 can predict the production operation rate of the product using the artificial neural network model learned by the derived variable learning unit.

[0063] The artificial neural network model according to this embodiment can be a gradient boosting model (GBM). That is, based on the derived variables extracted by the data preprocessing unit 120, the production operation rate can be predicted by learning the derived variables through the GBM series artificial neural network model. The GBM series artificial neural network models have excellent performance and, compared with other models, do not use a relatively large amount of learning resources, thus reducing the time required to predict the operation rate.

[0064] Furthermore, according to this embodiment, the performance of the operation rate prediction is improved as more effective derived variables are generated. However, as the number of derived variables increases, the time required to predict the operation rate may also increase. Therefore, learning can be performed centered on derived variables that reflect high importance or high importance weight values ​​during prediction.

[0065] The operation rate prediction data generated by the operation rate prediction unit 132 can be transmitted to the production planning unit 140. The production planning unit can then incorporate the operation rate prediction values ​​output by the artificial neural network unit 130 into the candidate routes and production candidate times for the products to be produced in the future, thereby formulating the optimal production plan.

[0066] As an experimental example, when users predict the operating rate and formulate production plans for products, the prediction accuracy of the operating rate is usually calculated to be 50% to 70%. However, when the operating rate is predicted using the artificial neural network model according to this embodiment, the prediction accuracy of the operating rate is 81% to 90%, which shows that the prediction accuracy of the operating rate is significantly improved.

[0067] In this way, when planning production across multiple production lines, more accurate production planning minimizes production delays, thereby improving operational efficiency and ensuring that products are produced within the required delivery time.

[0068] Furthermore, more accurate production plans require less time than before, thus improving the ease of changing previous production plans.

[0069] Furthermore, production is carried out according to the planned schedule when producing actual products, thus enabling efficient preparation of the materials and resources required for production.

[0070] The following is for reference Figure 7 A production planning method according to an embodiment of the present invention will be described. Figure 7 The content not shown in the image can be referred to Figures 1 to 6 The content shown and its explanation.

[0071] Figure 7This is a sequence diagram illustrating a production planning method according to an embodiment of the present invention.

[0072] Reference Figure 7 A production planning method according to an embodiment of the present invention includes: a step of acquiring production data of products produced in multiple production lines (S100); a step of generating derived variables based on the acquired production data of the products (S200); a step of inputting the derived variables into an artificial neural network model, wherein the artificial neural network model learns the input derived variables to predict the production operating rate of the products (S300); and a step of formulating an optimal production plan based on the predicted production operating rate (S400).

[0073] The step (S100) of acquiring production data of the product in the multiple production lines can adjust the time of acquiring the production data of the product to conform to the operating characteristics of the factory.

[0074] The step (S200) of generating derived variables based on the obtained product production data can set the importance order of the derived variables based on the product production data.

[0075] Furthermore, in the step (S300) of generating derived variables based on the obtained product production data, the importance weight value of the derived variables can be set based on the product production data.

[0076] The apparatus and / or system described above can be implemented using hardware components, software components, and / or combinations of hardware and software components. The apparatus and components described in the embodiments can be implemented using one or more general-purpose computers or special-purpose computers (e.g., processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field-programmable gate arrays (FPGAs), programmable logic units (PLUs), microprocessors, or any other device capable of executing instructions and responding). The processing device can execute an operating system (OS) and one or more software applications running on the operating system. Furthermore, the processing device can access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, there are instances where a single processing device is used in the description; however, those skilled in the art will recognize that a processing device can include multiple processing elements and / or multiple types of processing elements. For example, a processing device can include multiple processors or a processor and a controller. Furthermore, it can also be other processing configurations such as parallel processors.

[0077] Software may include computer programs, code, instructions, or combinations thereof, and may constitute a processing apparatus to operate as needed, or to independently or collectively instruct the processing apparatus. Software and / or data, in order to be interpretable by the processing apparatus or to provide commands or data to the processing apparatus, may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium, or device or transmitted signal wave. Software is distributed across computer systems connected via a network and may be stored or executed using distributed methods. Software and data may be stored on more than one computer-readable recording medium.

[0078] The method according to the embodiments can be implemented using program commands executable by various computer units and recorded on a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., individually or in combination. The program commands recorded on the medium may be program commands specifically designed and configured for the embodiments, or program commands known and available to those skilled in the art of computer software. Examples of computer-readable recording media may include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROMs and DVDs; magneto-optical media such as floppy disks; and hardware devices specifically configured to store and execute program commands, such as read-only memory (ROM), random access memory (RAM), and flash memory. Examples of program commands may include not only machine language code generated by a compiler, but also high-level language code executable by a computer using an interpreter or the like. The aforementioned hardware devices may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

[0079] As described above, although the embodiments have been explained with reference to specific examples and accompanying drawings, those skilled in the art can make various modifications and variations based on the above description. For example, the described techniques and methods can be performed in different orders and / or the constituent elements of the described systems, structures, devices, circuits, etc., can be combined or integrated with the described methods in different forms, or appropriate results can be achieved by substituting or replacing other constituent elements or equivalents.

[0080] Therefore, other implementations, other embodiments, and contents equivalent to the claims also fall within the scope of the appended claims.

Claims

1. A production planning method, comprising the following steps: Obtain production data for products manufactured across multiple production lines; Derived variables are generated based on the production data of the acquired products; The derived variables are input into an artificial neural network model, which learns the input derived variables to predict the production rate of the product; and To develop the optimal production plan, the production utilization rate is reflected in the forecast.

2. The production planning method according to claim 1, wherein, The production data includes production equipment characteristics, production time characteristics, installation point characteristics, and production cycle time characteristics.

3. The production planning method according to claim 1, wherein, In the step of acquiring production data of the product from the multiple production lines, the time for acquiring the production data is adjusted to conform to the operating characteristics of the factory.

4. The production planning method according to claim 1, wherein, The steps for generating derived variables based on the acquired product production data include the following steps: The importance order of derived variables is determined based on the product's production data.

5. The production planning method according to claim 1, wherein, The steps for generating derived variables based on the acquired product production data include the following steps: The importance weights of derived variables are determined based on the product's production data.

6. A production planning device, comprising: The data acquisition department acquires production data for products manufactured on multiple production lines. The data preprocessing department generates derived variables based on the acquired product production data. The artificial neural network unit learns the derived variables from the input to predict the production run rate of the product; and The production planning department formulates production plans for products based on the production operation rate predicted by the artificial neural network department.

7. The production planning apparatus according to claim 6, wherein, The artificial neural network unit includes: A derived variable learning unit receives derived variables generated by the data preprocessing unit and uses the input derived variables to perform learning of an artificial neural network model; and The operation rate prediction unit predicts the production operation rate of the product using an artificial neural network model learned by the derived variable learning unit.

8. The production planning apparatus according to claim 7, wherein, The artificial neural network model is a gradient boosting model.

9. The production planning apparatus according to claim 6, wherein, The data acquisition unit includes: The data acquisition time adjustment unit adjusts the time for acquiring the production data of the product to match the operating characteristics of the factory.

10. The production planning apparatus according to claim 6, wherein, The data preprocessing unit includes: The derived variable importance order setting unit sets the importance order of derived variables based on the production data of the product.

11. The production planning apparatus according to claim 6, wherein, The data preprocessing unit includes: The derived variable importance weight setting unit sets the importance weight values ​​of the derived variables based on the production data of the product.

Images