Integrated control scheme for optimal four-parameter continuous sample inspection (OCSP-3) and process yield index (Spk)

Abstract

The invention discloses an integrated control scheme of an optimal four-parameter continuous sampling inspection (OCSP-3) and a process yield index (Spk). The integrated control scheme comprises the following steps: (1) giving an average detection quality limit AOQL, a long-term average inspection number limit AFIL, a producer risk value and a consumer risk value; (2) collecting 100 latest production data and carrying out normality test; (3) calculating a process qualified rate estimator # imgabs0 # and a lower confidence limit q *; (4) comparing the estimated quantity of the qualified rate of the process with an effective working interval (AOQL, 10AOQL), and making decisions of exempting inspection, stopping production and executing an optimal scheme; (5) calculating a continuous qualified product number io3 and a test score fo3, and determining a sample size n and a key value s0 of the integrated control scheme; (6) executing an integrated control scheme (io3, fo3, 50, 4, n and s0), keeping recording n latest data to estimate a process yield index estimator # imgabs1 #, comparing the process yield index estimator # imgabs1 # with s0, and making decisions of continuing execution, recalculating scheme parameters and stopping production so as to ensure stable process quality; the method is suitable for the condition of limited inspection capability, the quality and cost requirements are met at the same time through an integrated control scheme, and the producer risk and the consumer risk are reduced. The integrated control scheme can be used for process quality control of production processes with tolerance upper limit and tolerance lower limit requirements and limited inspection capability on a production line.

Description

Technical Field

[0001] The present invention belongs to the field of product manufacturing process control, and is specifically an optimal four-parameter continuous sampling test (OCSP-3) and process yield index (S pk ). Aiming at the situation that both the producer risk and the consumer risk of the Dodge type CSP solution are high, the present invention uses the process yield index estimator distribution function to establish an inspection strategy that controls both types of risks at a given level. The integrated control scheme separates risk control from the CSP solution, and uses the metrological characteristics of the inspection data to control both the producer risk and the consumer risk at a given level through the risk control scheme while meeting the quality requirements. It can be used for process quality control of production processes with limited inspection capabilities that have upper and lower tolerance requirements on the production line. Background Art

[0002] Continuous sampling inspection is the identification and recognition of unit products when they pass the inspection point. Unit products that fail the random inspection or full inspection are eliminated, and the remaining unit products that have not been inspected continue to be sent down the production line. It is suitable for the inspection of work-in-progress that is passing through the inspection point and is not grouped for inspection. Generally, there is not enough storage space for work-in-progress, the inspection is easy to operate, the quality of the production process and raw materials is stable, and the inspection is non-destructive. In practice, it is more suitable for the quality inspection of transmission lines and processes under automated production methods. Work-in-progress refers to finished products, semi-finished products, components, raw materials, data or other physical objects that are passing through the inspection point. It is particularly pointed out that the continuous sampling plan does not need to be grouped into batches. It can be produced and inspected at the same time, saving the time spent on forming inspection batches and saving storage space. CSP-3 is one of the CSP series solutions, with four solution parameters, which realizes the most comprehensive optimization of the CSP series solutions. However, it was not selected into the standard due to the complexity of parameter formulation and operation procedures. Therefore, we propose an integrated control solution of optimal four-parameter continuous sampling inspection and process yield index. Summary of the invention

[0003] To achieve the above object, the present invention provides the following technical solutions: An optimal four-parameter continuous sampling test (OCSP-3) and process yield index (S pk ) includes the following steps: (1) Given the current average outgoing quality limit AOQL and the long-term average inspection number limit AFI L , producer risk value α and consumer risk value β parameters; (2) Collect the latest 100 (note: the initial sample size n = 100) production data for normality test. The normality test uses the normal fit (p-value). If p-value ≥ 0.05, that is, the production data obeys the normal distribution, the production process is a controlled process, and continue to step (3); otherwise, the production equipment should be adjusted. (3) Calculate the estimated process yield and the lower confidence limit q * ; The calculation formula for the process yield estimate is: in, Sample mean Sample variance USL is the upper tolerance limit, LSL is the lower tolerance limit, n is the number of samples, x i (i=1,...,n) is the sample observation value. The formula for calculating the lower confidence limit of the process qualified product rate is: (AOQL, 10AOQL] in, represents a chi-square distribution with n-1 degrees of freedom and quantile α. (4) Estimation of process qualified product rate Compare with the effective working range and make a judgment: If Exemption from inspection; Check and adjust; Continue with step (5). The calculation formula for the process yield estimate is: (5) Calculate the number of consecutive qualified products i for the optimal four-parameter continuous sampling test o3 And the test score f in the score test phase o3 , determine the sample size n of the integrated control solution and the key value s0 of the process yield index; Quality control parameters of the integrated control scheme i o3 and f o3 is the solution of the following system of equations: The risk control parameters n and s0 of the integrated control scheme are the solutions of the following system of equations: in, p IQL=AOQL / (1-AFI L ). (6) Implementation of integrated control scheme (i o3 ,f o3 ,50,4,n,s0); (7) Keep records of the latest n data and estimate and process yield index estimator S pk A natural estimate of is: (8) When and When , continue to execute step (7), if and Continue to execute step (2) and recalculate the solution parameters to ensure the process output quality AOQL, otherwise stop production for adjustment. Compared with the prior art, the present invention has the following advantages: 1. This optimal four-parameter continuous sampling inspection and process yield index integrated control solution fully utilizes the quality improvement function of OCSP to ensure that qualified process quality can be obtained regardless of how the process quality fluctuates. 2. This integrated control scheme of optimal four-parameter continuous sampling inspection and process yield index can control both types of risks at a given level at the same time, thereby meeting the risk control needs of process status judgment under quality and cost constraints, and making up for the defect that both types of risks in the CSP quality control scheme are very high and cannot be controlled at the same time. 3. This integrated control scheme of the optimal four-parameter continuous sampling inspection and process yield index separates the process state judgment risk from the CSP scheme, which is an innovation in risk control methods. The risk control scheme based on the process yield index realizes the simultaneous control and quantitative control of two types of risks, which improves the process risk control effect. 4. The parameters of the integrated control scheme of the optimal four-parameter continuous sampling inspection and process yield index have a one-to-one correspondence with the four types of constraints. The integrated control scheme does not increase the inspection workload; the counting characteristics of the inspection data are used to control the process quality, and the metrological characteristics of the inspection data are used to control the two types of risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0005] The following will be combined with the appended Figure 1, combined with examples, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of invention protection. See also Figure 1 As shown, a technical solution provided by the present invention is: An integrated control scheme for optimal four-parameter continuous sampling inspection and process yield index. When using, follow the steps below. (1) Given the current average inspection quality limit AOQL = 0.00143 and the long-term average inspection number limit AFI L = 0.5, producer risk value α = 0.01 and consumer risk value β = 0.01; (2) Collect the latest 100 production data (unit: mm) for normality test; The normality of the 100 data is p-value = 0.371 ≥ 0.05, that is, the production data obeys the normal distribution, the production process is a controlled process, and step (3) is continued; (3) Calculate the estimated process yield and the lower confidence limit q * ; The process yield estimate is: The lower confidence limit of the process qualified product rate is: q*=0.995224 (4) The estimated value of process failure rate is: If the system is in the valid working range, continue to step (5). (5) Calculate the number of consecutive qualified products i for the optimal four-parameter continuous sampling test o3 And the test score f in the score test phase o3 , determine the sample size n of the integrated control solution and the key value s0 of the process yield index; Quality control parameters of the integrated control scheme i o3 and f o3 By solving the simultaneous equations, we can obtain: i o3 =2394 f o3=0.0343 The risk control parameters n and s0 of the integrated control scheme can be obtained by solving the simultaneous equations: n = 1720 s0=1.1991 (6) Implement the integrated control scheme (2394, 0.0343, 50, 4, 1720, 1.1991) and keep records of the latest data to ensure stable process quality. The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An optimal four-parameter continuous sampling test (OCSP-3) and process yield index (S pk ) includes the following steps: (1) Given the current average outgoing quality limit AOQL and the long-term average inspection number limit AFI L , producer risk value α and consumer risk value β parameters; (2) Collect the latest 100 (note: the initial sample size n = 100) production data for normality test. The normality test uses the normal fit (p-value). If p-value ≥ 0.05, that is, the production data obeys the normal distribution, the production process is a controlled process, and continue to step (3); otherwise, production inspection and equipment adjustment should be carried out. (3) Calculate the estimated process yield and the lower confidence limit q * ; The calculation formula for the process yield estimate is: in, Sample mean Sample variance USL is the upper tolerance limit, LSL is the lower tolerance limit, n is the number of samples, x i (i=1,...,n) is the sample observation value. The formula for calculating the confidence limit of the process qualified product rate is: in, represents a chi-square distribution with n-1 degrees of freedom and quantile α. (4) Estimation of process qualified product rate Compare with the effective working range (AOQL, 10AOQL] and make a judgment: If Exemption from inspection; Check and adjust; Continue with step (5). The calculation formula for the process yield estimate is: (5) Calculate the number of consecutive qualified products i for the optimal four-parameter continuous sampling test o3 And the test score f in the score test phase o3 , determine the sample size n of the integrated control solution and the key value s0 of the process yield index; Quality control parameters of the integrated control scheme i o3 and f o3 is the solution of the following system of equations: The risk control parameters n and s0 of the integrated control scheme are the solutions of the following system of equations: in, p IQL =AOQL / (1-AFI L ). (6) Implementation of integrated control scheme (i o3 ,f o3 ,50,4,n,s0); (7) Keep records of the latest n data and estimate and process yield index estimator S pk A natural estimate of is: (8) When and When , continue to execute step (7), if and Continue to execute step (2) and recalculate the solution parameters to ensure the process output quality AOQL, otherwise stop production for adjustment.

2. The Optimal Four Parameter Continuous Sampling Test (OCSP-3) and Process Yield Index (S pk ) can be used for process quality control of production processes with limited inspection capabilities that have upper and lower tolerance requirements on the production line.

Images