JP2008027150A - Unit and method for predicting manufacturing load, computer program, and computer readable storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a prediction model for manufacture load which facilitate of production management job and operation management job simply and precisely in steel manufacturing industry having processes for producing various, small-lot products in large quantities. <P>SOLUTION: The model consists of a classification logic creation step (a step S4) for creating a logic which groups products whose product attributes, such as order information on the products and product specification information, are identical or within a fixed range as a variety group according to a logic creating determination tree; and a manufacturing load prediction model construction step (a step S6) for constructing a manufacturing load prediction model which predicts manufacturing loads for each variety group created on the basis of past manufacturing achievement data. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a production management technique in a manufacturing process of a manufacturing industry having a process for producing a large number of small lot products such as a steel manufacturing industry, and in particular, products having similar product attributes are grouped together to produce a plurality of varieties. A production load prediction device, a production load prediction method, a computer program, and a computer-readable storage that are suitable for use in improving productivity by predicting the production load for each product type group while defining them in groups. It relates to the medium.

  For example, steel products in the steel manufacturing industry have a wide variety of product attributes such as standards that define their materials and sizes according to customer applications, so their manufacturing specifications such as chemical composition and molding method, The heat treatment method has a wide variety of features. In addition, there is an increasing demand for compliance with delivery dates and shortened delivery times in accordance with customer's product use schedule.

  On the other hand, in the manufacturing process, from the viewpoint of improving productivity by mass production, it is required to produce a plurality of orders with the same manufacturing specification in batch units. The manufacturing process of steel products consists of a plurality of manufacturing facilities such as steel making, rolling, refining, and shipping, and the optimum lot conditions of each manufacturing facility are different. Therefore, the pursuit of improving the productivity of lots at a certain manufacturing facility will decrease the productivity of other manufacturing facilities, or the lot summarization at the upstream process will lead to the concentration of the manufacturing load at the downstream process, which will increase the in-process and the manufacturing period. It may cause. Therefore, it is required to create a lot considering the trade-off between manufacturing facilities. In addition, pre-fabrication for producing an appropriate-sized lot causes extra product inventory and a corresponding increase in work period. In the specification of the present application, the manufacturing load refers to a processing amount (manufacturing amount) in each manufacturing facility.

  In contrast, the steel industry often performs the following production management. The sales department receives an order inquiry from the customer and, based on the manufacturing specifications of the order and the requested delivery date, negotiates whether the order is accepted and if necessary, and then instructs the ironworks to produce the order information. Steelworks will consider more detailed manufacturing specifications based on order information instructed by the sales department. Based on the production specifications, the production lot and the production timing of each production facility are planned in consideration of the delivery date, the production period, and the productivity in the production process, and a production instruction is issued to each production facility.

  In this way, although rough delivery date negotiations are underway at the order receiving stage, the sales department receives an order due to the complexity of the manufacturing process and the large scale and complexity of handling large quantities of various types and small lot orders as described above. At present, it is difficult to properly negotiate the delivery date considering the unity of lots at each manufacturing facility and the manufacturing load.

  In addition, after order information is instructed from the sales department to the steelworks, that is, after the delivery date target is determined, the determination of the production start timing is an important planning factor. For example, in a steelmaking facility, it is desired to produce orders of the same component as much as possible from the viewpoint of productivity and yield, but on the other hand, it is necessary to level the production load in the processes after steelmaking. Therefore, it is necessary to devise an optimal plan with the entire manufacturing process as an integrated process, which makes it possible to achieve both large-scale production at steelmaking facilities and leveling of manufacturing load in downstream processes. In reality, however, the production flow is diversified because of the variety of products, and the production load of refining equipment (one of the production equipment) such as the maintenance of defects such as defects found in inspection equipment is the product. Since it is not determined until after the inspection, it is difficult to accurately predict the manufacturing load at the planning stage and to take measures. As a result, the production load after production increases, and the work period may increase due to an increase in work in progress.

  In such a situation, at the current production site, the production process manager classifies the products into broad product groups based on the characteristics of the production process, and uses the actual average production load for each product group to predict the production load. The method of using is often used.

  Further, in Patent Document 1, in an assembly processing line for high-mix low-volume production, product groups having similar requirements for the component configuration or production line of products are handled as groups of the same product, and the load on the production line is appropriately leveled. A production instruction leveling device has been proposed.

JP-A-10-138102 "Basics and mechanism of multivariate analysis that is well understood", Kazunori Yamaguchi et al., Hidekazu System, (2004)

  However, in the conventional method described above, since it is classified into groups of rough varieties based on the knowledge about the manufacturing process, the accuracy of predicting the manufacturing load is different even if the group of the same varieties has different manufacturing load depending on the product. Not always enough.

  In addition, in the production instruction quantity leveling device disclosed in Patent Document 1, a variety is used in that a plurality of varieties having similar requirements for product component configurations or production lines are handled as one varieties group (variety group). The conditions for creating a group are clear, but in a manufacturing process that handles large quantities of small-lot orders and a complex manufacturing flow, the number of types of products that have many product attributes and that have the appropriate scale for the entire manufacturing process There was a problem that it was difficult to clarify the conditions when creating a variety group.

  In addition, as a method to improve the prediction accuracy of manufacturing load, it may be possible to review the definition of product group, but it is necessary to study a large number of combinations of attributes on a trial and error basis, which is not realistic. There was a problem.

  In addition, as another method for improving the prediction accuracy of the manufacturing load, it is conceivable to increase the number of product group groups. However, for the same reason as described above, a large number of combinations of various attributes can be made on a trial and error basis. There was a problem that had to be considered and was not realistic.

  In view of the above problems, the present invention satisfies manufacturing requirements such as capacity constraints of each manufacturing facility in a manufacturing process in which a large variety of products are manufactured in large quantities through processing of a complicated manufacturing flow by a plurality of manufacturing facilities. It is an object of the present invention to provide a manufacturing load predicting device and the like that can quickly and accurately predict a manufacturing load for each product group that is indispensable when creating a satisfied production plan.

The production load prediction device of the present invention is based on past production record data for a production process having a series or branch structure composed of a plurality of production facilities for producing a plurality of products of a plurality of types in a small lot. A production load prediction device for predicting the production load of a product at each production facility, based on the production performance data, products having the same or a predetermined range of product attributes for a plurality of types of products Creating a plurality of product groups having the same product group and sorting the products into the product group, creating product classification logic using decision trees, and based on the manufacturing performance data, the product classification logic A production load prediction model creating means for constructing a production load prediction model for predicting the production load for each of the product type groups using Using the model, characterized in that it and a manufacturing load prediction means for deriving the manufacturing load of each manufacturing facility for the new production plan.
Further, in the manufacturing load prediction device of the present invention, the type classification logic creating means includes one or a plurality of product order information and manufacturing specification information included as explanatory variables of the decision tree based on the manufacturing performance data. It is characterized in that the product category logic is created using the product attribute and the actual value of the passing process pattern of each manufacturing facility as the objective variable.
The manufacturing load prediction method of the present invention is based on past manufacturing performance data for a manufacturing process having a series or branch structure composed of a plurality of manufacturing facilities for producing a plurality of products of a plurality of types in a small lot. A manufacturing load prediction method for predicting a manufacturing load of a product at each manufacturing facility based on the manufacturing performance data, and products having the same or a predetermined range of product attributes for a plurality of types of products Creating a plurality of product groups having the same product group and sorting the products into the product group, creating a product classification logic using a decision tree, and the product classification logic based on the manufacturing result data A production load prediction model creating step for constructing a production load prediction model for predicting a production load for each product group using Using the model, and having a manufacturing load prediction step of deriving a manufacturing load of each manufacturing facility for the new production plan.
Moreover, in the manufacturing load prediction method of the present invention, the type classification logic creating step includes one or a plurality of product order information and manufacturing specification information included as an explanatory variable of the decision tree based on the manufacturing performance data. It is characterized in that the product category logic is created using the product attribute and the actual value of the passing process pattern of each manufacturing facility as the objective variable.
Moreover, the computer program of this invention makes a computer perform each process of the manufacturing load prediction method described above.
A computer-readable storage medium according to the present invention stores the computer program described above.

  In the present invention, when defining a plurality of products whose manufacturing specifications of individual products are the same or within a certain range as a plurality of product groups, a decision tree creation logic is applied, and an objective variable at the time of decision tree creation As a result, the past process data, which is past performance data, is used to reduce trial and error in determining the product category logic and to construct a highly accurate manufacturing load prediction model. Then, by producing a production plan that matches the production capacity using the obtained production load prediction model, it is possible to obtain effects such as productivity improvement, work-in-process reduction, and work period reduction.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, a manufacturing process of a thick plate, which is a typical product in the steel industry, will be described as an example of a manufacturing process including a manufacturing process having a series or branched structure constituted by a plurality of manufacturing facilities. An example of the schematic configuration of the manufacturing equipment for the thick plate manufacturing process is shown in FIG.

  In FIG. 11, in the converter P <b> 1, molten steel in a high-temperature molten state is adjusted to a steel output component that is a chemical component of an intermediate steel product in units of, for example, about 300 [ton], and is output to a ladle. The production unit (lot unit) in the converter P1 is called charge.

  In the continuous casting facility P2, for example, a plate called a slab of about 20 [ton] units is obtained by continuously casting a plurality of charge intermediate products manufactured in the converter P1 and then cutting them at a specified length. To produce intermediate products. A series of production units in this continuous casting machine is called casting. Although it depends on the manufacturing specifications, the 8 to 12 charges are manufactured as one cast.

  In the rolling equipment P3, after heating the slab, the slab is rolled to a predetermined thickness and width, formed, sheared into a plurality of plates, and then inspected for surface quality and size.

  In the finishing equipment P4, the slab that could not be sheared to the size of the order specification by the rolling equipment is cut. In the finishing facilities P5 and P6, correction for quality assurance is performed. In the finishing equipment P7, care is taken to ensure quality. And the product which finished all the processes is arrange | positioned in warehouse P8.

  FIG. 1 shows a schematic configuration of the production load prediction apparatus of the present embodiment. Moreover, in order to also explain embodiment of the manufacturing load prediction method of this invention, the creation flow of the manufacturing load prediction model in the said manufacturing load prediction apparatus is shown in FIG.

  In FIG. 1, 1 is a type classification logic creation device, 2 is a design parameter setting device, 3 is a production load prediction model creation device, 4 is a production load prediction device, 5 is a production load display device, 6 is a production performance data collection device, Reference numeral 7 denotes a manufacturing process.

  In the manufacturing load predicting apparatus configured as described above, the manufacturing result information (manufacturing process (each manufacturing facility) and manufacturing date and time) in the process of manufacturing the thick plate is manufactured from each manufacturing facility in the manufacturing process 7. The data is collected through an I / O board (communication port) via a network such as a LAN in the process, or collected through a manufacturing result data collection device 6 configured by a mass storage device such as a DVD. (FIG. 2, step S1).

  The manufacturing performance information includes various information such as manufacturing specifications such as product composition, size, and manufacturing process, and order information such as delivery date and product quantity. Therefore, the manufacturing performance data collection device 6 extracts only information related to the manufacturing load prediction and performs performance data preprocessing such as excluding abnormal values, creates manufacturing performance data, and stores it in the manufacturing performance database D1. Stored (FIG. 2, step S2).

  Next, the design parameter setting device 2 configured by using an input / output device such as a keyboard and a mouse is necessary to create a decision tree used in the product category logic, such as the number of product groups that classify a plurality of products. Design parameters are set (FIG. 2, step S3).

  Here, the decision tree used to create the kind classification logic will be described. A decision tree is one of the data analysis methods, and is an analysis method that classifies the data as tree branches and leaves according to various conditions as shown in FIG. It is used for classification of information (for example, see pages 143 to 168 of Non-Patent Document 1). The decision tree is composed of multiple nodes that are a cluster of data, and starts from the root node (root node) that represents the entire data, and the proportion of data that has specific attributes at the end nodes (leaf nodes, leaf nodes) Are created while branching the nodes one after another so that there is a large amount of data, that is, biased data is included. The decision tree can be used for various predictions by using the obtained branch condition to the leaf node and past data (learning data) belonging to the leaf node. The attribute to be predicted is called “object variable”, and the attribute describing the data branch condition is called “explanatory variable”. In creating the decision tree, design parameter settings such as how to define objective variables and explanatory variables and how to determine the size of the decision tree (number of nodes and depth) were obtained. This is extremely important because it is closely related to the accuracy of decision tree prediction and ease of handling. Necessary design parameters as described above are set when creating a decision tree.

  Then, based on the pre-processed manufacturing performance data read from the manufacturing performance database D1 and the design parameters, the product classification logic creating device 1 is used to create the product classification logic composed of the decision tree, and the product classification logic Store in the database D2 (FIG. 2, step S4).

  The production load prediction model creation device 3 includes the product attributes included in the production performance data for the past predetermined period for each product extracted from the production performance database D1, and the products stored in the product classification logic database D2. Based on the classification logic, it is determined to which product group each product of the manufacturing result data belongs, and a product group name is assigned to each product information of the manufacturing result data as a product attribute (step S5 in FIG. 2). Then, the following processing is performed.

(A) Past production data for each product to which a product group name is assigned is classified by focusing on the product group name,
(A) By dividing the amount of processing generated at each manufacturing facility for each product group for a certain period in the past (for example, one year) by the total processing amount for each product group for that product period, for each product group and each manufacturing facility The production load prediction model is calculated from the occurrence rate,
(C) Stored in the production load prediction model database D3 as representing the production load of the production group in the production facility (FIG. 2, step S6).

  Next, an outline of a method of applying the manufacturing load prediction model to predict the manufacturing load of each manufacturing facility in the production planning work will be described with reference to FIG. The production load prediction device 4 shown in FIG. 1 reads out the production load prediction model from the production load prediction model database D3 (FIG. 3, step S7), and the production in which the planned production amount for each product group and each period is described. Plan information is read from the production plan database D4 (FIG. 3, step S8). The production plan information is created by a production planner in consideration of the order receipt status, equipment operation status, production cost, productivity, and the like, and is stored in the production plan database D4. The production plan information may be input from an external device via an I / O device such as a network. Then, by multiplying the obtained production plan information (scheduled production amount for each product group / period) by the production load of each production facility in the production load prediction model, the planned production plan is applied to each production facility. The amount of production load to be applied is predicted (FIG. 3, step S9), and the result is sent to and displayed on the production load display device 5 constituted by a computer = display, and production plan information is added or corrected. (FIG. 3, step S10). The production planner confirms the result (FIG. 3, step S11), and takes action such as correcting the production plan if necessary (FIG. 3, step S12).

  Further, by displaying the production result data after production based on the production plan on the production load display device 5 together with the predicted value, it is possible to compare the prediction and the result. Analyzing the cause of the prediction failure by this comparison, leading to various operational improvements, confirming the accuracy of the manufacturing load prediction model, rebuilding the product category logic or rebuilding the manufacturing load prediction model as necessary, It is also possible to maintain the prediction accuracy.

  It is to be noted that the object of the present invention is to supply a storage medium recording a computer program code of software that realizes the functions of the above-described embodiment to a system or apparatus, and the computer (CPU or MPU) of the system or apparatus stores it. Needless to say, this can also be achieved by reading and executing the program code stored in the medium.

  In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the computer program and the storage medium storing the program code constitute the present invention.

  As a storage medium for supplying the program code, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an OS (operating system) running on the computer based on the instruction of the program code. However, it is needless to say that a case where the function of the above-described embodiment is realized by performing part or all of the actual processing and the processing is included.

  Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

  Hereinafter, the Example in the manufacturing process of the thick board in the steel industry of this invention is described. The product category logic creation device 1 extracts product attribute information related to the production load prediction from the production result data stored in the production result database D1 in units of products (FIG. 2, step S1). A part of the extracted information is shown in FIG. Using the product “No.” assigned to each product as a key, manufacturing performance information such as the thickness and width of each product, material standards, presence or absence of specific manufacturing specifications, and presence or absence of processing results at each manufacturing facility have. For example, the product 1 has the standard A1, the production specification 1, and the product size is 12 [mm] in thickness, 1,000 [mm] in width, and 20,000 [mm] in length. In addition to the items listed in FIG. 4, there are a large number of product attributes, but all product attributes related to manufacturing load prediction are read out. For the reading period, for example, past manufacturing performance data is extracted for a predetermined period such as the last half year or one year.

  Next, performance data preprocessing is performed on the read data (FIG. 2, step S2). Since there is a case where information of a specific item may be lost due to some trouble in the read data, the presence or absence is checked and only significant data is extracted. Specifically, data in which the attribute information is blank, data in an unmanufactured state, data that has been started but has not been manufactured, and the like are removed.

  In order to predict the production load of each product type group, it is necessary to define a type group that can accurately capture the probability of processing (hereinafter referred to as “occurrence rate”) at each manufacturing facility based on the actual production data. There is. Here, the occurrence rate is a processing occurrence probability per unit amount of products belonging to the product type group. For example, “the occurrence rate of a certain kind group of finishing equipment (correction 1) P5 is 0.1” is processed by the finishing equipment P5 when 100 [ton] of a product belonging to the kind group is manufactured. The amount is 10 [ton]. In order to increase the prediction accuracy of the production load, it is extremely important to define a product group that has a substantially constant occurrence rate for any product belonging to the product group.

  In the present embodiment, the product attribute assigned in the manufacturing result data read from the manufacturing result database D1 is set as a candidate for an explanatory variable of the decision tree. It is natural that the objective variable is “occurrence rate of each manufacturing facility”, but the occurrence rate of each manufacturing facility is an amount that can be calculated for the first time by using past manufacturing performance data after defining the product group. Therefore, the decision tree creation logic cannot simply be applied with the objective variable as “occurrence rate of each manufacturing facility”.

  Therefore, in this embodiment, “passing process pattern of each manufacturing facility” is set as an objective variable. Specifically, the actual value of the presence / absence of passing through each of the manufacturing equipment for cutting, straightening 1, straightening 2 and care, which are the finishing equipment, is set to 1 when passing and 0 when not passing. The presence / absence of passage of each facility for all steps is represented by a passage step pattern of a combination of 1 and 0. An example is shown in FIG. Thus, by symbolizing the passage presence / absence of each manufacturing facility as a passage process pattern, it is possible to set the objective variable based only on the production result data without defining the product type group in advance. Since the passage process pattern has a high correlation with the occurrence rate of each manufacturing facility that is originally desired, the objective variable can be substituted without reducing accuracy.

  Furthermore, the accuracy can be further improved by setting the passing process pattern not only to pass the manufacturing facility but also to the number of passes. Whether to select only the presence / absence of passage or the number of passages may be selected in accordance with the characteristics of the manufacturing process to be predicted. For example, if the same facility is often passed multiple times, the number of passages is incorporated into the pattern, and if not, only the presence or absence of passage may be considered.

  However, assuming that all types of passing process patterns that have actually occurred are objective variables, the number of actual data decreases in the case of passing process patterns that occur infrequently, so the reliability of the data on the occurrence rate may be poor. . In such a case, the actual amount of generation for each passing process pattern is large, that is, only the main passing process pattern is the target variable, and the other passing process patterns with a small amount of generation are aggregated as “other processes”. It is preferable to use the objective variable. In this way, the main passing process pattern that imposes a load on the manufacturing equipment is modeled with high accuracy, and the other passing process patterns are used with less influence on the manufacturing equipment, while reducing the accuracy of the model. Therefore, it is possible to reduce the number of variety groups and facilitate the management of the variety groups. A specific example is shown in FIG. In this example, as a result of investigating the generation amount of the passing process pattern, 0000 passing process patterns that do not occur in all processes and patterns (1000, 0100, 0010, 0001) that passed only one of the processes are all. Since most of the passing process patterns accounted for, they are set as objective variables, and other passing process patterns are aggregated as a passing process pattern of “Others”. In this way, the types of objective variables can be reduced from a total of 16 patterns to 6 patterns while minimizing the effect on the accuracy of the model.

  As other design parameters, parameters related to the structure of the decision tree such as the number of leaf nodes of the decision tree to be created and the depth of the tree structure are set. In the present embodiment, an upper limit value of the number of data held by one leaf node is given. If the upper limit value is decreased, the number of leaf nodes increases, that is, the decision tree becomes larger and deeper. In this embodiment, the upper limit value of the number of data is 200.

  Next, a decision tree is created based on the design parameters described so far (FIG. 2, step S4). In creating the decision tree, commercially available decision tree creation software may be used. An example of the obtained decision tree is shown in FIG. At first, the thickness of the steel plate is used as an explanatory variable, and it branches depending on whether the plate thickness is thicker or thinner than 32 [mm], and when it is thin, it branches depending on whether the plate width is narrower than 800 [mm]. In this way, the terminal leaf node is reached by repeating the branching gradually. Representative objective variables are assigned to the terminal leaf nodes.

  Next, according to the created decision tree, a product group name is given to each product in the production result data for each product (FIG. 2, step S5). For example, following the decision tree branch condition of FIG. 7 based on the size and standard attributes of the product 1 in FIG. 4, the product group “1000” is obtained, and the product group name “1000” is assigned as the product 1 attribute. In the same manner, a variety group name is assigned to all products using a decision tree.

  Next, the production performance data is classified for each product type group, and the average value of the number of times of passing through each manufacturing facility for each product type group is calculated to create a production load prediction model for each product type group (step S6). An example of the manufacturing load prediction model is shown in FIG. The production load prediction model has an incidence [%] of each production facility for each product group. The created production load prediction model for each product group is stored in the production load prediction model database D3.

  Next, the case where the production load prediction model derived as described above is used for production planning work will be described. Here, the production planning work related to the production in the continuous casting equipment P2 in the steel making process will be described as an example.

  The continuous casting equipment P2 is equipment for continuously casting 8 to 12 molten steels whose components are adjusted in the converter P1. Since the component adjustment in the converter P1 is performed in units of about 300 [tons], orders from customers are aggregated into lots (charges) in units of 300 [ton] while considering the delivery date, and at the continuous casting facility P2. The daily casting schedule is formulated by concentrating on the cast (8 to 12 charges), which is the processing unit. FIG. 9 shows an example of a casting schedule that has been planned. Plan the daily production of steel output components.

  In planning the casting schedule, the productivity and delivery date in the continuous casting facility P2 are taken into account, but it is also important to consider the manufacturing load in the subsequent processes after the continuous casting facility P2. Therefore, the production load of each production facility is predicted based on the product type group and production load prediction model included in the once-designed casting schedule, and the casting schedule is validated by comparing with the capacity and operation schedule of each production facility. Actions such as changing the casting timing of products can be taken before production if there is a problem.

  As described above, when the production load prediction model is used to evaluate the validity of the production plan, the accuracy is important. FIG. 10 shows a comparison between the production load of the maintenance equipment P7 predicted using the production load prediction model used in this example and the actual value. The abscissa represents the period when the unit of 5 days is January, and the ordinate represents the results of the number of products generated in the treatment facility P7 and the predicted value. It can be seen that the prediction is accurate.

  In this way, by constructing a production load prediction model that can accurately predict the production load with manageable product group numbers, the quality of production management operations and operation management operations in the steel manufacturing industry that produces a wide variety of products is improved. Can be made.

  As described above, according to the present invention, a highly accurate manufacturing load prediction model can be easily obtained in a manufacturing process in which a large variety of products are manufactured through processing at a plurality of manufacturing facilities as in the steel industry. Therefore, it can be used for operation monitoring / prediction by improving production planning accuracy and detecting changes in the operation state. Therefore, the present invention is very useful in industries that are manufacturing industries such as the steel industry.

It is a figure showing schematic structure of the manufacturing load prediction apparatus which concerns on embodiment of this invention. It is the flowchart which showed the production method of the manufacturing load prediction model which concerns on embodiment of this invention. It is the flowchart which showed the method of applying the manufacturing load prediction model which concerns on embodiment of this invention to production planning work. It is a figure showing an example of the manufacture performance data in the Example of this invention. It is a figure showing an example of the symbolization of the passage process pattern in the Example of this invention. It is a figure showing an example of the setting of the objective variable in the Example of this invention. It is a figure showing an example of the decision tree for the kind division in the Example of this invention. It is a figure showing an example of the manufacturing load prediction model in the Example of this invention. It is a figure showing an example of the casting schedule of the casting installation in the Example of this invention. It is a figure showing the generation | occurrence | production prediction precision of the care installation in the Example of this invention. It is the schematic of an example of the thick board manufacturing process in the steel industry.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Type classification logic preparation apparatus 2 Design parameter setting apparatus 3 Manufacturing load prediction model preparation apparatus 4 Manufacturing load prediction apparatus 5 Manufacturing load display apparatus 6 Manufacturing performance data collection apparatus 7 Manufacturing process D1 Manufacturing performance database D2 Product classification logic database D3 Manufacturing load prediction Model database D4 Production plan database P1 Converter P2 Continuous casting equipment P3 Rolling equipment P4 Cutting equipment P5 Straightening equipment 1
P6 Correction equipment 2
P7 Care facilities P8 Warehouse

Claims (6)

Products in each manufacturing facility based on past manufacturing results data for manufacturing processes with a series or branch structure composed of multiple manufacturing facilities to produce multiple products of multiple types in small lots A production load prediction device for predicting the production load of
Based on the manufacturing performance data, a plurality of product groups having the same product attribute for a plurality of types of products or a product within a predetermined range are created as a plurality of product groups, and the products are classified into the product groups. Product classification logic creation means for creating product classification logic by tree;
Based on the manufacturing performance data, a manufacturing load prediction model creating means for building a manufacturing load prediction model for predicting a manufacturing load for each product group using the product classification logic;
A production load prediction device comprising: a production load prediction means for deriving a production load of each production facility for a new production plan using the production load prediction model.   The product category logic creating means uses one or more product attributes included in product order information and manufacturing specification information as explanatory variables of the decision tree based on the manufacturing performance data, and each manufacturing facility as a target variable. The production load predicting apparatus according to claim 1, wherein the type classification logic is created using the actual value of the passing process pattern. Products in each manufacturing facility based on past manufacturing results data for manufacturing processes with a series or branch structure composed of multiple manufacturing facilities to produce multiple products of multiple types in small lots A production load prediction method for predicting the production load of
Based on the manufacturing performance data, a plurality of product groups having the same product attribute for a plurality of types of products or a product within a predetermined range are created as a plurality of product groups, and the products are classified into the product groups. A variety classification logic creation process for creating a variety classification logic by tree,
A production load prediction model creating step for constructing a production load prediction model for predicting a production load for each of the product type groups using the product type classification logic based on the production result data;
A production load prediction method comprising: a production load prediction step of deriving a production load of each production facility for a new production plan using the production load prediction model.   The product category logic creation step uses one or more product attributes included in product order information and manufacturing specification information as explanatory variables of the decision tree based on the manufacturing performance data, and each manufacturing facility as a target variable. The production load prediction method according to claim 3, wherein the kind classification logic is created using the actual value of the passing process pattern.   A computer program causing a computer to execute each step of the manufacturing load prediction method according to claim 3.   A computer-readable storage medium storing the computer program according to claim 5.

Images