GB/T 40681.4-2021Statistical methods in monitoring process capability and performance - Part 4: Process capability estimates and performance measures (English PDF)

Foreword

The plan of GB/T 40681 "Statistical Methods for Monitoring Production Process Capability and Performance" is divided into the following 8 parts.

---Part 1.General principles and concepts;

---Part 2.Process capability and performance of time-dependent process models;

---Part 3.Research on machine performance of discrete product measurement data;

---Part 4.Process capability estimation and performance measurement;

---Part 5.Process capability and performance estimation of counting characteristics;

---Part 6.Multivariate normal process capability analysis;

---Part 7.Measuring process capability;

---Part 8.Equipment performance analysis of multi-state production process.

This part is Part 4 of GB/T 40681.

This section was drafted in accordance with the rules given in GB/T 1.1-2009.

This part uses the redrafting method to modify and adopt ISO 22514-4.2016 "Statistical Methods Capability and Performance in Process Management No. 4

Part. Process Capability Estimation and Performance Measurement.

Compared with ISO 22514-4.2016, this part has made the following structural adjustments.

---The 3.1 of ISO 22514-4.2016 is deleted, and the "assessment and measurement methods involved in this part" in 3.1 are only applicable to metrological

Data is not applicable to attribute data. For relevant content of attribute data, please refer to ISO 22514-5" to adjust to Chapter 1, and revise

Changed to "This section applies to the process capability and performance evaluation of measurement data in the common production process, and does not apply to attribute characteristics

Process capability and performance evaluation. "

The technical differences between this part and ISO 22514-4.2016 and the reasons are as follows.

--- Delete "P0.135%. 0.135% quantile", "P99.865%. 99.865% quantile", "pi. geometric constant" and "e. Euler constant" in 2.1

In addition, the reason is that the definition is repeated;

--- Modify the "Palpha.alpha quantile" in 2.1 to "Palpha standardized Pearson curve alpha quantile" to be more accurate;

--- Replace "sigmat" in 2.1 and the full text with "sigmat. total standard deviation of the process";

--- Delete "Y1, Y2" in 2.1 and the full text. The meanings are different in different figures, and they have been explained in the text without symbol explanation;

---Modify the expression "CpkL =0.86, CpkU =0.91" involving numerical calculation in 4.8 to the estimated form "CpkL =0.86,

CpkU=0.91", modified the incorrectly used estimated value symbol;

--- Estimate the parameters of µ and sigma2 in C.3.1 "µ=X=

Nsum

i=1

Xi" and "sigma2=

N-1sum

i=1

(Xi-X)2", amended to

"Mu=logX=

Nsum

i=1

logXi" and "sigma2=

N-1sum

i=1

(logXi-logX)2";

--- Modify "Xalpha=Y-xi(-log(1-alpha))-beta" in C.5 to "Xalpha=Y xi[-log(1-alpha)]

beta".

This section has made the following editorial changes.

---Modified the standard name;

--- Modified the non-standard expression of the data index in Table 3;

--- Modify "Kl0.95" in D.1.3 to "Kl";

--- Modify "Ku0.95" in D.1.3 to "Ku;

---The full text formula adds the serial number;

---Revised references.

This part is proposed and managed by the National Standardization Technical Committee for the Application of Statistical Methods (SAC/TC21).

Drafting organizations of this section. Beijing University of Aeronautics and Astronautics, China National Institute of Standardization, Inner Mongolia Mengniu Dairy (Group) Co., Ltd.,

Huzhou Rongke Building Material Technology Co., Ltd., Shanxi Aerospace Tsinghua Equipment Co., Ltd., Huzhou Mingfeng Enterprise Management Consulting Co., Ltd., Beijing

Electronic Engineering General Research Institute, the 12th Research Institute of China Electronics Technology Group Corporation, Liaocheng Zhuoqun Auto Parts Co., Ltd.

Introduction

Many organizations have already embarked on a continuous process improvement strategy. In order to comply with the strategy, any organization needs to evaluate the performance of its key processes.

Capability and performance. The methods given in this section are intended to help carry out this aspect of management. Managers need to continue to pay attention to these assessments in order to

Carry out continuous improvement activities when needed.

Recently, this part has been greatly developed. The most fundamental change is to conceptualize the capability conditions and performance conditions named in this section.

The fundamental difference lies in whether the process characteristics are in a statistically controlled state (in a statistically controlled state, it is called capability, otherwise it is called performance). therefore,

Two sets of indicators corresponding to capabilities and performance are given in the relevant clauses. Because it has been found in the industry that due to improper calculation and publication

Some companies are misled in terms of their true capabilities, so it is increasingly necessary to make a strict distinction between capabilities and performance.

The progress of this part lies in the transition from general conditions to specific conditions, and this improvement makes the generalized formula present a more specific form of expression.

A large number of existing documents describe the importance of understanding the work process in any organization, for example, the production process or the information processing process.

With the intensification of sales competition between enterprises, in addition to the sales price of important products or services, during the use of products or services, purchase

The expenses that buyers will bear are also becoming more and more important. The goal of each supplier is no longer to just meet the specifications, but to continue to reduce

Mutations.

In a world where competition continues to increase, continuous improvement can reduce the loss of product failure and the cost of survival. Because as the variation decreases,

The cost of evaluation will decrease, the demand for testing products may disappear or the sampling frequency may be reduced.

In order for an organization to be able to evaluate the capabilities and performance of its suppliers, process capability and performance evaluation is necessary. These organizations will find the package

The indicators contained in this section are useful in this regard.

Quantifying the variation of a process can make judgments about its suitability and ability to meet certain given requirements. The following paragraph

And clauses provide philosophical principles that need to be understood to determine the capabilities or performance of a process.

All processes have inherent variability. This section does not specifically explain the meaning, reason, source and how of inherent variation

The process is based on the premise of the stable existence of inherent variation.

Process owners need to work hard to understand the source of their process variation. Will draw process flow diagrams, identify process input and output methods, and factors

Fruit diagrams (fishbone diagrams) are used together reasonably to help identify these process variations.

The distinction between short-term variation and long-term variation is very important for the use of this section.

Process capabilities determined by long-term variability may vary greatly.

When considering short-term variation, it can only be carried out based on the shortest-term variation, sometimes called machine research, which is described in ISO 22514-3.

The methods needed to carry out such studies are beyond the scope of this section; it should be pointed out that these studies are important and useful.

It should be pointed out that this section only provides the point estimate of the process capability index. In view of the actual demand, it is recommended to calculate the position of these indexes as much as possible.

Confidence interval.

Statistical methods for production process capability and performance monitoring

Part 4.Process capability estimation and performance measurement

1 Scope

This part of GB/T 40681 gives the commonly used process capability and performance for the two types of normal distribution and non-normal distribution.

Metrics provide an evaluation method of process capability and performance.

This section applies to the process capability and performance evaluation of metrological data in common production processes, and does not apply to the process capability of quantitative data.

Force and performance evaluation.

2 Symbols and abbreviations