Manufacturing support device, manufacturing support method, and program

Abstract

To allocate more surplus material slabs to appropriate orders without limiting the possibility of allocating surplus materials to orders each having a manufacturing record.SOLUTION: A manufacturing support device for assisting in applying billets to orders includes: a quality prediction unit that predicts quality characteristic values as of when the billet becomes a product based on attribute values and processing conditions of the billet; an optimization calculation unit that optimizes the processing conditions such that the quality characteristic values predicted by the quality prediction unit satisfy specifications of quality characteristic values required in the order; and an output unit for outputting information indicating a combination of the billet and the order for which processing conditions can be calculated in the optimization calculation unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a manufacturing support device, a manufacturing support method, and a program. [Background technology]

[0002] Steel billets (slabs, billets, blooms, etc.) produced by a continuous casting machine are assigned to an order lot for each steel lot, for example, in units of 300 tons. Therefore, there may be surplus steel (hereinafter, sometimes referred to as surplus slabs) that are not assigned to an order lot within the steel lot. Since the surplus steel becomes inventory, it is allocated to lower-order orders with relatively loose requirements when possible. However, since the surplus steel was originally manufactured to meet the strict requirements of higher-order orders, if a high-cost steel billet is allocated to a lower order, a loss occurs due to the price difference. Steel billets that are not allocated to lower-order orders are used as inexpensive slabs or scrapped, so the loss becomes even larger. If more surplus steel could be allocated to higher-order orders, the loss could be reduced, but in a typical steel billet production site, surplus steel exists in units of hundreds of sheets, and orders also exist in units of hundreds, so it is difficult to manually consider whether to allocate all combinations.

[0003] In this regard, Patent Document 1 describes a method of allocating slabs to orders, which enables surplus materials to be allocated to orders quickly and in a timely manner. Specifically, slabs are classified using codes assigned based on manufacturing records, while orders are classified using codes assigned based on the specifications and dimensions of the steel plate products related to the orders. Each classification code of an order is associated in advance with a classification code of a slab that satisfies the requirements, and slabs are allocated to orders by associating the classification codes with each other. Here, the classification code of a slab is determined based on the slab shape, slab weight, chemical composition of the slab, slab width, slab location, slab maintenance method, and the presence or absence of slab defects. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2000-353007 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, even with the technology described in the above Patent Document 1, it is difficult to sufficiently reduce surplus materials not used for orders. In Patent Document 1, the classification codes of billets and orders are associated with each other based on past performance, i.e., performance of billets manufactured to satisfy the requirements of an order, so that billets manufactured under manufacturing conditions with no past performance may not have an order associated with them by the classification code, and as a result, may not be used for an order. Since there is not always one type of manufacturing condition for billets that satisfies the required characteristics of an order, a billet may satisfy the required characteristics even if there is no similar past performance, and in such a case, there is a possibility that an opportunity to use surplus materials for an order may be missed by making an association within the scope of past performance as described above.

[0006] Therefore, an object of the present invention is to provide a manufacturing support device, a manufacturing support method, and a program that are capable of allocating more surplus materials to appropriate orders, without limiting the possibility of allocating surplus materials to orders that actually have a production history. [Means for solving the problem]

[0007] According to one aspect of the present invention, there is provided a manufacturing support device that supports the allocation of steel billets to orders, the manufacturing support device comprising: a quality prediction unit that predicts quality characteristic values ​​of the steel billets when they are made into products based on attribute values ​​and processing conditions of the steel billets; an optimization calculation unit that optimizes the processing conditions so that the quality characteristic values ​​predicted by the quality prediction unit satisfy the specifications of the quality characteristic values ​​required in the order; and an output unit that outputs information indicating combinations of the steel billets and the orders for which processing conditions can be calculated by the optimization calculation unit.

[0008] According to another aspect of the present invention, there is provided a manufacturing support method for supporting the allocation of steel billets to orders, the manufacturing support method including: a quality prediction step of predicting quality characteristic values ​​of the steel billets when they are made into products based on attribute values ​​and processing conditions of the steel billets; an optimization calculation step of optimizing the processing conditions so that the predicted quality characteristic values ​​satisfy the specifications of the quality characteristic values ​​required in the order; and an output step of outputting information indicating combinations of the steel billets and the orders for which processing conditions can be calculated in the optimization calculation step.

[0009] According to yet another aspect of the present invention, there is provided a program for causing a computer to function as a manufacturing support device that supports the allocation of steel billets to orders, the manufacturing support device comprising: a quality prediction unit that predicts quality characteristic values ​​of the steel billets when they are made into products based on attribute values ​​and processing conditions of the steel billets; an optimization calculation unit that optimizes the processing conditions so that the quality characteristic values ​​predicted by the quality prediction unit satisfy the specifications of the quality characteristic values ​​required in the order; and an output unit that outputs information indicating combinations of the steel billets and the orders for which processing conditions can be calculated by the optimization calculation unit.

[0010] According to the above-described configuration, whether or not a steel billet can be allocated to an order is determined by solving an optimization problem using a quality prediction model. Therefore, the possibility of allocating surplus materials to orders can be determined broadly, not just for materials that have actual manufacturing records, and more surplus materials can be allocated to appropriate orders. [Brief description of the drawings]

[0011] [Figure 1] 1 is a diagram showing a schematic configuration of a manufacturing support device according to an embodiment of the present invention; [Diagram 2] FIG. 11 is a diagram illustrating an example of manufacturing performance data. [Diagram 3] FIG. 4 is a diagram showing an example of billet data. [Figure 4] FIG. 13 is a diagram showing an example of non-production order data. [Diagram 5]FIG. 11 is a diagram for explaining extraction of combination candidates. [Figure 6] FIG. 11 is a diagram for explaining extraction of combination candidates. [Figure 7] FIG. 11 is a diagram for explaining extraction of combination candidates. [Figure 8] FIG. 11 is a diagram for explaining extraction of combination candidates. [Figure 9] FIG. 11 is a diagram for explaining extraction of combination candidates. [Figure 10] FIG. 11 is a diagram showing examples of extracted combination candidates. [Figure 11] FIG. 13 is a diagram illustrating an example of constraint conditions. [Figure 12] FIG. 13 is a diagram illustrating an example of an evaluation function. [Figure 13] FIG. 13 is a diagram showing an example of a display of a matching result between a steel billet and an order. [Figure 14] 1 is a flowchart showing a schematic process procedure of a manufacturing support method according to an embodiment of the present invention. [Figure 15] FIG. 13 is a diagram showing a schematic configuration of a manufacturing support device according to a modified example of an embodiment of the present invention. [Figure 16] 1 is a graph showing the results of applying actually existing surplus materials to unproduction orders as an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In this specification and the drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0013] Fig. 1 is a diagram showing a schematic configuration of a manufacturing support device according to an embodiment of the present invention. The manufacturing support device 100 shown in Fig. 1 is configured by an information processing device such as a large-scale computer such as a mainframe or an open system, or a personal computer. The manufacturing performance data 102, the quality prediction model 106, the steel piece data 108, and the non-production order data 110 are stored in various memories such as ROM and RAM, a hard disk, or a recording medium such as a CD-ROM. Alternatively, some or all of these data may be stored in another information processing device and transmitted to the manufacturing support device 100 by communication using an electric communication line such as a LAN or the Internet.

[0014] The functions of the manufacturing performance data acquisition unit 101, quality prediction model construction unit 103, quality prediction model storage unit 105, billet data acquisition unit 107, order data acquisition unit 109, billet-order combination candidate extraction unit 111, constraint condition setting unit 113, evaluation function setting unit 115, optimization calculation unit 117 and matching result display unit 119 included in the manufacturing support device 100 are realized by the processor of an information processing device (computer) operating in accordance with a program stored in a memory or recording medium.

[0015] Note that each unit of the manufacturing support apparatus 100 does not have to be realized by a single information processing device, and may be realized by distributing it among a plurality of information processing devices. Specifically, for example, the manufacturing performance data acquisition unit 101, the quality prediction model construction unit 103, and the quality prediction model storage unit 105, and the billet data acquisition unit 107, the order data acquisition unit 109, the billet-order combination candidate extraction unit 111, the constraint condition setting unit 113, the evaluation function setting unit 115, the optimization calculation unit 117, and the matching result display unit 119 may each be realized by a different information processing device.

[0016] In this embodiment, the manufacturing support device 100 executes a process of filling orders with surplus steel billets (slabs, billets, blooms, etc.) that are intermediate products, which are manufactured by a steelmaking process and a rolling process. Specifically, the manufacturing performance data acquisition unit 101 acquires manufacturing performance data 102. Here, FIG. 2 is a diagram showing an example of the manufacturing performance data 102. As shown in FIG. 2, the manufacturing performance data 102 corresponds to design variables indicating attribute values ​​such as components and processing conditions such as rolling methods, and actual values ​​of quality variables indicating quality characteristic values ​​such as TS (tensile strength), YP (yield point), and EL (elongation), as well as product attributes such as the steel type of the product determined from these actual values, for each product (product ID) that has been manufactured in the past. Note that the design variables may include variables other than the attribute values ​​and processing conditions, as long as they are variables related to the quality design of the product. Similarly, the quality variables may include variables other than mechanical property values ​​such as TS as long as they relate to product quality, and may include variables such as flatness and the presence or absence of defects.

[0017] Based on the acquired manufacturing performance data 102, the quality prediction model construction unit 103 constructs a quality prediction model that predicts the quality characteristic value (e.g., mechanical characteristic value such as TS) of a previously manufactured steel billet from its attribute value (e.g., composition) and processing condition (e.g., rolling method) when the steel billet is made into a product. Here, the processing condition includes the manufacturing condition in the post-process for turning the steel billet (e.g., slab) into a product, specifically, in the process such as rolling and heat treatment. In this embodiment, the quality prediction model is constructed by a Bayesian estimation method, and predicts the probability distribution of the quality characteristic value from the attribute value and processing condition of the steel billet. When there are multiple quality variables, multiple quality prediction models may be constructed, or a single quality prediction model capable of predicting the probability distribution of each of the multiple quality variables may be constructed. The quality prediction model storage unit 104 stores the quality prediction model 106 constructed by the quality prediction model construction unit 103 in the memory of the manufacturing support device 100 or a removable recording medium.

[0018] In this embodiment, the quality prediction model construction unit 103 constructs a quality prediction model using a known Gaussian process regression method, which is a type of Bayesian estimation method. Here, the Bayesian estimation method is a method for estimating the average value and standard deviation of the output y, assuming that the coefficients a and b also have uncertainty (a to N(μ, σ)) when the relationship between the output y and the input x is expressed in the form of y=ax+b. In the Gaussian process regression method, the output (quality variable) y is estimated based on the input (design variable) x=[x1, , x N ] covariance matrix K=Φ(x)Φ(x) on the feature space (feature vector Φ(x)) T =[k(x i ,x j )] (i,j) Under the assumption that the distribution follows a Gaussian distribution, the new data point x is calculated using the following equation (1): * Output of y * The kernel basis function k(x, x') uses a radial basis function (RBF) as shown in equation (2), and optimizes the hyperparameter θ for each input variable by variational inference. Note that functions other than RBF may also be used as the kernel basis function.

[0019]

number

[0020] For example, when production performance data 102 relating to a predetermined number of products has been accumulated, a quality prediction model 106 is constructed and stored by the functions of the production performance data acquisition unit 101, the quality prediction model construction unit 103, and the quality prediction model storage unit 105. Thereafter, at any time, such as when a new order is generated, the billet data acquisition unit 107, the order data acquisition unit 109, the billet-order combination candidate extraction unit 111, the constraint condition setting unit 113, the evaluation function setting unit 115, the optimization calculation unit 117, and the matching result display unit 119 output information useful for allocating surplus materials to unmanufactured orders.

[0021] The billet data acquiring unit 107 acquires billet data 108. FIG. 3 is a diagram showing an example of the billet data 108. As shown in FIG. 3, the billet data 108 associates the steel type, weight, components, number of days in stock, and alloy cost with each surplus material (bill ID) that is in stock at the time of processing. The number of days in stock is the number of days that the billet stays in stock after it is manufactured. The alloy cost is the manufacturing cost per unit weight of the alloy that constitutes the billet. The order data acquiring unit 109 acquires unmanufactured order data 110. FIG. 4 is a diagram showing an example of the unmanufactured order data 110. As shown in FIG. 4, the unmanufactured order data 110 associates the billet data 110 with the remaining number of invoices, unit weight and plate thickness, component specifications, and quality characteristic value specifications that are required as final products with each order (order ID) that is not manufactured at the time of processing or that is not manufactured and the allocation of billets is not yet determined. Among these, the component specifications prescribe the content of elements such as C and Mn, carbon equivalent, etc., and the quality characteristic value specifications prescribe mechanical characteristic values ​​such as TS, YP, and EL, etc. These specifications may be prescribed by a range determined by both an upper limit value and a lower limit value, as in the illustrated example, or may be prescribed by a range determined by either an upper limit value or a lower limit value alone.

[0022] The billet-order combination candidate extraction unit 111 extracts combination candidates that are relatively likely to satisfy the order specifications from the viewpoints of (1) unit weight and remaining invoice quantity, (2) component specifications, and (3) quality characteristic values, which will be described below, for combinations of surplus billets included in the billet data 108 and orders included in the unmanufactured order data 110. As already described, there can be several hundred pieces of surplus billets and orders, respectively, so by preferentially extracting possible combination candidates, the processing load of the subsequent optimization calculation unit 117 and the like can be reduced.

[0023] (1) Unit weight and billing balance The billet-order combination candidate extraction unit 111 calculates the remaining weight (= billing balance x unit weight) that needs to be produced for each order ID in the unproduction order data 110, and extracts candidate combinations of billets and orders in which the weight of the billets recorded in the billet data 108 is equal to or greater than the remaining weight of the order. Here, Figs. 5, 6, 7, 8, and 9 are diagrams for explaining the extraction of candidate combinations. As in the example shown in Fig. 5, the combination of order ID "order1" and billet ID "slab1" is a candidate because the remaining weight of the order (20) is less than the weight of the billet (22), but the combination of order ID "order2" and billet ID "slab1" is not a candidate because the remaining weight of the order (30) exceeds the weight of the billet (22).

[0024] (2) Ingredient specifications The billet-order combination candidate extraction unit 111 refers to the component specifications (upper and lower limits of the content of components such as C and Mn, carbon equivalent, etc.) for each order ID in the unmanufactured order data 110, and extracts candidate combinations of billets and orders whose components recorded in the billet data 108 satisfy the component specifications of the order.

[0025] (3) Quality characteristic values The billet-order combination candidate extraction unit 111 predicts possible quality characteristic values ​​of each of the surplus billets included in the billet data 108 based on past production performance data (which may be the same data as the above-mentioned production performance data 102, or may be different data). Specifically, for example, it predicts the range of the quality characteristic values ​​based on the average value and standard deviation calculated by either of the prediction methods (3-1) or (3-2) described below.

[0026] (3-1) Prediction method 1 The billet-order combination candidate extraction unit 111 calculates the average value and standard deviation required to predict the possible range of quality characteristic values ​​(TS (tensile strength), YP (yield point), and EL (elongation)) for each type of billet, as shown in Fig. 6, based on past production performance data. For example, the billet-order combination candidate extraction unit 111 may collectively calculate the average value and standard deviation for each of the same steel type (A, B, etc.) for the quality characteristic values ​​for each billet (product ID) included in the past production performance data 102 shown in Fig. 2. This makes it possible to predict the average value and standard deviation of each quality characteristic value of the surplus material (for each billet ID) from the steel type of the surplus material included in the billet data 108 in Fig. 3.

[0027] (3-2) Prediction method 2 The billet-order combination candidate extraction unit 111 constructs a prediction model having explanatory variables and objective variables as shown in FIG. 7 based on past production performance data. In this prediction model, the explanatory variables correspond to the component contents and carbon equivalent, and the objective variable is the probability distribution of quality characteristic values. That is, in this case, the prediction model predicts the probability distribution of quality characteristic values ​​for the components of the billet. Unlike the above-mentioned quality prediction model, the prediction model used for prediction here does not use the processing conditions of the billet to predict the quality characteristic values. However, like the quality prediction model, it is possible to construct a prediction model using, for example, a Bayesian estimation method. If the prediction model is constructed in this way, it is possible to predict the average value and standard deviation of each quality characteristic value of the surplus material (for each billet ID) from the components of the surplus material included in the billet data 108 in FIG. 3 based on the prediction model.

[0028] 8 for TS (tensile strength), the billet-order combination candidate extraction unit 111 calculates the possible range of the quality characteristic value, specifically, the "average value ±α × standard deviation", from the average value and standard deviation of the quality characteristic value for each billet ID predicted by the above prediction method 1 or prediction method 2. For example, when α = 1.960, it is possible to calculate the possible range of the quality characteristic value with a probability of 95%.

[0029] The billet-order combination candidate extraction unit 111 compares the possible range of quality characteristic values ​​for each billet ID calculated by either or both of (3-1) and (3-2) above with the specifications of the quality characteristic values ​​for each order ID included in the unproduction order data 110 in FIG. 4, and extracts billet and order combination candidates whose predicted ranges overlap with the specifications of the quality characteristic values ​​of the order. An example of combination candidate extraction by such a determination is shown in FIG. 9. The combination of billet ID "slab1" and order ID "order1" is a candidate because the ranges of the predicted quality characteristic values ​​(TS (tensile strength), YP (yield point), and EL (elongation)) all overlap with the specifications of the quality characteristic values ​​of the order, but the combination of billet ID "slab1" and order ID "order2" is not a candidate because the range of TS (tensile strength) among the quality characteristic values ​​does not overlap with the specifications of the quality characteristic values ​​of the order.

[0030] 10 is a diagram showing examples of combination candidates extracted by the billet-order combination candidate extraction unit 111. Note that when the value of α used by the billet-order combination candidate extraction unit 111 is large, the number of combination candidates increases and the search region becomes wider, but the processing load for solving the optimization problem by the optimization calculation unit 117 increases. Therefore, the value of α is preferably set according to the computer resources available to the optimization calculation unit 117 and the allowable calculation time.

[0031] The constraint condition setting unit 113 sets constraint conditions for a variable x of an optimization problem, which will be described later, for each combination candidate extracted by the steel billet-order combination candidate extraction unit 111. FIG. 11 is a diagram showing an example of constraint conditions set by the constraint condition setting unit 113. In the illustrated example, the decision variable x related to processing conditions in the post-process, specifically, the heating temperature and rolling temperature in the rolling process, roll For each of these decision variables, upper and / or lower limits are set as constraints. roll The constraints are set based on, for example, equipment specifications, equipment status, production schedules, and the like.

[0032] The evaluation function setting unit 115 sets an evaluation function for evaluating the cost related to manufacturing (hereinafter, referred to as manufacturing cost). Specifically, the evaluation function setting unit 115 sets a decision variable x roll (Specifically, this includes formulating the equation of the evaluation function J, setting constants, etc.). FIG. 12 is a diagram showing an example of constants related to manufacturing costs (hereinafter, also referred to as cost constants) included in the evaluation function J set by the evaluation function setting unit 115. In the illustrated example, roll For each of the heating temperature, rolling temperature, and cooling temperature included in the table, the manufacturing cost per unit (1°C) (i.e., the value of the cost constant) is set. In this case, the cost constant is a coefficient multiplied by the value of each temperature. For the heating temperature, a positive value of 1 is set because the higher the heating temperature, the more time and fuel are required for heating, and for the rolling temperature, a negative value of -2 is set because the lower the temperature, the more waiting time occurs. For the cooling temperature, 0 is set because the process has a small effect on the manufacturing cost.

[0033] The optimization calculation unit 117 uses the quality prediction model 106 constructed and stored by the processes of the manufacturing performance data acquisition unit 101, the quality prediction model construction unit 103, and the quality prediction model storage unit 105 to calculate the condition x ∈ [x slab ,x order ,x roll ] is given, the quality characteristic value (here TS, YP and EL) y = [y TS ,y YP ,y EL Specifically, the optimization calculation unit 117 solves the optimization problem formulated as follows. Here, among the above conditions x, x slab is a condition determined based on the steel piece data 108 such as the component content, and x order is a condition determined based on the non-production order data 110 such as plate thickness, and x roll are the processing conditions in the post-process, specifically, the manufacturing conditions such as the heating temperature and rolling temperature that can be adjusted in the rolling process. Condition x slab and condition x order is a constant that has already been determined, while the processing conditions in the post-process xroll is an undetermined decision variable. Therefore, by the optimization calculation unit 117 solving the optimization problem as described above, the processing conditions x roll is optimized under the constraint conditions set by the constraint condition setting unit 113.

[0034]

number

[0035] In the above formula, L Y ,U Y are the lower and upper limits of the specification Y ∈ [TS, YP, EL] of the quality characteristic value of the order, and are determined based on the unproduction order data 110. roll,i lower ,x roll,i upper are variables x roll,i The constant c is determined by the constraint condition setting unit 113. roll,i is the variable x roll,i is a cost constant, which is determined by the evaluation function setting unit 115 as described above.

[0036] On the other hand, in the above formula, F Y (x,q) is the quality characteristic value y=[y TS ,y YP ,y EL ] predictive distribution p(y|x)=p(y|x slab ,x order ,x roll ) and outputs the predicted value of the quality characteristic value y when the condition x and the cumulative probability density q are given. Y We will explain (x,q) later. Y ,U Y ,α1 Y ,α2 Y The range using is given as a constraint. Here, the condition x is x slab ,x order are constants determined based on the billet data 108 and the unproduction order data 110, respectively. Y(x,q) is the decision variable x roll It changes depending on.

[0037] Furthermore, in the above formula, α1 Y ,α2 Y are the lower limit L of the specification Y of the quality characteristic value of the order, respectively. Y and the upper limit U Y The function w(α) is a weighting function for imposing a large penalty on the evaluation function J when the specification Y of the quality characteristic value of the order is not satisfied. Specifically, for example, the function w(α) is set to return 1 when the value of α is non-negative, and 10 when the value of α is negative. The optimization calculation unit 117 solves the optimization problem according to the above formula for each of the combination candidates of surplus materials and unmanufactured orders extracted by the billet-order combination candidate extraction unit 111. α1 Y ,α2 Y When both are non-negative, the processing conditions x that satisfy the quality characteristic value are roll has been solved, and the surplus material can be used to fulfill the orders of the combination candidates.

[0038] The matching result display unit 119 displays the matching result between the surplus material and the unmanufactured order on the display of the manufacturing support device 100 or an externally connected device based on the calculation result by the optimization calculation unit 117. Specifically, the matching result display unit 119 displays the processing condition x that satisfies the quality characteristic value in the optimization calculation unit 117. roll The system outputs information showing the combination of billets and orders for which it was possible to calculate the quality characteristic value. FIG. 13 is a diagram showing an example of the displayed matching results. In the example shown, the matching result display unit 119 displays the number of days of stay, alloy cost, and steel type of the billet data 108 for each billet ID of the surplus material, and displays a matrix of "application possibility" for each order ID of the unmanufactured order. roll For the combination of billets and orders for which the quality characteristic values ​​are obtained, an icon (filled in square) indicating that the combination can be used is displayed, and the processing conditions x that satisfy the quality characteristic values ​​through optimization calculations are displayed. rollFor combinations for which no allocation is obtained and combinations not extracted as combination candidates by the order combination candidate extraction unit 111, an icon (□) indicating that allocation is not possible is displayed. In addition, the value of α described for the billet-order combination candidate extraction unit 111 may be selectable as the "reliability".

[0039] When a combination of surplus material and an order that is deemed to be applicable is selected in the above display, the predicted range of the quality characteristic values ​​of the billet calculated by the billet-order combination candidate extraction unit 111 for that combination and the specification of the quality characteristic values ​​of the order are displayed as "predicted strength" and "order specification". In addition, the processing condition x obtained by the optimization calculation is displayed as "rolling condition guidance". roll The values ​​of the manufacturing conditions of the rolling process that correspond to the above are displayed. By referring to this information, the user can understand the orders to which surplus material can be allocated, and can appropriately determine the orders to which the surplus material will actually be allocated, taking into account the manufacturing conditions of the rolling process.

[0040] FIG. 14 is a flowchart showing a schematic processing procedure of a manufacturing support method according to an embodiment of the present invention. In the following, a case where the manufacturing support method is executed in the manufacturing support device 100 described with reference to FIG. 1 will be described. First, the manufacturing performance data acquisition unit 101 acquires the manufacturing performance data 102 (step S101). Next, the quality prediction model construction unit 103 constructs a quality prediction model (step S103), and the quality prediction model storage unit 105 stores the quality prediction model 106 (step S105). As already described, the processing of steps S101 to S105 for constructing and storing the quality prediction model 106 may be executed in advance, for example, when the manufacturing performance data 102 related to a predetermined number of products is accumulated, and then, at any timing, such as when a new order is generated, the processing of steps S107 to S121 for outputting information for allocating surplus materials to unmanufactured orders using the stored quality prediction model 106 may be executed.

[0041] In the process of allocating surplus materials to unproduction orders, the billet data acquisition unit 107 acquires billet data 108 (step S107), and the order data acquisition unit 109 acquires unproduction order data 110 (step S109). The order of steps S107 and S109 is not limited to the example shown in the figure, and the order may be reversed or they may be executed in parallel. Next, the billet-order combination candidate extraction unit 111 extracts combination candidates between surplus materials included in the billet data 108 and orders included in the unproduction order data 110 (step S111).

[0042] Furthermore, the constraint condition setting unit 113 sets constraint conditions for the post-process (step S113), and the evaluation function setting unit 115 sets an evaluation function (step S115). Note that step S115 for setting the evaluation function can be executed even without the billet data 108 or the unproduction order data 110, and therefore may be executed at any time before step S117. The optimization calculation unit 117 executes optimization calculation based on the data acquired and calculated above (step S117). The above steps S113 to S117 are executed for all the extracted combination candidates (step S119). Note that it is not necessary to repeat all of steps S113 to S117 as many times as the number of combination candidates, and for example, steps S113 and S115 may be executed first for all combination candidates.

[0043] When the optimization calculation is performed, the matching result display unit 119 displays the matching results between surplus materials and unmanufactured orders based on the calculation results on the display of the manufacturing support device 100 or an externally connected device (step S121).

[0044] According to one embodiment of the present invention as described above, by solving an optimization problem using the quality prediction model 106 constructed based on the manufacturing history data 102, it is determined whether or not surplus material recorded as billet data 108 can be allocated to an order in the unmanufactured order data 110. Since the quality prediction model 106 is capable of predicting the probability distribution of the quality characteristic values ​​of the final product from the billet attribute values ​​and processing conditions, it is possible to predict the possibility that surplus material can be allocated to an order over a wide range of combinations of component contents of billets and specifications of quality characteristic values ​​of orders for which, for example, manufacturing history data 102 does not exist. Therefore, in this embodiment, more surplus material can be allocated to appropriate orders.

[0045] Fig. 15 is a diagram showing a schematic configuration of a manufacturing support device according to a modified embodiment of the present invention. The manufacturing support device 200 shown in Fig. 15 is configured similarly to the manufacturing support device described above with reference to Fig. 1, but includes a billet data acquisition unit 107, an order data acquisition unit 109, a billet-order combination candidate extraction unit 111, a constraint condition setting unit 113, an evaluation function setting unit 115, an optimization calculation unit 117, and a matching result display unit 119, and does not include a manufacturing performance data acquisition unit, a quality prediction model construction unit, and a quality prediction model storage unit. In the illustrated example, an optimization calculation is performed using a quality prediction model 106 that has already been constructed and stored using another device, and a process is executed to output information for allocating surplus materials to unmanufactured orders.

[0046] FIG. 16 is a graph showing the result of applying an actually existing surplus slab to an unmanufactured order as an embodiment of the present invention. When an optimization calculation was performed using the quality prediction model 106 for surplus slabs of eight types of steel A to H that were not applied to orders by the conventional method and had been in stock for 100 days or more, it was found that for all seven orders order1 to order7, there were surplus slabs that were determined to be applicable because they satisfied the specifications of the quality characteristic values ​​of the orders (TS and YP) with a probability of 90% or more. This included an example in which the specifications of the quality characteristic values ​​of an order of a non-heat-treatment steel type were satisfied by manufacturing a surplus slab of a heat-treatment steel type under the manufacturing conditions of a specific rolling process, and it was found that the possibility of applying surplus to an order can be determined widely, not only for those with a manufacturing record.

[0047] Although the preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, the present invention is not limited to these examples. It is clear that a person skilled in the art in the technical field to which the present invention pertains can come up with various modified or altered examples within the scope of the technical ideas described in the claims, and it is understood that these also naturally fall within the technical scope of the present invention. [Explanation of symbols]

[0048] 100, 200... manufacturing support device, 101... manufacturing performance data acquisition unit, 102... manufacturing performance data, 103... quality prediction model construction unit, 104... quality prediction model storage unit, 105... quality prediction model storage unit, 106... quality prediction model, 107... billet data acquisition unit, 108... billet data, 109... order data acquisition unit, 110... unmanufactured order data, 111... billet-order combination candidate extraction unit, 113... constraint condition setting unit, 115... evaluation function setting unit, 117... optimization calculation unit, 119... matching result display unit, 200... manufacturing support device.

Claims

1. A manufacturing support device that supports the application of steel billets to orders, comprising: a quality prediction unit that predicts a quality characteristic value of the steel billet when the steel billet is made into a product based on an attribute value of the steel billet and processing conditions; an optimization calculation unit that optimizes the processing conditions so that the quality characteristic value predicted by the quality prediction unit satisfies a quality characteristic value specification required in the order; an output unit that outputs information indicating a combination of the steel billet and the order for which processing conditions can be calculated in the optimization calculation unit; A manufacturing support device comprising:

2. The manufacturing support device according to claim 1 , wherein the quality prediction unit predicts the quality characteristic value using a quality prediction model constructed by a Bayesian estimation method.

3. a combination candidate extraction unit that extracts combination candidates that are relatively likely to satisfy specifications required in the orders from all combinations of the steel billets and the orders, The manufacturing support device according to claim 1 or 2, wherein the optimization calculation unit optimizes the processing conditions for the extracted combination candidates.

4. 4. The manufacturing support device according to claim 3, wherein the combination candidate extraction unit extracts combination candidates between the billet and the order in which the weight of the billet is equal to or greater than a remaining weight requested in the order.

5. 5. The manufacturing support device according to claim 3, wherein the candidate combination extraction unit extracts a candidate combination between the billet and the order, the component of which satisfies a component specification required in the order.

6. 6. The manufacturing support device according to claim 3, wherein the combination candidate extraction unit calculates a possible range of quality characteristic values ​​for each steel type of the steel billet based on past manufacturing performance data, and extracts combination candidates between the steel billet and the order, the range of which overlaps with specifications of quality characteristic values ​​required in the order.

7. 6. The manufacturing support device according to claim 3, wherein the combination candidate extraction unit uses a Bayesian estimation method to construct a prediction model that predicts a probability distribution of quality characteristic values ​​for components of the steel billet based on past manufacturing performance data, and extracts combination candidates between the steel billet and the order, in which a possible range of the quality characteristic values ​​predicted by the prediction model overlaps with a specification of the quality characteristic value required in the order.

8. A constraint condition setting unit that sets constraint conditions for the processing conditions, The manufacturing support device according to claim 1 , wherein the optimization calculation unit optimizes the processing conditions under the constraint conditions.

9. An evaluation function setting unit that sets an evaluation function for the processing conditions, The manufacturing support device according to claim 1 , wherein the optimization calculation unit optimizes the processing conditions based on the evaluation function.

10. The manufacturing support device according to claim 1 , wherein the output unit displays, in a matrix, combinations of the processing conditions that allowed optimization for each steel piece or for each order.

11. The manufacturing support device according to claim 1 , wherein the output unit displays the optimized machining conditions for a combination in which the machining conditions were optimized.

12. A quality prediction model construction unit that constructs a quality prediction model based on past manufacturing performance data, The manufacturing support device according to claim 1 , wherein the quality prediction unit predicts the quality characteristic value by using the quality prediction model.

13. 1. A manufacturing support method for supporting the fulfillment of steel billets to orders, comprising the steps of: a quality prediction step of predicting a quality characteristic value of the steel billet when the steel billet is made into a product based on the attribute values ​​of the steel billet and processing conditions; an optimization calculation step of optimizing the processing conditions so that the predicted quality characteristic value satisfies the quality characteristic value specifications required in the order; an output step of outputting information indicating a combination of the steel billet and the order for which processing conditions can be calculated in the optimization calculation step; A manufacturing support method comprising:

14. A manufacturing support device that supports the application of steel billets to orders, comprising: a quality prediction unit that predicts a quality characteristic value of the steel billet when the steel billet is made into a product based on an attribute value of the steel billet and processing conditions; an optimization calculation unit that optimizes the processing conditions so that the quality characteristic value predicted by the quality prediction unit satisfies a quality characteristic value specification required in the order; an output unit that outputs information indicating a combination of the steel billet and the order for which processing conditions can be calculated in the optimization calculation unit; Manufacturing support device equipped with A program that makes a computer function as a

Images