A transformation tree method for intelligently generating quality improvement solutions

Abstract

This invention discloses a transformation tree method for intelligently generating quality improvement solutions. This method calculates the transformation degree by obtaining parameters, optional parameter values, and their corresponding qualified production results during product production. Then, based on the transformation degree and optional parameter values, a transformation tree is constructed and reversely traversed to obtain transformation solutions. Finally, the transformation solutions are ranked by their merit to obtain the optimal production solution. The production solutions generated using this method can better adapt to the needs of real-world production environments.

Description

Technical Field

[0001] The present invention relates to the field of production quality improvement, and more specifically, to a transformation tree method for intelligently generating quality improvement solutions. Background Art

[0002] In today's increasingly competitive market, production management has become a crucial cornerstone for business survival and growth. By deeply integrating information technology into every aspect of production and using models to optimize these processes, companies can improve the consistency and stability of product quality and save costs.

[0003] While existing technologies have improved product quality and provided some control over costs, they still face numerous shortcomings. Some models optimize only for the final result, leading to overfitting and ignoring potential for improvement in intermediate steps. Others focus solely on idealized scenarios, ignoring environmental variations and resulting in a lack of adaptability. Consequently, existing models struggle to adapt to real-world production scenarios.

[0004] Prior art discloses a method for improving quality consistency. This method first obtains material parameters, then analyzes the factors causing unstable quality, and finally proposes targeted improvement measures based on these factors. This method only adjusts individual factors and does not consider overall improvements. Summary of the Invention

[0005] In view of the defect that the quality and cost adjustment models of production links in the existing technology are difficult to adapt to the scenarios of actual production applications, the present invention proposes a transformation tree method for intelligently generating quality improvement solutions.

[0006] The primary purpose of the present invention is to solve the above technical problems, and the technical solutions of the present invention are as follows:

[0007] A transformation tree method for intelligently generating quality improvement solutions, comprising:

[0008] S1: Obtain several production parameters in the product production process, the optional value of each production parameter and the corresponding number of qualified products and unqualified products, and divide the several production parameters into adjustable parameters and non-adjustable parameters;

[0009] S2: Calculate the conversion degree of each adjustable parameter based on the number of qualified products and unqualified products corresponding to the optional value of each adjustable parameter;

[0010] S3: Set the non-adjustable parameters as the root node, and build a transformation tree based on the transformation degree and optional values ​​of each adjustable parameter. Each branch of the transformation tree corresponds to a production plan.

[0011] S4: Select a specific production plan, traverse the corresponding tree branch in reverse order, determine several transformation plans, and calculate the quality of each transformation plan;

[0012] S5: Sort all transformation schemes according to their excellence to obtain an excellence sequence;

[0013] S6: Test the transformation scheme results that are at the top of the goodness sequence. If they do not meet expectations, new test results are fed back, and samples are replenished and generated again. The scheme that meets expectations is output as the optimal production scheme.

[0014] Furthermore, in step S1, after dividing the plurality of production parameters into adjustable parameters, the step further includes performing a discreteness judgment on the adjustable parameters. If the adjustable parameters are continuous values, a parameter discretization operation is performed.

[0015] Furthermore, in step S3, constructing the transformation tree includes:

[0016] S301: Obtain a root node as a root node group; sort the conversion degrees of various adjustable parameters in ascending order to form an adjustable parameter sequence;

[0017] S302: sequentially selecting an adjustable parameter from the adjustable parameter sequence as a first adjustable parameter;

[0018] S303: deriving corresponding child nodes from the nodes of each root node group according to the optional value of the first adjustable parameter;

[0019] S304: Using the derived child nodes as a new root node group;

[0020] S305: Replace the first adjustable parameter and execute step S303 until each adjustable parameter in the adjustable parameter sequence is traversed.

[0021] Furthermore, in step S4, the reverse order traversal includes:

[0022] S401: Using the leaf node of the tree branch corresponding to the production plan as the second node;

[0023] S402: The parent node corresponding to the second node is used as the third node; and the nodes excluding the second node from the child nodes of the third node are used as a child node group;

[0024] S403: Determine whether there is a node in the child node group; if so, execute step S404; if not, execute step S406;

[0025] S404: Select a node in the child node group as the first node, and replace the parameter optional value represented by the first node with the parameter optional value corresponding to the second node in the production plan to form a new production plan;

[0026] S405: Remove the first node from the child node group and execute step S403;

[0027] S406: Determine whether the third node is a root node; if so, execute step S408; if not, execute step S407;

[0028] S407: Set the third node as the new second node, and the parent node of the third node as the new third node; remove the new second node from the child nodes of the new third node as a new child node group; and execute step S403;

[0029] S408: The obtained new production plan is used as a transformation plan.

[0030] Furthermore, before step S5, the method further includes:

[0031] Determine whether the number of conversion schemes reaches a preset number; if so, directly execute step S5; otherwise, increase the optional values ​​of production parameters and / or adjustable parameters and repeat steps S1-S4.

[0032] Furthermore, the calculation formula of goodness is as follows:

[0033] O=α1k1+α2k2

[0034] Among them, O represents goodness, α1 and α2 represent weights, k1 represents the conversion cost correlation, and k2 represents the conversion degree.

[0035] Furthermore, the calculation formula for conversion cost correlation is as follows:

[0036]

[0037] Among them, k1(x) represents the conversion cost correlation, ρ represents the extension distance calculation operation, D represents the extension bit value calculation operation, x0 represents the minimum conversion cost, x represents the conversion cost, X represents the range of conversion cost that meets the production requirements, and X0 represents several ranges of X with smaller conversion costs. Indicates the range of X and the production requirements that are met in a specific case. Indicates the selectable range for all conversion costs.

[0038] Furthermore, the calculation formula of conversion degree is as follows:

[0039]

[0040] Among them, ai The increment of qualified products after the i-th adjustable parameter is adjusted from one optional value to another optional value is calculated by multiplying the frequency of the optional value by the confidence of the other optional value, b i The total amount of unqualified products before the i-th adjustable parameter is adjusted from one optional value to another.

[0041] Furthermore, in step S1, the parameters include processing method parameters, transportation method parameters, storage method parameters and loss rate.

[0042] A conversion tree device for intelligently generating a quality improvement solution, comprising:

[0043] Production parameter classification module: obtains several production parameters in the product production process, the optional value of each production parameter and the corresponding number of qualified products and unqualified products, and divides several production parameters into adjustable parameters and non-adjustable parameters;

[0044] Conversion degree calculation module: calculates the conversion degree of each adjustable parameter based on the number of qualified products and unqualified products corresponding to the optional values ​​of each adjustable parameter;

[0045] Conversion tree construction module: Set the non-adjustable parameters as the root node, and build a conversion tree based on the conversion degree and optional values ​​of each adjustable parameter. Each branch of the conversion tree corresponds to a production plan;

[0046] Reverse traversal module: selects a specific production plan, traverses the corresponding tree branch in reverse order, determines several transformation plans, and calculates the quality of each transformation plan;

[0047] Optimality ranking module: sorts all transformation schemes according to their optimality to obtain the optimality sequence;

[0048] Optimal selection module: The transformation scheme that is at the top of the optimality sequence is selected as the optimal production scheme.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] This invention constructs a transformation tree to rank the importance of production line parameter improvements. New transformation plans are generated by traversing the transformation tree in reverse order. By introducing a merit ranking algorithm, this invention comprehensively considers the cost and efficiency of each plan to select the optimal production solution. Overall, the solutions generated by this invention are more adaptable to actual production scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A flow chart of a conversion tree method for intelligently generating a quality improvement solution provided in an embodiment.

[0052] Figure 2The flowchart of constructing the transformation tree provided in the embodiment.

[0053] Figure 3 A reverse order traversal flowchart is provided for an embodiment.

[0054] Figure 4 A structural diagram of a conversion tree device for intelligently generating a quality improvement solution provided in an embodiment. DETAILED DESCRIPTION

[0055] The accompanying drawings are for illustrative purposes only and are not to be construed as limiting this patent;

[0056] In order to better illustrate this embodiment, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product size;

[0057] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0058] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0059] Example

[0060] like Figure 1 As shown, a transformation tree method for intelligently generating a quality improvement solution includes:

[0061] S1: Obtain several production parameters in the product production process, the optional value of each production parameter and the corresponding number of qualified products and unqualified products, and divide the several production parameters into adjustable parameters and non-adjustable parameters;

[0062] S2: Calculate the conversion degree of each adjustable parameter based on the number of qualified products and unqualified products corresponding to the optional value of each adjustable parameter;

[0063] S3: Set the non-adjustable parameters as the root node, and build a transformation tree based on the transformation degree and optional values ​​of each adjustable parameter. Each branch of the transformation tree corresponds to a production plan.

[0064] S4: Select a specific production plan, traverse the corresponding tree branch in reverse order, determine several transformation plans, and calculate the quality of each transformation plan;

[0065] S5: Sort all transformation schemes according to their excellence to obtain an excellence sequence;

[0066] S6: Test the transformation scheme results that are at the top of the goodness sequence. If they do not meet expectations, new test results are fed back, and samples are replenished and generated again. The scheme that meets expectations is output as the optimal production scheme.

[0067] It should be noted that the conversion tree method for intelligently generating quality improvement solutions of the present invention mainly involves a production solution generation method based on a conversion tree.

[0068] It should be noted that this method can combine and extract different transformation schemes from mixed parameters, and the transformation schemes are screened to meet the needs of actual production.

[0069] In a specific embodiment, the method is applied to cold shaping processes such as cutting and stamping, and the feed distance, feed speed, cutting speed, and surface pressure parameters in the production process are collected as several production parameters in the product production of step S1.

[0070] In a specific embodiment, the method is applied to thermal processes such as forging and welding, and the heating temperature, heating time, and material quantity parameters in the production process are collected as several production parameters in step S1 product production.

[0071] In a specific embodiment, the method is applied to the mixing process of raw materials in different proportions, and the amounts, mixing sequences, and mixing methods of different components in the mixing process are collected as several production parameters in the product production in step S1.

[0072] In a specific embodiment, if there are multiple non-adjustable parameters, the information gain ratios of the non-adjustable parameters may be calculated and the root node may be constructed by arranging the information gain ratios in descending order.

[0073] The formula for information gain is as follows:

[0074]

[0075] Furthermore, in step S1, after dividing the plurality of production parameters into adjustable parameters, the step further includes performing a discreteness judgment on the adjustable parameters. If the adjustable parameters are continuous values, a parameter discretization operation is performed.

[0076] In a specific embodiment, the method is applied to the mixing process of raw materials with different proportions. The mixing of different components in the mixing process is collected. A discrete value index needs to be established, the mixing proportion is used as a separate variable, and the mixing proportion value is used as a discrete optional value.

[0077] Furthermore, if Figure 2 As shown, in step S3, constructing the transformation tree includes:

[0078] S301: Obtain a root node as a root node group; sort the conversion degrees of various adjustable parameters in ascending order to form an adjustable parameter sequence;

[0079] S302: sequentially selecting an adjustable parameter from the adjustable parameter sequence as a first adjustable parameter;

[0080] S303: deriving corresponding child nodes from the nodes of each root node group according to the optional value of the first adjustable parameter;

[0081] S304: Using the derived child nodes as a new root node group;

[0082] S305: Replace the first adjustable parameter and execute step S303 until each adjustable parameter in the adjustable parameter sequence is traversed.

[0083] Furthermore, if Figure 3 As shown, in step S4, the reverse order traversal includes:

[0084] S401: Using the leaf node of the tree branch corresponding to the production plan as the second node;

[0085] S402: The parent node corresponding to the second node is used as the third node; and the nodes excluding the second node from the child nodes of the third node are used as a child node group;

[0086] S403: Determine whether there is a node in the child node group; if so, execute step S404; if not, execute step S406;

[0087] S404: Select a node in the child node group as the first node, and replace the parameter optional value represented by the first node with the parameter optional value corresponding to the second node in the production plan to form a new production plan;

[0088] S405: Remove the first node from the child node group and execute step S403;

[0089] S406: Determine whether the third node is a root node; if so, execute step S408; if not, execute step S407;

[0090] S407: Set the third node as the new second node, and the parent node of the third node as the new third node; remove the new second node from the child nodes of the new third node as a new child node group; and execute step S403;

[0091] S408: The obtained new production plan is used as a transformation plan.

[0092] In a specific embodiment, the method is applied to shaping cold processing such as cutting and stamping. It is assumed that the conversion degree of cutting speed, feed distance and surface pressure increases in sequence, and a conversion tree is constructed thereby. A parameter combination of feed speed category 0, cutting speed category 2, feed distance category 3 and surface pressure category 1 is selected as a production plan and recorded as (0, 2, 3, 1). When traversing in reverse, first adjust the leaf node parameter, that is, the surface pressure, and after modifying it to category 2, a new production plan (0, 2, 3, 2) is obtained. Similarly, after completing the modification of the surface pressure, reverse traverse to adjust the feed distance to category 2, forming a new production plan (0, 2, 2, 1). And so on, until all adjustable parameters are modified.

[0093] In a specific embodiment, this method is used in thermal processes such as forging and welding. It is assumed that the conversion degree of the heating temperature is less than the heating time, and a conversion tree is constructed accordingly. A parameter combination of 2 types of material quantity, 4 types of heating temperature, and 3 types of heating time is selected as the production plan and recorded as (2,4,3). When traversing in reverse, first adjust the leaf node parameter, that is, the heating time. After changing it to type 1, a new production plan (2,4,1) is obtained. Similarly, after completing the modification of the heating time, reverse traverse to adjust the heating temperature to type 3, forming a new production plan (2,3,3). And so on, until all adjustable parameters are modified.

[0094] Furthermore, before step S5, the method further includes:

[0095] Determine whether the number of conversion schemes reaches a preset number; if so, directly execute step S5; otherwise, increase the optional values ​​of production parameters and / or adjustable parameters and repeat steps S1-S4.

[0096] Furthermore, the calculation formula of goodness is as follows:

[0097] O=α1k1+α2k2

[0098] Among them, O represents goodness, α1 and α2 represent weights, k1 represents the conversion cost correlation, and k2 represents the conversion degree.

[0099] In a specific embodiment, α1=0.5, α2=0.5.

[0100] In a specific embodiment, it is necessary to ensure that k1∈[-1,1] and k2∈[-1,1]. Therefore, k1 and k2 need to be normalized. The specific formula is as follows:

[0101]

[0102] Furthermore, the calculation formula for conversion cost correlation is as follows:

[0103]

[0104] Among them, k1(x) represents the conversion cost correlation, ρ represents the extension distance calculation operation, D represents the extension bit value calculation operation, x0 represents the minimum conversion cost, x represents the conversion cost, X represents the range of conversion cost that meets the production requirements, and X0 represents several ranges of X with smaller conversion costs. Indicates the range of X and the production requirements that are met in a specific case. Indicates the selectable range for all conversion costs.

[0105] In a specific embodiment, the operation of calculating the extension distance is as follows:

[0106] Assume that the extension distance is ρ(x,x0,N), N is the interval [a,b] of axis x, then the minimum conversion cost x0 can be regarded as a point in axis x.

[0107] When x0 is to the left of the midpoint of the interval:

[0108]

[0109] When x0=a:

[0110]

[0111] When x0 is to the right of the midpoint of the interval:

[0112]

[0113] When x0=b:

[0114]

[0115] When x0 is at the midpoint of the interval:

[0116]

[0117] The calculation operation of the extension bit value is as follows:

[0118] D(x,x0,N,M)=ρ(x,x0,M)-ρ(x,x0,N)

[0119] In a specific embodiment, if the minimum conversion cost x0 = 2500, the conversion cost range X0 is [2300, 2700], the conversion cost meets the production requirement range X is set to [1500, 3000], the conversion cost meets the production requirement range and the production requirement range under specific circumstances is [1000,3500], then:

[0120]

[0121] k1(2900)=0.5

[0122] Furthermore, the calculation formula of conversion degree is as follows:

[0123]

[0124] Among them, a i The increment of qualified products after the i-th adjustable parameter is adjusted from one optional value to another optional value is calculated by multiplying the frequency of the optional value by the confidence of the other optional value, b i The total amount of unqualified products before the i-th adjustable parameter is adjusted from one optional value to another.

[0125] Furthermore, in step S1, the parameters include processing method parameters, transportation method parameters, storage method parameters and loss rate.

[0126] like Figure 4 As shown, a conversion tree device for intelligently generating a quality improvement solution includes:

[0127] Production parameter classification module: obtains several production parameters in the product production process, the optional value of each production parameter and the corresponding number of qualified products and unqualified products, and divides several production parameters into adjustable parameters and non-adjustable parameters;

[0128] Conversion degree calculation module: calculates the conversion degree of each adjustable parameter based on the number of qualified products and unqualified products corresponding to the optional values ​​of each adjustable parameter;

[0129] Conversion tree construction module: Set the non-adjustable parameters as the root node, and build a conversion tree based on the conversion degree and optional values ​​of each adjustable parameter. Each branch of the conversion tree corresponds to a production plan;

[0130] Reverse traversal module: selects a specific production plan, traverses the corresponding tree branch in reverse order, determines several transformation plans, and calculates the quality of each transformation plan;

[0131] Optimality ranking module: sorts all transformation schemes according to their optimality to obtain the optimality sequence;

[0132] Optimal selection module: The transformation scheme that is at the top of the optimality sequence is selected as the optimal production scheme.

[0133] The same or similar reference numerals correspond to the same or similar components;

[0134] The terms used in the drawings to describe positional relationships are for illustrative purposes only and should not be construed as limiting this patent;

[0135] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the claims of the present invention.

Claims

1. A transformation tree method for intelligently generating quality improvement solutions, characterized in that: include: S1: Obtain several production parameters in the product production process, the optional value of each production parameter and the corresponding number of qualified products and unqualified products, and divide the several production parameters into adjustable parameters and non-adjustable parameters; S2: Calculate the conversion degree of each adjustable parameter based on the number of qualified products and unqualified products corresponding to the optional value of each adjustable parameter; S3: Set the non-adjustable parameters as the root node, and build a transformation tree based on the transformation degree and optional values ​​of each adjustable parameter. Each branch of the transformation tree corresponds to a production plan. S4: Select a specific production plan, traverse the corresponding tree branch in reverse order, determine several transformation plans, and calculate the quality of each transformation plan; S5: Sort all transformation schemes according to their excellence to obtain an excellence sequence; S6: Test the transformation solutions that are at the top of the optimality sequence. If they do not meet expectations, new test results are fed back, and samples are replenished and generated again. The solution that meets expectations is output as the optimal production solution. The calculation formula of goodness is as follows: Among them, O represents goodness, and represents the weight, represents the conversion cost correlation, Indicates the degree of conversion; The calculation formula for conversion cost correlation is as follows: in, represents the conversion cost correlation, ρ Indicates the extension distance calculation operation, D Indicates the extension bit value calculation operation, x 0 represents the minimum conversion cost. x represents the conversion cost, X Indicates that the conversion cost meets the production requirements. X 0 represents several ranges of X with relatively low conversion costs. express X and meet the production requirements in specific cases, Indicates the selectable range of all conversion costs; The conversion rate is calculated as follows: in, It represents the increment of qualified products after the i-th adjustable parameter is adjusted from the first optional value to the second optional value. The increment is calculated by multiplying the frequency of the first optional value by the confidence level of the second optional value. It represents the total amount of unqualified products before the i-th adjustable parameter is adjusted from one optional value to another.

2. The method for intelligently generating a transformation tree for a quality improvement solution according to claim 1, characterized in that: In step S1, after dividing a number of production parameters into adjustable parameters, the step also includes performing a discreteness judgment on the adjustable parameters. If the adjustable parameters are continuous values, a parameter discretization operation is performed.

3. The method for intelligently generating a transformation tree for a quality improvement solution according to claim 1, characterized in that: In step S3, constructing a transformation tree includes: S301: Obtain a root node as a root node group; sort the conversion degrees of various adjustable parameters in ascending order to form an adjustable parameter sequence; S302: sequentially selecting an adjustable parameter from the adjustable parameter sequence as a first adjustable parameter; S303: deriving corresponding child nodes from the nodes of each root node group according to the optional value of the first adjustable parameter; S304: Using the derived child nodes as a new root node group; S305: Replace the first adjustable parameter and execute step S303 until each adjustable parameter in the adjustable parameter sequence is traversed.

4. The method for intelligently generating a transformation tree for a quality improvement solution according to claim 1, wherein: In step S4, the reverse order traversal includes: S401: Using the leaf node of the tree branch corresponding to the production plan as the second node; S402: The parent node corresponding to the second node is used as the third node; and the nodes excluding the second node from the child nodes of the third node are used as a child node group; S403: Determine whether there is a node in the child node group; if so, execute step S404; if not, execute step S406; S404: Select a node in the child node group as the first node, and replace the parameter optional value represented by the first node with the parameter optional value corresponding to the second node in the production plan to form a new production plan; S405: Remove the first node from the child node group and execute step S403; S406: Determine whether the third node is a root node; if so, execute step S408; if not, execute step S407; S407: Set the third node as the new second node, and the parent node of the third node as the new third node; remove the new second node from the child nodes of the new third node as a new child node group; and execute step S403; S408: The obtained new production plan is used as a transformation plan.

5. The method for intelligently generating a transformation tree for a quality improvement solution according to claim 1, characterized in that: Before step S5, the method further includes: Determine whether the number of conversion schemes reaches a preset number; if so, directly execute step S5; otherwise, increase the optional values ​​of production parameters and / or adjustable parameters and repeat steps S1-S4.

6. The method for intelligently generating a transformation tree for a quality improvement solution according to claim 1, characterized in that: In step S1, the parameters include processing method parameters, transportation method parameters, storage method parameters and loss rate.

7. A conversion tree device for intelligently generating a quality improvement solution, used to implement a conversion tree method for intelligently generating a quality improvement solution according to any one of claims 1 to 6, characterized in that: include: Production parameter classification module: obtains several production parameters in the product production process, the optional value of each production parameter and the corresponding number of qualified products and unqualified products, and divides several production parameters into adjustable parameters and non-adjustable parameters; Conversion degree calculation module: calculates the conversion degree of each adjustable parameter based on the number of qualified products and unqualified products corresponding to the optional values ​​of each adjustable parameter; Conversion tree construction module: Set the non-adjustable parameters as the root node, and build a conversion tree based on the conversion degree and optional values ​​of each adjustable parameter. Each branch of the conversion tree corresponds to a production plan; Reverse traversal module: selects a specific production plan, traverses the corresponding tree branch in reverse order, determines several transformation plans, and calculates the quality of each transformation plan; Optimality ranking module: sorts all transformation schemes according to their optimality to obtain the optimality sequence; Optimal selection module: The transformation scheme that is at the top of the optimality sequence is selected as the optimal production scheme.

Images