JP6923420B2 - Production process analysis method - Google Patents

Description

The present invention relates to a method for analyzing a production process for producing a product or service.

The production process of manufacturing a product includes a process by an automated manufacturing facility and a manual process by an operator. In addition, the production process that provides services such as parts cleaning and analysis of clinical trial results in drug development includes a process using a cleaning device and an analysis device and a manual process by a worker.

In the production process of manufacturing products and providing services, the quality of products and services varies from lot to lot depending on the operating conditions of equipment, the working conditions of workers, the handling methods of raw materials and products, and the like.

Principal component analysis and cluster analysis are applied to product data and process data related to the product manufacturing process to divide the lot of the manufacturing process into multiple groups and contribute to the superiority or inferiority between the groups in order to suppress the variation in product quality. There are known data that identify the inhibitory factor (see, for example, Patent Document 1).

In such a manufacturing process analysis method, the manufacturing process can be efficiently improved by identifying the impeding factors that contribute to the superiority or inferiority between the groups, so that the quality of the product between lots can be improved.

Japanese Patent No. 5956094

By the way, products and services are produced through various processes. In the analysis method as described above, the hindrance factor is identified from the process data of the entire production process, and there is a problem that it is not enough to improve the entire process in consideration of the state of each process, for example.

Therefore, a technical problem to be solved in order to stabilize the quality of the product manufactured through a plurality of processes and the service provided arises, and an object of the present invention is to solve this problem. do.

In order to achieve the above object, the production process according to the present invention is a method for analyzing a production process in which a product or service is produced, which is composed of a plurality of processes, and is a state quantity indicating a state for each process and the product. Alternatively, a step of collecting data including at least a quality item indicating the quality of service for each lot of the production process, a step of dividing each process into a plurality of groups according to the state quantity of the process, and each combination of the groups. A step of dividing the lot into a plurality of routes, a step of determining the superiority or inferiority of the route according to the quality item of the route, and a step of specifying a suitable combination of the groups according to the superiority or inferiority of the route. ,including.

According to this configuration, the process is divided into a plurality of groups according to the state quantity of the process, the lot is divided into a plurality of routes for each combination of the process groups, and the superiority or inferiority of the route is judged according to the quality item of the route. By doing so, it is possible to select a combination of process states that contributes to improving the quality of products and services.

The present invention divides a process into a plurality of groups according to the state quantity of the process, divides a lot into a plurality of routes for each combination of process groups, and determines the superiority or inferiority of the route according to the quality item of the route. Therefore, it is possible to select a combination of process states that contributes to improving the quality of products and services.

The flowchart which shows the analysis method of the production process which concerns on one Embodiment of this invention. The graph which shows the principal component score and the group for each lot in the 1st step. The graph which shows the principal component score and the group for every lot in the 2nd step. The graph which shows the principal component score and the group for every lot in the 3rd step. The graph which shows the principal component score and the group for every lot in the 4th step. The graph which shows the principal component score and the group for every lot in the 5th step. The graph which shows the principal component score and the group for every lot in the 6th step. A table showing the combination of groups and product data for each process. The figure which shows the combination of the group of each process corresponding to Case 4-7 of FIG.

An embodiment of the present invention will be described with reference to the drawings. In the following, when referring to the number, numerical value, quantity, range, etc. of components, it is limited to the specific number unless it is clearly stated or in principle it is clearly limited to the specific number. It is not a thing, and it may be more than or less than a specific number.

The analysis method according to the present embodiment is applied to a process of manufacturing a product (product) or a process of providing a service (hereinafter, collectively referred to as a "production process"). Production processes include processes that consist only of mechanical equipment and all processes are automated, processes that include manual work processes by workers, and manufacturing processes that are automated by mechanical equipment and manual work by workers. Includes processes that include steps.

In the following, a case where this analysis method is applied to a fiber production line, which is an example of a production process, will be described as an example. It goes without saying that the production process to which the present invention is applied is not construed as being limited to the fiber production line, and includes other production lines and processes for providing services.

Although the fiber production line is composed of a plurality of steps, the following description will be made on the assumption that the production line has undergone six steps (first to sixth steps) in a simplified manner. The "process" may be set for each individual process or may be set collectively for a plurality of processes.

Each device constituting the production line is provided with a sensor (not shown) for measuring various values. The measurement target of the sensor is the input amount of the raw material, the temperature in the processing device that performs various processing such as heating, the processing amount per unit time, and the like. The sensor sends the measured value to the control device that regulates the manufacturing equipment that composes the production line.

The control device performs the processing described later based on the process data indicating the production conditions of the product measured by the sensor and the product data (quality data) indicating the quality of the product. Process data is a factor that can affect the quality of products and is a state quantity that indicates the state of the process. Composition, etc.) etc. are included.

FIG. 2 is a flowchart showing the procedure of the analysis method of the production process according to the present embodiment.

First, with respect to the operational production process, the control device collects the process data and the product data measured by the sensor (step S1). In this step S1, process data and product data for each lot are stored in the control device.

Next, the process data and product data collected in step S1 are standardized and converted into intermediate functions (step S2).

The process data standardization process performed in step S2 is known, and specifically, the control device calculates based on the mathematical formula 1.

Next, the principal component load and the principal component score are obtained based on the intermediate variables obtained in step S2 (step S3). In step 3, first, the correlation coefficient matrix in the intermediate variable is created, and the eigenvalues and eigenvectors of the correlation coefficient matrix are derived. In the correlation coefficient matrix, when the intermediate variables are x1, x2, x3 ..., The first principal component PC1 is expressed as shown by Equation 2. Further, the Nth principal component PCn is expressed as shown by Equation 3. Then, the correlation coefficient matrix is formed by using the coefficients a11, a12, a13 ... As the elements in the first row and the coefficients an1, an2, an3 ... As the elements in the nth row.

Next, the principal component score is obtained from the eigenvector of the correlation coefficient matrix. In addition, the contribution rate of each principal component is obtained from the eigenvalues of the correlation coefficient matrix. The contribution rate of the principal component is obtained by dividing the eigenvalues by the sum of the eigenvalues. Here, the first principal component, the second principal component, and the Nth principal component are determined from the one having the larger eigenvalue.

Specifically, the control device determines the values of the first principal component PC1, the second principal component PC2, and the like based on the intermediate variables x1, x2, x3 of each lot and each coefficient of the correlation coefficient matrix, that is, The principal component score is calculated. FIG. 2 is a graph in which the principal component scores are plotted in a coordinate system in which the first principal component is on the horizontal axis and the second principal component is on the vertical axis for the state quantity in the first step.

Next, the control device applies cluster analysis to the principal component scores shown in FIG. 2 to divide each lot into a plurality of groups (step S4). "Cluster analysis" is a method of classifying analysis target data (clusters) into a plurality of groups focusing on similarity, and hierarchical clustering, classification optimization clustering, and the like are known. The "similarity" that the cluster analysis pays attention to in this example refers to the distance between the principal component scores of each lot. In this example, aggregated hierarchical clustering, which is one of hierarchical clustering, was used. In addition, Ward's method, which provides a stable solution, was used as the method for calculating the distance between clusters. The "Ward's method" is to select the cluster that minimizes the increase in the sum of squared deviations when the two clusters are merged. For example, when clusters A and B are merged to generate cluster C, the sums of deviation squares Sa, Sb, and Sc in clusters A, B, and C are expressed as formulas 4 to 6, respectively.

According to Equations 4 to 6, the sum of squared deviations Sc in the cluster C is as follows.

ΔSab in Equation 7 means an increment of the sum of squared deviations when clusters A and B are merged to form cluster C. Therefore, clustering is advanced by selecting and merging clusters so that ΔSab is minimized at each merging stage.

In this embodiment, as shown in FIG. 2, the groups 1 to 6 are divided into 6 groups 1 to 6 in the five-dimensional space from the first to the fifth eigenvectors. Groups 1 to 6 correspond to clusters 1 to 6 in FIG. 2, respectively. Further, the number of groups is not limited to 6, and may be 5 or less or 7 or more as long as it is easy to handle.

Similarly, in the second to sixth steps, the principal component scores are plotted on the graph and divided into a plurality of groups. 3 to 7 are graphs showing principal component scores and groups (clusters) for the state quantities of the second to sixth steps.

Next, the lot generation process is divided into a plurality of routes for each combination of each group in the first to sixth steps (step S5). In this embodiment, since it was found that there are 16 types of routes based on the past production results, the routes are divided into 16 routes. FIG. 8 shows the breakdown of all routes. In addition, "Case" in FIG. 8 corresponds to the above-mentioned route, and the numbers in FIG. 8 correspond to the numbers of the groups in the first to sixth steps. Further, FIG. 9 shows routes corresponding to Cases 4 to 7 as an example of a combination of groups constituting the route.

Next, the superiority or inferiority of the product data is determined for each route (step S6). In this step S6, the control device calls the intermediate variables obtained from the product data (visual inspection defect rate, etc.) for each lot belonging to Cases 1 to 16 of FIG. 8, and determines the superiority or inferiority of these product data.

The superiority or inferiority of the product data is preferably based on the average value in a plurality of lots (lot groups) divided for each route. As a result, the variation in the product data of the lot group within the route is leveled, and the tendency of the quality of the product data between the routes can be grasped from a broad perspective.

Further, the superiority or inferiority of the product data may be determined based on the magnitude of the deviation of the product data in the route and the magnitude of the difference (range) between the maximum value and the minimum value, and the average value, deviation, R value, etc. may be determined by 2. You may judge by combining two or more. As a method of determining the superiority or inferiority of product data by combining the average value and the deviation, for example, when the average value in the route is the same, it is conceivable to determine that the deviation in the route is small as superior. As a result, it is possible to grasp the tendency of superiority or inferiority of product data between routes in consideration of the variation of product data within the route.

Then, the control device compares the product data for each route and determines the superiority or inferiority. In FIG. 8, the visual inspection defect rate as product data in each route was evaluated on a three-point scale of excellent, good, and acceptable according to their superiority and inferiority.

Next, a suitable combination of the groups of the first to sixth steps is specified according to the superiority or inferiority of the product data for each route (step S7). For example, comparing Cases 4 and 5 determined to be excellent with Cases 6 and 7 determined to be good, these are the first, second, fourth, and sixth steps, although they are in the same group. Step 3 is a different group. Further, it is presumed that the fifth step does not significantly affect the product data regardless of whether it is in groups 1 and 2. That is, it can be seen that it is preferable to combine the group 1 of the third step in the route that has passed through the group 2 of the first step and the group 2 of the second step.

Further, when the Case 14 determined to be excellent and the Case 15 determined to be good are compared, they are in the same group in the first to fourth and sixth steps, but in different groups in the fifth step. be. That is, the group 5 of the first step, the group 3 of the second step, the group 4 of the third step, the group 2 of the fourth step, the group 1 of the fifth step, and the group 1 of the sixth step. It can be seen that the combination is preferable. The "suitable combination" in the present embodiment means a combination that contributes to the improvement of the product data, and does not mean only the combination in which the product data is the best.

When setting the manufacturing conditions of each process based on the combination selected in step S7, the average value of the process data of each group is set as the initial value of the manufacturing conditions, and the quality of the product data at this time is confirmed. However, it is preferable to fine-tune the manufacturing conditions. By setting the average value of the process data to the initial value, a preferable combination of the groups of the first to sixth steps can be reproduced without deviating from the past state, that is, without imposing an excessive load on the manufacturing equipment and the like. can do.

It should be noted that the present invention can be modified in various ways as long as it does not deviate from the spirit of the present invention, and it is natural that the present invention extends to the modified ones.

The "production process" in the present invention includes a process of manufacturing a product (goods) or a process of providing a service. That is, the production process is not limited to the process of manufacturing a product. Services provided in the production process include, for example, services such as parts cleaning and analysis of clinical trial results in drug development.

Claims (1)

A production process analysis method that uses a control device to analyze a production process that is composed of multiple processes and produces a product or service.
A step of collecting data including at least a state quantity indicating the state of each process and a quality item indicating the quality of the product or service for each lot of the production process.
A step of dividing each process into a plurality of groups according to the state quantity of the process, and
A step of dividing the production process of the lot into a plurality of routes for each combination of the process order of the group to which each process of the lot belongs, and a step of dividing the production process of the lot into a plurality of routes.
A step of determining the superiority or inferiority of the route according to the quality item of the route,
A step of identifying a suitable combination of the groups according to the superiority or inferiority of the route.
A method of analyzing a production process, which comprises.

Images