JP2010128679A - Manufacturing work period model creation device, delivery time achievement rate predication device, manufacturing start target period calculation device, manufacturing work period model creation method, delivery time achievement rate prediction method, manufacturing start target period calculation method, program and computer-readable storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To set a proper manufacturing work period on which manufacturing actual situations are reflected in forming a production plan, and to form a plan to achieve a target delivery time achievement rate without excessively increasing a work in process, and to preevaluate the validity/invalidity of a production plan by predicting the delivery time achievement rate of the formed production plan. <P>SOLUTION: A manufacturing work period model is created by model on the basis of manufacturing ordering information such as the size or delivery time of a produced product and the past manufacturing result information such as a manufacturing work period result or a processing step result and a model classification logic. Also, product ordering information to be the object of a production plan is classified by model on the basis of the model classification logic, and a model-categorized period margin period being a difference between the delivery time of an order and a manufacturing start scheduled period is calculated, and a model-categorized delivery time achievement rate is calculated from the model-categorized manufacturing work period model and model-categorized work period margin period-categorized order quantity, and the delivery time achievement rate of a whole production plan is calculated by calculating the weighted linear sum of the model-categorized delivery time achievement rate. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention provides, for example, production order planning in a manufacturing plant that manufactures a plurality of products with different manufacturing specifications such as processing steps and processing order by processing in a plurality of processes, such as product order information such as delivery date, order quantity, and product specifications. Estimate the required manufacturing lead time of products and the delivery rate achievement of the planned production plan, calculate the manufacturing start target time, etc. The present invention relates to a device, a manufacturing start target time calculation device, a manufacturing work model creation method, a delivery date achievement rate prediction method, a manufacturing start target time calculation method, a program, and a computer-readable storage medium.

  Product manufacturing processes in many industries, including the steel industry, require that products be manufactured appropriately according to the contents of orders from customers. For example, in the steel industry, it is required to deliver a product having product order information requested by a customer (for example, size, strength, presence / absence of surface coating, etc.) in a required amount on a required date. In addition, with the diversification of customer needs, there is a growing tendency to require so-called “multi-product low-volume production” in which product order information, order quantities, and delivery dates vary for each order. Are required to produce products that satisfy all product order information at low cost. More recently, the ability to deliver products in the so-called “short construction period”, where the period from ordering to delivery is shorter than before, has become significant as one of the added value of products, leading to an increase in manufacturing costs. In addition, the manufacturing industry is required to have manufacturing ability that can withstand the demands of these customers.

  When drafting a production plan that meets these requirements, the desired date for starting the upstream process is usually determined by going back from the delivery date to the standard work schedule (standard lead time) for each order. While satisfying as much as possible, the production plan is drawn up in consideration of other planning conditions such as production lot and production capacity.

  Further, in Patent Document 1, manufacturing apparatuses are clustered based on the similarity of the pre-processing waiting time distribution of each manufacturing apparatus, the pre-processing waiting time representative model distribution of each cluster is calculated, and the sum of the representative model distributions is obtained. Therefore, a method for predicting the expected value and variance of the entire construction period has been proposed.

  Further, in Patent Document 2, in order to calculate the safety stock quantity, the probability Pb that the customer's requested delivery date for an article is shorter than the lead time L is calculated, and the lead time L exceeds the customer requested delivery date The average value LL of the difference between the lead time L and the customer requested delivery date is calculated, the inventory adjustment period N is corrected using the average value LL, the standard deviation σ of the demand for the article, the corrected inventory adjustment period A method for calculating the safety stock quantity from N, probability Pb, shipping frequency F, and safety coefficient k has been proposed.

  Further, in Patent Document 3, logic for grouping products having product attributes such as product order information and manufacturing specification information that are the same or within a certain range as a product group is created by a decision tree creation logic, and past manufacturing is performed. A method for predicting the manufacturing load for each product type group based on actual data has been proposed.

Japanese Patent Laid-Open No. 2004-30088 International Publication No. 2004/022463 Pamphlet JP 2008-27150 A

  In an actual manufacturing plant, there are restrictions on the processing capacity of the processing process, waiting for processing due to the in-process of each process, and there are various operational variables such as failure, so it is necessary for actual manufacturing. The period often varies even with the same product order information. Accordingly, even if the production start target time of the upstream process is set by going back from the delivery date with the standard work schedule described above, the delivery date may not be maintained even if the manufacture is started in the upstream process at the production start target time.

  In addition, in order to prevent such a delay in delivery time, in calculating the standard work schedule, in addition to the work schedule required for manufacturing, a surplus work schedule for absorbing the above work schedule fluctuations may be added. When the desired work date in the most upstream manufacturing process is calculated using the standard work schedule including the spare work schedule in this way, the manufacturing is started at a timing considerably earlier than the actually required manufacturing work schedule. The early start of production results in in-process work in progress and inventory in the warehouse, and these in-process and stock increase lead time. In addition, the manufacturing capacity is hindered due to the increased handling load accompanying the occupation of the product storage area, resulting in a vicious cycle of extending the construction period.

  Moreover, in patent document 1, the expected value and dispersion | distribution of the whole work period are estimated by superimposing the work period distribution for every manufacturing equipment using equipment use frequency information, but in a complicated manufacturing factory, after product inspection In many cases, the number of times the equipment is used is unknown before production, such as by adding a process. In addition, in manufacturing processes with various types of products, the equipment that becomes a bottleneck varies depending on the product, so that there is a problem that the work period distribution information for each manufacturing equipment varies depending on the order configuration.

  Further, in Patent Document 2, the lead time distribution is calculated for each article. However, in a manufacturing process with a large amount of products and various types of products, a large scale is needed to handle the lead time distribution for each article. There is a problem that it is necessary to handle information, and the calculation time becomes too long, so that the production planning work cannot be executed.

  Patent Document 3 proposes a method for creating a product group based on a product classification logic created using a decision tree creation logic and predicting a manufacturing load for each product. There is no.

  The present invention solves the above-mentioned problems, and an appropriate manufacturing period that reflects manufacturing ability is set at the time of production planning in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes. By setting for each product type and calculating the necessary manufacturing lead time within a practical time period, it is possible to formulate a plan to achieve the target delivery rate without excessively increasing the work in progress, and the planned production It is intended to be able to evaluate the quality of the production plan before the start of manufacturing by predicting the delivery rate of the plan before the start of manufacturing.

In order to solve the above problems, according to one aspect of the present invention, when product order information of a product to be manufactured is input for a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes. A manufacturing period model creation device for creating a manufacturing period model for outputting the manufacturing period of the product to be manufactured,
Production including at least one product order performance information including at least size, mass, and delivery date as product attributes for at least one or more already manufactured products by the manufacturing factory, manufacturing work period results of the manufactured products, and processing results in the above process An input unit for inputting achievement information;
Based on the product order record information, a product type classification logic is created that receives the input of the product order record information or the product order information and creates a product type classification logic for classifying the manufactured product or the product to be manufactured into a plurality of types. And
A product classification logic storage unit that stores the product classification logic created by the product classification logic creation unit;
Using the product classification logic stored in the product classification logic storage unit, the product classification unit for classifying the manufactured product into a plurality of product types,
A manufacturing period model creation unit that creates a manufacturing period model that inputs the type classified by the type classification unit and outputs the manufacturing period, using at least the manufacturing performance information of the already manufactured products classified by the type classification unit When,
A manufacturing period model storage unit for storing the manufacturing period model created by the manufacturing period model creation unit;
There is provided a manufacturing period model creation device characterized by comprising at least.

In addition, in the manufacturing construction model creation device,
The manufacturing period model creation department
A group setting unit for setting a plurality of predetermined groups different from the product type for the manufactured products from the manufacturing performance information;
Based on the production order record information, a group-by-product generation rate generation unit that calculates the rate of occurrence of each group by product type classified by the product type classification unit;
Based on the production performance information, a group-specific manufacturing period model creation unit that creates a manufacturing period model with the above group as an input and a necessary manufacturing period as an output;
Create a production period model for each type by creating a production period model with the above type as an input and the production period as an output by calculating the weighted linear sum of the above group-specific production period model with the group-specific occurrence rate as a weight And
May be provided.

In addition, in the manufacturing construction model creation device,
The predetermined group for the manufactured product set by the group setting unit may be a processing process pattern represented by a combination of presence or absence of processing in the plurality of processes.

In addition, in the manufacturing construction model creation device,
The above product classification logic creation part
An explanatory variable setting unit for setting one or more product attributes included in the product order information as explanatory variables;
A target variable setting unit that creates a predetermined group different from the product type for the manufactured product from the manufacturing performance information, and sets the group as a target variable;
A design parameter input unit for inputting design parameters when creating the product type classification logic;
Based on the explanatory variable, the objective variable, and the design parameter, a decision tree creating unit that creates a variety classification logic by a decision tree construction method;
May be provided.

In addition, in the manufacturing construction model creation device,
The manufacturing period model creation unit may generate a manufacturing period model that receives the type classified by the type classification unit as an input and outputs the frequency distribution of the manufacturing period for the type.

In addition, in the manufacturing construction model creation device,
The manufacturing period model creation unit may generate a manufacturing period model that receives the type classified by the type classification unit and outputs a function that approximates the frequency distribution of the manufacturing period for the type.

In order to solve the above problems, according to another aspect of the present invention, in the production planning work in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, A delivery date achievement rate prediction device that predicts a delivery date achievement rate that is a ratio of the total order quantity of the order quantity that is manufactured by the delivery date, using a creation device,
The above-mentioned product order information including at least the size, mass, and delivery date of the product to be manufactured that is the target of forecasting the delivery date, and the above-mentioned production start schedule for an order for which the production start time is set out of the orders for the product to be manufactured Based on the product order information and the product type classification logic, the product type information is read from the input unit for reading the production plan information including at least the time, and the product type classification logic storage unit of the production period model creation device. The difference between the delivery date of the product order information of the product to be manufactured classified by product type by the order data classification unit and the order data classification unit classified by product type and the production start time of the production plan information For each product type, calculate the work period margin period and calculate the order quantity for each product type and each work period margin period from the product order information. And another order quantity calculation unit period allowance period,
Read the manufacturing period model from the manufacturing period model storage section of the manufacturing period model creation device, and calculate the delivery date achievement rate for each type from the manufacturing period model and the order quantity for each type and for each period margin Delivery time achievement rate calculation department,
A delivery date achievement rate calculation unit for calculating a delivery date achievement rate of the production plan by calculating a weighted linear sum of the delivery date achievement rate by product type with the order quantity for each product type and work period margin period as a weight; and
A delivery date achievement rate predicting device is provided.

In order to solve the above problems, according to another aspect of the present invention, in the production planning work in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, A production start target time calculation device that calculates a target time to start manufacturing based on the delivery date of an order using a creation device,
An input unit for inputting the product order information including at least the size, mass, and delivery date of the product to be manufactured, which is the target for calculating the manufacturing start target time, and the target delivery date achievement rate for each type of the product to be manufactured;
An order data classification unit that reads the type classification logic from the type classification logic storage unit of the manufacturing period model creation device, and classifies the product to be manufactured according to the type based on the product order information and the type classification logic;
Read out the manufacturing period model for each type from the manufacturing period model storage unit of the manufacturing period model creation device, and based on the manufacturing period model for each type and the target delivery time achievement rate for each type, the necessity of the product to be manufactured A manufacturing period calculation unit for calculating the manufacturing period,
A production start target time calculation unit that calculates a target value of the start time of manufacture of the order by going back from the delivery date of the order by the required manufacturing period; and
There is provided a manufacturing start target time calculation device characterized by comprising:

In order to solve the above-mentioned problem, according to another aspect of the present invention, product order information of a product to be manufactured is input for a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes. A manufacturing period model creating method for creating a manufacturing period model for outputting the manufacturing period of the product to be manufactured,
Manufacturing including at least one product order performance information including at least size, mass, and delivery date as product attributes for at least one or more already manufactured products by the manufacturing factory, manufacturing construction time results of the manufactured products, and results of presence or absence of processing in the above process An input step for inputting the result information by the input unit;
Based on the product order record information, the type classification logic creation unit generates a type classification logic for receiving the input of the product order record information or the product order information and classifying the manufactured product or the product to be manufactured into a plurality of types. The kind classification logic creation step to create,
A product classification logic storage step for storing the product classification logic created in the product classification logic creation step in the product classification logic storage unit;
Using the kind classification logic stored in the kind classification logic storage step, the kind classification step of classifying the manufactured product into a plurality of kinds by the kind classification unit;
Using the production result information of the already manufactured products classified in the product classification step, create a production period model with the production period model as input and the production period model as the output. A production period model creation step to
A manufacturing period model storage step for storing the manufacturing period model created in the manufacturing period model creation step in the manufacturing period model storage unit;
Is provided, and a manufacturing period model creation method is provided.

In order to solve the above problems, according to another aspect of the present invention, in the production planning work in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, A delivery date achievement rate prediction method for predicting a delivery date achievement rate, which is a ratio of the order quantity completed by the delivery date to the total order quantity using a creation device,
The above-mentioned product order information including at least the size, mass, and delivery date of the product to be manufactured that is the target of forecasting the delivery date, and the above-mentioned production start schedule for an order for which the production start time is set out of the orders for the product to be manufactured The production plan information including at least the time is read by the input unit, and the order data classification unit reads the product classification logic from the product classification logic storage unit of the manufacturing period model creation device, and the product to be manufactured is the product. An order data classification step for classifying by product type based on the order information and the product type classification logic;
The difference between the delivery date of the product order information of the product to be manufactured classified by product type in the order data classification step and the scheduled production start time of the production plan information by the order quantity calculation unit by construction period margin period by product type Calculate a certain work period margin period, and from the product order information, calculate the order quantity for each product type and each work period margin period.
The delivery date achievement rate calculation unit for each product type reads the above-mentioned production schedule model from the production schedule model storage unit of the above-mentioned production schedule model creation device. A delivery rate achievement rate calculation step for each product type to calculate the delivery rate achievement rate of
The delivery date achievement rate calculated by the delivery date achievement rate calculation unit by calculating the delivery date achievement rate of the above production plan by calculating the weighted linear sum of the delivery rate achievement rate by product type with the order quantity by product type and construction period margin period as the weight A calculation step;
The delivery date achievement rate prediction method characterized by including at least is provided.

In order to solve the above problems, according to another aspect of the present invention, in the production planning work in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, A production start target time calculation method for calculating a target time to start manufacturing based on a delivery date of an order using a creation device,
An input step for inputting the product order information including at least the size, mass, and delivery date of the product to be manufactured, which is the target for calculating the manufacturing start target time, and the target delivery date achievement rate for each type of the product to be manufactured using the input unit;
Order data classification unit reads the type classification logic from the type classification logic storage unit of the manufacturing period model creation device, and orders data for classifying the product to be manufactured according to the type based on the product order information and the type classification logic A classification step;
The manufacturing period calculation unit reads out the manufacturing period model for each type from the manufacturing period model storage unit of the manufacturing period model creation device, and based on the manufacturing period model for each type and the target delivery rate achievement rate for each type, A necessary manufacturing period calculation step for calculating the necessary manufacturing period of the product to be manufactured,
A production start target time calculation step of calculating a target value of the start time of manufacture of the order by a manufacture start target time calculation unit by going back from the delivery date of the order by the required manufacturing period; and
The manufacturing start target time calculation method characterized by including at least.

In order to solve the above problem, according to another aspect of the present invention, a computer includes:
A program for executing each step of the manufacturing period model creation method is provided.

  In order to solve the above problems, according to another aspect of the present invention, there is provided a computer-readable storage medium characterized by storing the program.

  As described above, according to the present invention, when planning a production plan in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, an appropriate manufacturing period that reflects the manufacturing ability is set for each type. By setting it to, and calculating the required manufacturing lead time within a practical time period, it is possible to formulate a plan to achieve the target delivery date without excessively increasing the work in progress, and By predicting the delivery rate achievement rate before the start of manufacturing, the quality of the production plan can be evaluated before the start of manufacturing.

  Exemplary embodiments of the present invention will be described below in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

  In the following, as an example of a manufacturing plant that manufactures multiple products of different varieties by processing in multiple processes, a manufacturing plant for thick steel plates (thick plates), which is a typical product in the steel industry, is given as an example. Each embodiment of the present invention will be described. However, it goes without saying that this example of a manufacturing factory does not limit the scope of application of the manufacturing period model creation device and the like according to each embodiment of the present invention. In addition, before describing the manufacturing period model creation device and the like according to each embodiment of the present invention, first, a manufacturing factory in the steel industry exemplified here will be described.

<Outline of manufacturing plant>
First, with reference to FIG. 11, an example of a schematic configuration of a manufacturing factory for a thick steel plate (thick plate) which is a representative product in the steel industry will be described. In addition, in FIG. 11, as a structure of this manufacturing factory, by showing the process performed by each structure, and the flow (the arrow in FIG. 11) of the work-in-process (product in the middle of manufacture) flowing through each process, This manufacturing plant is shown conceptually.

  In the converter process 1001, a steel output component that is a chemical component of a steel intermediate product (molten steel) in a high-temperature molten state is adjusted in units of, for example, about 300 [ton], and the steel is discharged into a molten steel pan. The steel output unit in the converter 1001 is called “charge”.

  In the continuous casting process 1002, the molten steel produced in the converter process 1001 is continuously cast for a plurality of charges and then cut to a specified length, for example, a plate called “slab” of about 20 [ton] units. To produce intermediate products. A series of production units in the continuous casting process 1002 is referred to as “cast”. Although it depends on manufacturing specifications, for example, approximately 8 to 12 charges are manufactured as one cast.

  In the rolling process 1003, the slab manufactured in the continuous casting process 1002 is heated and then molded to a predetermined thickness and width. In the refining (cutting) step 1004, cutting to the size of the custom specification is performed, in the refining (correcting) step 1005, correction for ensuring the quality of the shape and the like is performed, and in the refining (managing) step 1006, the quality is ensured. The products that have undergone care and finished all processing are placed in the warehouse 1007. A product cut to the size of the product order information is called a “plate”.

  Here, assuming that the period from the start of manufacturing in the converter process 1001 to the placement of the product in the warehouse 1007 is the “manufacturing period”, the manufacturing period is the processing period of the work process before starting work and the net work period in each processing process The sum of Here, in the correction process 1005 and the care process 1006, whether or not it is necessary is determined after manufacturing in the rolling process 1003. Depending on the shape of the product, the correction process 1005 is performed again after the process in the correction process 1005. In some cases, it is difficult to accurately set the manufacturing period for each product at the time of production planning, because reprocessing such as processing in the manufacturing process and processing in the care process 1006 may be necessary.

  Therefore, at the time of production planning, the production start target time is determined by going back the standard work period for each product, but it is difficult to accurately predict the production work period for each product, so the standard work period is set with a margin. It tends to be. As a result, the manufacturing tends to precede the delivery date, leading to an increase in work in progress and inventory, which may cause a variation in the manufacturing period.

  One of the objects of each embodiment of the present invention is that, as described above, an appropriate manufacturing period that reflects the actual manufacturing condition is set for each product type, and a plan for obtaining a target delivery time achievement rate without excessively increasing the work in progress is made. At the same time, it is possible to evaluate the quality of the production plan in advance by predicting the delivery date achievement rate of the planned production plan before the start of production.

<Manufacturing period model creation apparatus and method of first embodiment>
FIG. 1 shows a schematic configuration of the manufacturing period model creating apparatus according to the first embodiment of the present invention, and FIG. 2 shows each step of the manufacturing period model creating method performed using the manufacturing period model creating apparatus.

For example, in the input unit 1 including information input means such as a keyboard, an electronic file drive, and a network I / O, at least the product size, the mass, and the delivery date of a manufactured product already manufactured in the manufacturing factory are used as product attributes. The product order record information to be included and the production record information including at least the manufacturing period record of the manufactured product and the record of the presence or absence of processing in the process are input (step S1 in FIG. 2).
An example of product order record information is shown in FIG. 12, and an example of production record information is shown in FIG.
As shown in the product order record information of FIG. 12, for example, the product 1 has the standard A1, the production specification 1 “Yes”, the product thickness 12 [mm], the product width 1,000 [mm], and the product length 2,000. [Mm] indicates that the delivery date is 10 [day]. Moreover, as shown in the manufacturing performance information of FIG. 13, for example, the product 1 has a manufacturing work history of 14 [day], and represents that the finishing process is performed only in the cutting process. That is, the product order record information and the production record information are the results accumulated for each manufactured product, and the production order record information represents information about the ordered product, and the manufacture record information manufactures the product. Represents actual performance data.

  Then, the product type classification logic creation unit 2 creates product type classification logic based on at least one of the product order record information and the production result information (step S2 in FIG. 2). The type classification logic is stored in a format such as (step S3 in FIG. 2). The type classification logic created and stored here is a logic for classifying a product into a plurality of types when product order record information on an already manufactured product or product order information on a product to be manufactured is input. This type classification logic is created by this type classification logic creation unit 2 based on at least one of product order record information and production record information for products already manufactured at the manufacturing factory. Here, an already manufactured product corresponding to information used when creating this logic is referred to as an “already manufactured product”, and a product to be manufactured is also referred to as a “manufactured product”. However, the names of manufactured products and products to be manufactured are for convenience of explanation and do not limit information for each product. In other words, the product order record information and the manufacture record information on the already manufactured product may be information that is calculated or set experimentally or trially as well as the product actually created in the manufacturing factory. Moreover, the information produced based on the track record in the same kind of manufacturing factory may be sufficient. The product to be manufactured means a product for which the manufacturing period is to be predicted, and it is not necessary to actually manufacture the product. In addition, in order to confirm the authenticity of the model, the already manufactured product becomes the product to be manufactured. It can be. The “product order information” for the product to be manufactured is information including at least the product size, mass, and delivery date of the product to be manufactured as product attributes, similarly to the product order record information for the manufactured product.

  As an example of the method for creating the product type classification logic, there is a method in which, for example, the product order record information standard itself is used as a product type, or the presence or absence of the manufacturing specification 1 in the product order record information is used as a product type. Further, the product type classification logic may be defined using a plurality of product order record information. An example of the product type classification logic using a plurality of product order record information is shown in FIG. The type 1 indicates that the product thickness is 25 [mm] or less, the standard is “A1”, and the product width is 2000 [mm] or less. That is, the product type classification logic creation unit 2 creates the product type classification logic by setting the product type for each piece of information or a combination of a plurality of information included in the product order record information. Therefore, the type classification logic created can specify the type of the product to be manufactured based on the same one information or a combination of a plurality of information included in the product order information.

  Then, the product classifying unit 4 classifies the already manufactured products into products using the product order record information in accordance with the product classifying logic stored in the product classifying logic storage unit 3 (step S4 in FIG. 2).

  Then, the manufacturing period model creation unit 5 uses at least the manufacturing performance information of the already manufactured products classified by the type classification unit 4 to create a manufacturing period model for each type with the type as an input and the manufacturing period as an output ( In step S5 in FIG. 2, the manufacturing period model storage unit 6 stores the manufacturing period model in the form of a database, a file, or the like (step S6 in FIG. 2).

  The manufacturing period model creation unit 5 according to the present embodiment creates a manufacturing period model for each product type directly from the manufacturing performance information classified for each product type by the product type classification logic as a first example of the manufacturing method model creation method. Use the method you want. FIG. 15 shows a first example of a manufacturing period model in which the type is input and the manufacturing period is output. First, as described above, the product classification unit 4 classifies the already manufactured products according to the product types defined by the product type classification logic illustrated in FIG. 14 based on the product order record information. Then, by calculating the average value, standard deviation, maximum value, and mode value of the manufacturing period results included in the manufacturing result information classified for each type, the type is input, and the average value and standard deviation of the manufacturing period results are input. Create a manufacturing period model that outputs the maximum and mode values.

  In addition, the case where the 1st example of the creation method was used was demonstrated in the manufacturing period model creation by the manufacturing period model creation part 5 which concerns on this embodiment. However, the manufacturing period model creation method is not limited to the first example, and may be a method of defining a predetermined group. Then, next, the case where the 2nd example of the manufacturing construction period model creation method using this group is used as 2nd embodiment of this invention is demonstrated.

<Manufacturing period model creation apparatus and method of second embodiment>
The manufacturing construction model creation unit 5 according to the second embodiment of the present invention, as a second example of the manufacturing construction model creation method, is based on the product order record information as well as the production record information of the manufactured products. Uses a production lead model creation method by defining another group. More specifically, first, the manufacturing construction model creation unit 5 defines a group different from the product type. Then, the manufacturing period model creation unit 5 calculates the product type and the group generation rate for each product type from the product order record information and the manufacture record information. At the same time, the manufacturing period model creation unit 5 calculates the group manufacturing period model for each group from the manufacturing performance information, and the linear sum of the group manufacturing period model weighted by the group generation rate by type You may make it produce the manufacturing period model according to a kind. FIG. 16 shows an example of the group generation rate by product type, and FIG. 17 shows an example of the manufacturing period model by group with the group as an input and the manufacturing period as an output. FIG. 17 illustrates an example in which the average value of the manufacturing period results is used as the manufacturing period model. Here, the manufacturing period of the product 1 in the case illustrated in FIGS. 16 and 17 is calculated by the following equation.

Model 1 manufacturing period model = class 1 group 1 incidence x group 1 manufacturing period + class 1 group 2 incidence x group 2 manufacturing period + class 1 group 3 incidence x group 3 manufacturing period + class 1 group 4 incidence x Group 4 manufacturing period = 0.5 x 24 + 0.2 x 15 + 0.1 x 10 + 0.2 x 20
= 20 (Formula 1)

  In the second example of the manufacturing period model creation method according to the present embodiment, it is possible to create the manufacturing period model of varieties 2, 3 and the like by the same calculation.

  Note that the group for the manufactured product defined in the second example is a classification different from the product type, and represents a group in which the product can be classified by predetermined information regarding the product itself or the manufacturing process. Examples of this group include varieties such as processing process patterns expressed by combinations of presence / absence of processing in a plurality of processes, components of products manufactured in manufacturing processes (such as steel output components) and standards, etc. Includes different product attributes. Below, for convenience of explanation, a specific example of the manufacturing process model creation device of the present embodiment in the case where processing process patterns are used as this group will be described. However, when another group is used, it can be implemented by replacing the “processing step pattern” with another group, which will be described below.

  FIG. 3 shows a schematic configuration of the manufacturing period model creating apparatus according to the second embodiment of the present invention, and FIG. 4 shows each step of the manufacturing period model creating method performed using the manufacturing period model creating apparatus. 3 and 4 is based on the manufacturing period model creation device of the first embodiment, and the manufacturing period model creation unit 5 further includes a processing process pattern. It includes a setting unit 51, a type-specific processing process pattern occurrence rate generation unit 52, a type-specific manufacturing process model creation unit 53, and a type-specific manufacturing process model creation unit 54. Therefore, in the following, this new configuration will be mainly described, and description of the same configuration as the first embodiment will be simply touched or omitted.

  For example, in the input unit 1 including information input means such as a keyboard, an electronic file drive, and network I / O, product order record information and manufacturing record information are input (step S1 in FIG. 4).

  The product type classification logic generation unit 2 generates product type classification logic based on at least one of the product order record information and the production result information (step S2 in FIG. 4). The product type classification logic storage unit 3 stores a database, a file, and the like. The product type classification logic is stored in a format (step S3 in FIG. 4).

  Then, the product classification unit 4 classifies the already manufactured products into products using the product order record information according to the product classification logic stored in the product classification logic storage unit 3 (step S4 in FIG. 4).

  And the process pattern setting part 51 is an example of a group setting part, and sets a process pattern from the manufacture results information of manufactured products based on the results of the process for each product (step S51 in FIG. 4). . This process pattern is represented by a combination of presence / absence of a process of a product. Specifically, for example, the processing process pattern setting unit 51 refers to manufacturing performance information of an already manufactured product, and determines whether or not each of the processing processes of cutting, correction 1, correction 2, and care, which are the adjustment processes, is processed. The result value is 1 when processed, 0 when not processed, and a process step pattern is expressed by a combination of the presence or absence of processing in each processing step for all the steps. Then, the process step pattern setting unit 51 extracts a plurality of different process step patterns from the manufacturing performance information of the plurality of manufactured products. It is desirable that the processing process pattern setting unit 51 identifies and extracts all processing process patterns when possible. Such a process pattern for an already manufactured product is also referred to herein as a “process pattern result”. FIG. 18 shows an example of the processing process results for each product and the set processing process pattern. For example, as shown in FIG. 18, since the product 1 is processed only in the cutting process, the processing process pattern is “1000”.

  The type-specific process step pattern generation rate creation unit 52 is an example of a type-specific group generation rate creation unit, and the rate of occurrence of each type of process step pattern classified by the type classification unit 4 based on manufacturing order record information. Is calculated. More specifically, the product-specific processing process pattern occurrence rate creation unit 52 further classifies product order performance information for manufactured products classified by product type by the product classification unit 4 for each processing process pattern, As the sum of the order mass for each type of process step pattern (also referred to as order weight, for example, the number of products in addition to the mass), the order mass for each type of process step pattern is calculated. Then, by dividing the order mass by the corresponding process step pattern by product type by the sum of the order mass by product type, the process rate pattern generation rate by product type is calculated (step S52 in FIG. 4). FIG. 19 shows an example of the order mass for each type of processing process pattern, and FIG. 20 shows an example of the rate of occurrence of the processing process pattern for each type. In FIG. 19, the processing process pattern occurrence rate of product type 1 in FIG. 20 is calculated by dividing the order mass for each product process pattern of product type 1 by the total order mass of product 1 of 100 [ton]. 80 [%] of the order mass belonging to the product type 1 indicates that the processing process pattern is “0000”, “1000” is 10 [%], and the others are 5 [%].

  And the manufacturing process model creation part 53 classified by process process pattern is an example of the manufacturing process model model creation part classified by group, and produces the manufacturing process model for every process process pattern based on manufacturing performance information (step S53 of FIG. 4). ). As an example of a manufacturing process model for each processing process pattern, with the processing process pattern as an input and the required manufacturing process period as an output, for example, the average value, standard deviation, and maximum value of the manufacturing process results of the product group for each processing process pattern Statistical values such as mode values may be set. FIG. 21 shows an example of a manufacturing process model for each process pattern.

  Then, the product-specific manufacturing process model creation unit 54 calculates the weighted linear sum of the process process pattern-specific manufacturing process model using the product-specific process process pattern occurrence rate as a weight (step S54 in FIG. 4). The manufacturing period model storage unit 6 stores the manufacturing period model in the form of a database, a file, or the like (step S6 in FIG. 4). Formula 2 shows an example of creating a production period model for product type 1 when using an average value as the production period model.

Production period of product type 1 = "0000" work period x "0000" occurrence rate + "1000" work period x "1000" occurrence rate + "1010" work period x "1010" occurrence rate + "1111" work period x "1111" occurrence rate ... (Formula 2)

  In addition, when the numerical example of FIG.20 and FIG.21 is applied to the said Formula 2, and the manufacturing period for the kind 1 is calculated, it will become the following Formula 3. FIG. However, in the example of this Formula 3, the average value of the results of the manufacturing period was used as the necessary manufacturing period calculated by the manufacturing period model for each processing period pattern. When other groups (such as steel output components and standards) are used, data as shown in FIG. 20 and FIG. 21 for that group is created and numerical examples of the data are applied to Equation 1 to obtain Equation 2 An equivalent production period calculation formula can be obtained.

Production period of product type 1 = 0.8 x 5 + 0.1 x 8 + 0.05 x 10 + 0.05 x 15
= 6.05 (Formula 3)

<Manufacturing period model creation apparatus and method of third embodiment>
FIG. 5 shows a schematic configuration of the manufacturing period model creating apparatus according to the third embodiment of the present invention, and FIG. 6 shows each step of the manufacturing period model creating method performed using the manufacturing period model creating apparatus. 5 and 6 is based on the manufacturing period model creation device of the first embodiment or the second embodiment, and further includes a kind classification logic creation unit. 2 includes an explanatory variable setting unit 21, an objective variable setting unit 22, a design parameter input unit 23, and a decision tree creation unit 24. Therefore, in the following, this new configuration will be mainly described, and description of the configuration similar to the first embodiment or the second embodiment will be simply touched or omitted.

  For example, in the input unit 1 including information input means such as a keyboard, an electronic file drive, and a network I / O, product order record information and manufacturing record information are input (step S1 in FIG. 6).

  The explanatory variable setting unit 21 sets one or a plurality of product attributes included in the product order record information as explanatory variables (step S21 in FIG. 6). FIG. 12 shows an example of product attributes.

  Moreover, the objective variable setting part 22 sets the process process pattern demonstrated in said 2nd embodiment produced based on the manufacture performance information of a manufactured product as an objective variable (step S22 of FIG. 6). Since the manufacturing lead time has a high correlation with the processing step pattern, products having similar processing step patterns can be grouped as varieties, whereby products having similar manufacturing lead times can be grouped as varieties.

  Then, the design parameter setting unit 23 sets design parameters necessary when creating the product type classification logic (step S23 in FIG. 6).

Here, a decision tree construction method used to create the kind classification logic will be described.
A decision tree is one of data analysis methods, and is an analysis method for classifying data like tree branches and leaves according to various conditions as shown in FIG. It is used for information classification and the like (for example, see Non-Patent Document 1: “Basics and Mechanism of Multivariate Analysis That Can Be Elucidated”, Kazunori Yamaguchi et al., Hidekazu System, (2004) pages 143-168). 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.

  In the design parameter setting unit 23, parameters relating 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 as design parameters. In this example, it is assumed that 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 example, the upper limit value of the number of data is 100. However, the upper limit value of the number of data is not limited to this example, and it is desirable that the upper limit value is appropriately determined in consideration of prediction accuracy and ease of handling. Also, the design parameter is not limited to the upper limit value, and may be another parameter.

  Then, the decision tree creation unit 24 creates a variety classification logic by the above-described decision tree construction method based on the explanatory variable, the objective variable, and the design parameter (step S24 in FIG. 6).

  The product classification logic storage unit 3 stores the product classification logic in the form of a database, a file, or the like (step S3 in FIG. 6).

  An example of a decision tree constructed by the decision tree construction method is shown in FIG. For example, a leaf including many products having a process pattern of “0000” includes a product group having a product width of 10 mm or less, a product width of over 800 mm, and a standard of “A2”. This indicates that the product width is a product group exceeding 10 [mm] and not more than 32 [mm], a product width exceeding 800 [mm], and a standard other than “A1”. In addition, “others” is a set of objective process variables that aggregate process process patterns with a small number of appearances in the process pattern, and is set for the purpose of suppressing an increase in the number of varieties. The other types of the product classifying unit 4, the manufacturing period model creation unit 5, the manufacturing period model storage unit 6, and the subsequent creation flow (S4, S5, S6 in FIG. 6) are the first implementation of the present invention. This is the same as the embodiment or the second embodiment.

<Manufacturing period model creation apparatus and method of fourth embodiment>
The manufacturing period model created by the manufacturing period model generation apparatus according to the fourth embodiment of the present invention is input with the type or processing process pattern as an input, and the frequency distribution of the manufacturing period (may be a probability distribution) is output. An example is shown in FIG. However, in FIG. 23, the frequency distribution is shown as a normalized value (probability distribution).

  Here, the manufacturing period model of FIG. 23 is a manufacturing period model corresponding to a certain product type or processing process pattern, and the manufacturing time model as illustrated in FIG. 23 is created for the number of types or processing patterns. In the first to third embodiments, as shown in FIGS. 15, 17, and 21, a manufacturing period model that inputs a product type or a processing process pattern and outputs an average value or a mode value of the manufacturing period results. However, in the present embodiment, as illustrated in FIG. 23, a manufacturing period model is created from manufacturing period result information, with the type or processing process pattern as input and the occurrence frequency distribution of the required manufacturing period as output. There is a feature to do. By creating a model as an occurrence frequency distribution, it is possible to have information on manufacturing period variations by product type or processing process pattern.

  It should be noted that the manufacturing period model creation device of this embodiment is configured by changing the average value or mode value of the manufacturing period results in the first to third embodiments into the frequency distribution of the necessary manufacturing period. Therefore, detailed description of each configuration of the manufacturing period model creation device is omitted.

<Manufacturing period model creation apparatus and method of fifth embodiment>
FIG. 24 shows an example of a manufacturing period model that is input by the type or processing process pattern created by the manufacturing period model creating apparatus of the fifth embodiment of the present invention and that outputs the manufacturing period. However, in FIG. 24, the frequency distribution is shown as a normalized value (probability distribution).

  Here, the manufacturing period model of FIG. 24 is a manufacturing period model corresponding to a certain product type or processing process pattern, and the manufacturing time model as illustrated in FIG. 24 is created for the number of product types or the number of processing patterns. In the fourth embodiment, the occurrence frequency distribution of the necessary production period is directly used as the production period model. However, in the fifth embodiment, the occurrence period distribution of the necessary production period is approximated by some function. There is a feature in the model. The line graph shown in FIG. 24 is obtained by approximating the occurrence frequency distribution shown in FIG. Here, the lognormal distribution is a statistical distribution in which the logarithmic value of the random variable has a normal distribution, and is defined by the distribution represented by Equation 4. Here, x represents a random variable, σ represents a standard deviation of the random variable, and μ represents an average of the random variable.

... (Formula 4)

  FIG. 24 represents a lognormal distribution with an average of 1.5 and a standard deviation of 0.7. In addition, a logarithmic normal distribution frequency distribution in which the manufacturing period is a random variable, the average is 1.5, and the standard deviation is 0.7 and the cumulative frequency distribution that is the accumulation are numerically shown in the manufacturing period model column of FIG. In the example of FIG. 26, the manufacturing period model is configured by 21 random variables and their frequency distribution. However, when the lognormal distribution is used, the frequency distribution can be expressed by only two parameters of average and standard deviation. By approximating with functions in this way, the amount of information in the manufacturing period model database can be reduced.

  Further, by changing the occurrence frequency distribution of the required manufacturing period in the fourth embodiment to an approximate function, it is possible to configure the manufacturing period model creating apparatus of the present embodiment. Detailed description of the configuration and the like is omitted.

  As in the fourth embodiment and fifth embodiment, even when the actual value of the occurrence frequency distribution and its function approximation are used as the manufacturing period model for each process pattern, the process pattern is generated for each product type. By superimposing the manufacturing period models for each processing process pattern using the rate as a weight, a manufacturing period model for each product type can be created.

<Delivery Time Achievement Prediction Device and Method of Each Embodiment>
FIG. 7 shows a schematic configuration of the delivery date achievement rate prediction apparatus of each embodiment of the present invention, and FIG. 8 shows each step of the delivery date achievement rate prediction method implemented using the delivery date achievement rate prediction apparatus. In addition, this delivery date achievement rate prediction device is a delivery rate achievement rate that is a ratio of the order quantity completed by the delivery date to the total order quantity using the production period model creation device described in the first to fifth embodiments. Predict. The delivery date achievement rate prediction apparatus will be described as follows.

  For example, the input unit 100 including information input means such as a keyboard, an electronic file drive, and a network I / O inputs product order information and production plan information of a product to be manufactured that is a delivery date prediction target (FIG. 8). Step S101). The production plan information is information including at least the scheduled production start time of the product to be manufactured.

  The order data classification unit 101 reads out the product type classification logic from the product type classification logic storage unit 3 of the manufacturing period model creation device, and determines the product to be manufactured for each product type based on the product order information and the read product type classification logic. (Step S102 in FIG. 8).

  The order quantity calculation unit 102 according to the work period margin period for each product type is the difference between the delivery date of the product order information of the products scheduled for manufacture classified by product type by the order data classification unit 101 and the production start time of the production plan information. A margin period is calculated. Further, the order quantity calculation unit 102 by construction period margin period by product type further orders from the product order information by product type and by construction period margin period (hereinafter also referred to as “order quantity by construction period margin period”). Is calculated (step S103 in FIG. 8). FIG. 25 shows an example of the order quantity for each work period margin period by product type.

  Then, the delivery date achievement rate calculation unit 103 by product type reads the manufacturing time model from the manufacturing time model storage unit 6 of the manufacturing time model creation device, and delivers the delivery date by product type based on the manufacturing time model and the order quantity by product time margin period. An achievement rate is calculated (step S104 in FIG. 8). More specifically, the delivery date achievement rate calculation unit 103 by product type calculates a manufacturing work period (not only an average value etc., but also a frequency distribution or an approximate function) from one manufacturing work time model. Then, the delivery date achievement rate calculation unit 103 by product type calculates the total amount that would have been calculated by the delivery date of the product order information based on the manufacturing period, and calculates the ratio of the total amount to the total order amount. The delivery date achievement rate calculation unit 103 for each product performs such calculation for each product, thereby calculating the delivery rate achievement rate for each product.

  Here, an example of calculating the delivery rate achievement rate by product type will be described with reference to FIG. However, here, the case where the manufacturing period model calculates the frequency distribution of the manufacturing period will be described as an example. The manufacturing period in the manufacturing period model column of FIG. 26 corresponds to the manufacturing period that is the horizontal axis of FIG. 23 or FIG. 24, and the frequency distribution of FIG. 26 is the frequency distribution that is the vertical axis of FIG. It corresponds. In FIG. 26, since there are 10 [ton] orders for which the work period margin period is 0 [day] and the cumulative probability of the orders whose work period is 0 [day] is 0, the predicted value of the delivery date achievement amount is 10 [ton]. ] × 0 [ton] with a cumulative probability of 0. Similarly, there is an order with a construction period margin of 1 [day] of 100 [ton], and the cumulative probability of an order with a construction period of 1 [day] is 0.01. ] × cumulative probability of 0.01 is 1 [ton]. Thus, the predicted value of the delivery date achievement amount is calculated for all the work period margin periods, and when the total is taken, it becomes 5863 [ton], and the ratio of the predicted value of the delivery date achievement amount to the total order mass 7220 [ton] (that is, The delivery rate achievement rate by product type is calculated as 80 [%].

  Then, the delivery date achievement rate calculation unit 104 calculates the weighted linear sum of the delivery date achievement rate by product type for all the ordered products by using the order quantity by work period margin period by product type as a weight. The delivery rate achievement rate of the plan is calculated (step S105 in FIG. 8). The delivery date achievement rate storage unit 105 stores the delivery date achievement rate in the form of, for example, a database or a file (step S106 in FIG. 8).

  An example of calculating the delivery date achievement rate of the entire production plan will be described with reference to FIG. First, the ratio of the order mass for each product type to the total product order mass is obtained. Since the order mass 7220 [ton] for the product type 1 and the total product order mass is 54180 [ton], the order mass ratio is 0.13. This is calculated for all varieties, and this is used as a weight, and the delivery date achievement rate of each product type is added to obtain the delivery date achievement rate of 85% for all varieties.

  Here, the cumulative frequency distribution for each manufacturing period is calculated based on FIG. 26. However, when the manufacturing period model is created by logarithmic normal distribution approximation, the frequency distribution is calculated using Equation 4, and the cumulative value is calculated. It goes without saying that the cumulative frequency distribution can be calculated by calculating. Further, when using an average value or the like as the manufacturing period, it is possible to calculate the delivery date achievement rate from the average value and standard deviation of the normal distribution, assuming that the manufacturing period distribution is a normal distribution.

<Production start target time calculation apparatus and method of each embodiment>
FIG. 9 shows a schematic configuration of the manufacturing start target time calculation apparatus of the present invention, and FIG. 10 shows each step of the manufacturing start target time calculation method performed using the manufacturing start target time calculation apparatus. The manufacturing start target time calculation device calculates the target time to start manufacturing based on the delivery date of the order by using the manufacturing period model creation device described in the first to fifth embodiments. The delivery date achievement rate prediction apparatus will be described as follows.

  For example, in the input unit 200 including information input means such as a keyboard, an electronic file drive, and a network I / O, product order information of a product to be manufactured and a target delivery rate achievement rate for each product type to which the product is applicable. Is input (step S201 in FIG. 10).

  The order data classification unit 201 reads the type classification logic from the type classification logic storage unit 3 of the manufacturing period model creation apparatus, and classifies the products to be manufactured for each type based on the product order information and the type classification logic (see FIG. 10 step S202).

  Then, the manufacturing lead time calculation unit 201 reads the manufacturing lead time model for each product type from the manufacturing lead time model storage unit 6 of the manufacturing lead time model creation device, and sets the manufacturing lead time model for each product type and the target delivery date achievement rate for each product type. Based on this, the necessary manufacturing period of the product scheduled for manufacture classified for each product type is calculated (step S203 in FIG. 10). More specifically, the manufacturing period calculation unit 201 calculates a manufacturing period (not only an average value or the like, but also a frequency distribution or an approximate function) from one manufacturing period model. Then, the manufacturing lead time calculation unit 201 calculates the number of days (that is, the required manufacturing lead time) necessary to satisfy the target delivery date achievement rate based on the manufacturing lead time.

  Then, the manufacturing start target time calculation unit 203 calculates the manufacturing start target time for each product order information by going back by the necessary manufacturing work period calculated from the delivery date for each product order information (step 204 in FIG. 10), and starts the manufacturing. The target time storage unit 204 stores the manufacturing start target time in the form of a database, a file, or the like (step S205 in FIG. 10).

  A manufacturing start time model of the type to which the product order information belongs is given in FIG. 26, the delivery date is 34 [day], and the target delivery time achievement rate is 95 [%]. explain. First, it can be seen from the manufacturing period model in FIG. 26 that the period for which the cumulative frequency distribution is 95 [%] (0.95) or more is 14 [day]. Therefore, the production start target time is calculated as 20 [day] by going back 14 [day] from the delivery date.

  Here, the cumulative frequency distribution for each manufacturing period is calculated based on FIG. 26. However, when the manufacturing period model is created by logarithmic normal distribution approximation, the frequency distribution is calculated using Equation 4, and the cumulative value is calculated. It goes without saying that the cumulative frequency distribution can be calculated by calculating. Further, when using an average value or the like as the manufacturing period, it is possible to calculate the necessary manufacturing period from the average value and standard deviation of the normal distribution, assuming that the manufacturing period distribution is a normal distribution.

<Example of effects according to each embodiment of the present invention>
According to each embodiment of the present invention, by classifying products according to product type classification logic and creating a manufacturing period model for each product type based on manufacturing results, appropriate manufacturing reflecting the actual manufacturing status within practical time. The construction period can be predicted.

  Also, paying attention to the high correlation between the process pattern and the manufacturing process period, the same process can be achieved by creating a manufacturing process model for each product type as the linear sum of the process process pattern occurrence rate by product type and the manufacturing process model by process process pattern. Because it is possible to suppress the variation in the manufacturing lead time of the products included in the process pattern, further improve the prediction accuracy of the manufacturing lead time, and include the processing pattern pattern generation rate by type in the manufacturing lead time model Therefore, it is possible to predict the effect on the manufacturing period when the processing pattern of the product type is changed for each product type.

  In addition, by applying the decision tree construction method to the kind classification logic, it is possible to suppress the variation in the manufacturing period for each kind, and to further improve the prediction accuracy of the manufacturing period.

  In addition, by using a production period model for each product type based on production results, when planning production, it is possible to perform the actual production within the practical time without performing a detailed production simulation that requires calculation time. In addition, it is possible to accurately predict the delivery rate achievement rate.

  In addition, by using the production period model for each product type based on the production results, it is possible to calculate the necessary production period with no excess or deficiency in line with the actual situation, and calculate the appropriate production start target time. become able to.

  In addition, it is possible to evaluate the quality of the production plan in advance, and it is possible to make an optimal production plan that takes productivity, delivery date, cost, and the like into consideration.

  As mentioned above, although preferred embodiment of this invention was described in detail, referring an accompanying drawing, it cannot be overemphasized that this invention is not limited to this example. It is obvious that a person having ordinary knowledge in the technical field to which the present invention pertains can come up with various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that these also belong to the technical scope of the present invention.

  For example, the series of processes described in the above embodiments may be executed by dedicated hardware, but may be executed by software. When the series of processes is performed by software, the above series of processes can be realized by causing a general-purpose or dedicated computer to execute the program. The computer includes a CPU (Central Processing Unit), a recording device such as a HDD (Hard Disk Drive), a ROM (Read Only Memory), a RAM (Random Access Memory), a LAN (Local Area Network), etc. Communication devices, input devices such as a mouse / keyboard, magnetic disks such as flexible disks, optical disks such as various CDs (Compact Discs), MOs (Magneto Optical) disks, DVDs (Digital Versatile Discs), semiconductor memories, etc. Drives that read and write to removable storage media, etc., and display devices such as monitors, and audio from speakers and headphones An output device such as an output device. The computer may execute the above-described series of processes by executing a program recorded in a recording device / removable storage medium or a program acquired via a network. In addition, each of the above-described embodiments can of course be implemented as a computer-readable recording medium such as the above-described removable storage medium or recording device that records the program.

BRIEF DESCRIPTION OF THE DRAWINGS It is the figure which showed schematic structure of the apparatus in 1st embodiment of the manufacturing construction model creation apparatus of this invention. It is the flowchart which showed the outline of the manufacturing construction model creation procedure in 1st embodiment of the manufacturing construction model creation apparatus of this invention. It is the figure which showed schematic structure of the apparatus in 2nd embodiment of the manufacturing construction period model creation apparatus of this invention. It is the flowchart which showed the outline of the manufacturing construction model creation procedure in 2nd embodiment of the manufacturing construction model creation apparatus of this invention. It is the figure which showed schematic structure of the apparatus in 3rd embodiment of the manufacturing construction period model production apparatus of this invention. It is the flowchart which showed the outline of the manufacturing construction model creation procedure in 3rd embodiment of the manufacturing construction model creation apparatus of this invention. It is the figure which showed schematic structure of the apparatus in embodiment of the delivery date achievement rate prediction apparatus of this invention. It is the flowchart which showed the outline of the delivery date achievement rate prediction procedure in embodiment of the delivery date achievement rate prediction apparatus of this invention. It is the figure which showed schematic structure of the apparatus in embodiment of the manufacture start target time calculation apparatus of this invention. It is the flowchart which showed the outline of the manufacturing start target time calculation procedure in embodiment of the manufacture start target time calculation apparatus of this invention. It is the schematic of an example of the thick board manufacturing process in the steel industry. It is the figure which showed an example of product order information. It is the figure which showed an example of manufacture performance information. It is the figure which showed an example of the kind classification logic. It is the figure which showed an example of the manufacturing period model according to a kind. It is the figure which showed an example of the group generation rate according to kind. It is the figure which showed an example of the manufacturing construction period model according to group. It is the figure which showed an example of the process pattern corresponding to a process process performance. It is the figure which showed an example of the order mass according to the process process pattern classified by kind. It is the figure which showed an example of the process process pattern generation rate according to kind. It is the figure which showed an example of the manufacturing process period model according to a process process pattern. It is the figure which showed an example of the kind classification logic by a decision tree. It is the figure which showed an example of the manufacturing construction period generation frequency distribution. It is the figure which showed an example of the function approximation of the manufacturing construction period generation frequency distribution. It is the figure which showed an example of the order mass according to construction period margin period according to a kind. It is the figure which showed an example which calculates a delivery date achievement rate based on the manufacturing time model of the kind 1, and the order mass according to a work time margin period. It is the figure which showed an example which calculates the delivery date achievement rate of all the varieties.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Input part 2 Product classification logic creation part 21 Explanation variable setting part 22 Objective variable setting part 23 Design parameter setting part 24 Decision tree creation part 3 Product classification logic storage part 4 Product classification part 5 Manufacturing process model creation part 51 Processing process pattern setting Unit 52 Process-specific pattern generation rate creation part 53 Production process-period model creation part 54 Process-period production model creation part 6 Production-period model creation part 6 Production-period model storage part 100 Input part 101 Order data classification part 102 Order mass calculation unit 103 Product delivery date achievement rate calculation unit 104 Delivery date achievement rate calculation unit 105 Delivery date achievement rate storage unit 200 Input unit 201 Order data classification unit 202 Manufacturing lead time calculation unit 203 Manufacturing start target time calculation unit 204 Manufacturing start target time storage Part 1001 Converter 1002 Continuous casting 1003 Rolling 1004 Refinement (cutting)
1005 Refinement (correction)
1006 Refinement (care)
1007 Warehouse S1 Input step S2 Product classification logic creation step S21 Explanation variable setting step S22 Objective variable setting step S23 Design parameter setting step S24 Decision tree creation step S3 Product classification logic storage step S4 Product classification step S5 Production period model creation step S51 Processing step Pattern setting part step S52 Production process pattern generation rate creation step by product type Production period model creation step by step process pattern production step S54 Production period model creation step by product type Production step model storage step S101 Input step S102 Order data classification step S103 Production period by product type Order quantity calculation step by margin period S104 Delivery date achievement rate calculation step by type S105 Delivery date achievement rate calculation step S106 Delivery date achievement rate storage step S20 Input Step S202 order data classification step S203 need manufacturing period calculation step S204 manufactured undertake target timing calculation step S205 manufactured undertake target timing storage step

Claims (13)

For a manufacturing plant that manufactures multiple products of different varieties by processing in multiple processes, create a manufacturing lead model to output the manufacturing lead time of the product to be manufactured when product order information for the product to be manufactured is input A manufacturing period model creation device,
Manufacturing including at least one product order performance information including at least size, mass, and delivery date as product attributes for at least one or more ready-made products by the manufacturing factory, manufacturing work history results of the manufactured products, and results of presence or absence of processing in the process An input unit for inputting achievement information;
Based on the product order record information, a type classification logic is created for receiving the input of the product order record information or the product order information and creating a type classification logic for classifying the manufactured product or the product to be manufactured into a plurality of types. And
A product type classification logic storage unit for storing the product type classification logic created by the product type classification logic creation unit;
A product classifying unit that classifies the manufactured product into a plurality of products using the product classifying logic stored in the product classifying logic storage unit;
A manufacturing period model creating unit that creates a manufacturing period model that inputs the type classified by the type classifying unit and outputs the manufacturing period using at least the manufacturing performance information of the manufactured products classified by the type classifying unit When,
A manufacturing period model storage unit that stores the manufacturing period model created by the manufacturing period model creation unit;
An apparatus for creating a manufacturing period model characterized by comprising at least In the manufacturing construction model creation device according to claim 1,
The manufacturing construction model creation unit
A group setting unit for setting a plurality of predetermined groups different from the product type for the manufactured products from the manufacturing performance information;
Based on the production order record information, a group-by-product generation rate creation unit that calculates the rate of occurrence of each of the groups by product type classified by the product type classification unit;
Based on the manufacturing performance information, a group-specific manufacturing period model creation unit that creates a manufacturing period model that inputs the group and outputs a necessary manufacturing period; and
Create a production period model for each type by creating a production period model with the type as an input and the production period as an output by calculating a weighted linear sum of the group production period model with the group-specific occurrence rate as a weight And
A manufacturing construction model creation device characterized by comprising: In the manufacturing period model creation device according to claim 2,
The manufacturing period model creation apparatus characterized in that the predetermined group for the manufactured product set by the group setting unit is a processing process pattern represented by a combination of presence or absence of processing in the plurality of processes. In the manufacturing construction model creation device according to any one of claims 1 to 3,
The product type classification logic creation unit
An explanatory variable setting unit for setting one or more product attributes included in the product order information as explanatory variables;
A target variable setting unit that creates a predetermined group different from the product type for the ready-made product from the manufacturing performance information, and sets the group as a target variable;
A design parameter input unit for inputting design parameters when creating the product type classification logic;
Based on the explanatory variable, the objective variable, and the design parameter, a decision tree creating unit that creates a variety classification logic by a decision tree construction method;
A manufacturing construction model creation device characterized by comprising: In the manufacturing period model creation device according to any one of claims 1 to 4,
The manufacturing period model creation unit is characterized in that the manufacturing period model creation unit receives the kind classified by the kind classification unit and creates a manufacturing period model that outputs a frequency distribution of the production period for the kind. . In the manufacturing period model creation device according to any one of claims 1 to 4,
The manufacturing period model creation unit is configured to input a type classified by the type classification unit and to create a manufacturing period model that outputs a function approximating the frequency distribution of the manufacturing period for the type. Construction model creation device. A production plan planning operation in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, and is manufactured by the delivery date using the manufacturing period model creation device according to any one of claims 1 to 6. A delivery date achievement rate predicting device that predicts a delivery date achievement rate that is a ratio of the completed order quantity to the total order quantity,
The above-mentioned product order information including at least the size, mass, and delivery date of the product to be manufactured, which is a target for predicting the delivery date, and the above-mentioned production start schedule for an order for which a manufacturing start time is determined among orders for the product to be manufactured An input unit for reading production plan information including at least a time, and reading the product type classification logic from the product type classification logic storage unit of the manufacturing period model creation device, and the product to be manufactured based on the product order information and the product type classification logic The order data classifying unit that classifies by product type, and the difference between the delivery date of the product order information of the product to be manufactured classified by product type by the order data classifying unit and the scheduled production start time of the production plan information By work type margin period, and from the product order information, calculate the order quantity for each product type and each work period margin period And another order quantity calculation unit period allowance period,
Read the manufacturing period model from the manufacturing period model storage unit of the manufacturing period model creation device, and calculate the delivery date achievement rate for each type from the manufacturing period model and the order quantity for each type and for each period margin Delivery time achievement rate calculation department,
A delivery date achievement rate calculation unit for calculating a delivery date achievement rate of the production plan by calculating a weighted linear sum of the delivery date achievement rate by product type with the order quantity by the product type and the work period margin period as a weight;
A delivery date achievement rate prediction device characterized by comprising: In a production plan drafting operation in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, using the manufacturing period model creation device according to any one of claims 1 to 6, Based on the manufacturing start target time calculation device for calculating the target time to start manufacturing,
An input unit for inputting the product order information including at least the size, mass, and delivery date of the product to be manufactured, which is the target for calculating the production start target time, and the target delivery date achievement rate for each type of the product to be manufactured;
An order data classification unit that reads the type classification logic from the type classification logic storage unit of the manufacturing period model creation device, and classifies the product to be manufactured for each type based on the product order information and the type classification logic;
Read out the manufacturing period model for each type from the manufacturing period model storage unit of the manufacturing period model creation device, and based on the manufacturing period model for each type and the target delivery time achievement rate for each type, the necessity of the product to be manufactured A manufacturing period calculation unit for calculating the manufacturing period,
A production start target time calculation unit for calculating a target value of the start time of manufacture of the order by going back from the delivery date of the order by the required manufacturing period; and
A production start target time calculation apparatus comprising: For a manufacturing plant that manufactures multiple products of different varieties by processing in multiple processes, create a manufacturing lead model to output the manufacturing lead time of the product to be manufactured when product order information for the product to be manufactured is input A manufacturing period model creation method,
Manufacturing including at least one product order performance information including at least size, mass, and delivery date as product attributes for at least one or more ready-made products by the manufacturing factory, manufacturing work history results of the manufactured products, and results of presence or absence of processing in the process An input step for inputting the result information by the input unit;
Based on the product order record information, a type classification logic creation unit generates a type classification logic for receiving the input of the product order record information or the product order information and classifying the manufactured product or the product to be manufactured into a plurality of types. The kind classification logic creation step to create,
A product classification logic storage step for storing the product classification logic created in the product classification logic creation step in a product classification logic storage unit;
Using the kind classification logic stored in the kind classification logic storing step, the kind classification step for classifying the manufactured product into a plurality of kinds by the kind classification unit;
Using the manufacturing result information of the manufactured products classified in the product classification step, at least the production time model created by the production time model creation unit that inputs the product classified in the product classification step and outputs the production time A production period model creation step to
A manufacturing period model storing step for storing the manufacturing period model created in the manufacturing period model creating step in a manufacturing period model storage unit;
A production period model creation method characterized by comprising at least A production plan planning operation in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, and is manufactured by the delivery date using the manufacturing period model creation device according to any one of claims 1 to 6. A delivery date achievement rate prediction method for predicting a delivery date achievement rate that is a ratio of a completed order quantity to a total order quantity,
The above-mentioned product order information including at least the size, mass, and delivery date of the product to be manufactured, which is a target for predicting the delivery date, and the above-mentioned production start schedule for an order for which a manufacturing start time is determined among orders for the product to be manufactured An input step for reading production plan information including at least the time by an input unit, and an order data classification unit for reading the product type classification logic from the product type classification logic storage unit of the manufacturing period model creation device, and the product to be manufactured as the product Order data classification step for classifying by product type based on the order information and the product type classification logic;
The difference between the delivery date of the product order information of the product to be manufactured classified by product type in the order data classification step and the scheduled production start time of the production plan information by the order quantity calculation unit by construction period margin period by product type Calculate a certain work period margin period, and from the product order information, calculate the order quantity for each product type and each work period margin period;
The production period delivery rate achievement rate calculation unit reads the manufacturing period model from the manufacturing period model storage unit of the manufacturing period model creation device, and determines the classification according to the type from the manufacturing period model and the order quantity according to the type and the margin period. A delivery rate achievement rate calculation step for each product type to calculate the delivery rate achievement rate of
The delivery date achievement rate calculated by the delivery date achievement rate calculation unit by calculating the delivery date achievement rate of the production plan by calculating the weighted linear sum of the delivery date achievement rate by product type with the order quantity by the product type and the work period margin period as a weight A calculation step;
The delivery date achievement rate prediction method characterized by including at least. In a production plan drafting operation in a manufacturing factory that manufactures a plurality of products of different varieties by processing in a plurality of processes, using the manufacturing period model creation device according to any one of claims 1 to 6, Based on a manufacturing start target time calculation method for calculating a target time to start manufacturing,
An input step for inputting the product order information including at least the size, mass, and delivery date of the product to be manufactured, which is the target for calculating the production start target time, and the target delivery date achievement rate for each type of the product to be manufactured by an input unit;
Order data that reads the product type classification logic from the product type classification logic storage unit of the manufacturing period model creation device by the order data classification unit, and classifies the product to be manufactured for each product type based on the product order information and the product type classification logic A classification step;
The manufacturing period calculation unit reads out the manufacturing period model for each type from the manufacturing period model storage unit of the manufacturing period model creation device, and based on the manufacturing period model for each type and the target delivery time achievement rate for each type, A necessary manufacturing period calculation step for calculating a necessary manufacturing period of the product to be manufactured; and
A production start target time calculating step of calculating a target value of the start time of manufacture of the order by a manufacture start target time calculating unit by going back from the delivery date of the order by the required manufacturing period; and
The manufacturing start target time calculation method characterized by including at least. On the computer,
The program for performing each step of the manufacturing construction period model creation method in any one of Claims 9-11.   A computer-readable storage medium storing the program according to claim 12.