- By
**hilda** - Follow User

- 99 Views
- Uploaded on

Download Presentation
## PowerPoint Slideshow about ' Chapter 35' - hilda

**An Image/Link below is provided (as is) to download presentation**

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript

### Chapter 35

Statistical Process Control

Objectives

- Define and describe the objectives of Statistical Process Control (SPC).
- Select and use control charts in various situations.
- Interpret control charts.

Special versus Common Causes

- Every process has variation.
- The sources of process variation can be divided into two categories: special and common.
- Common causes (natural variation) are those inherent in the process and generally not controllable by process operators – example: Variation in raw material.
- Special causes include unusual events that the operator can adjust – example: tool wear, broken equipment.
- The principal purpose of control charts is to help the operator recognize the presence of special causes so that corrective action can be taken in a timely manner.

Selection of Variables

- When a control chart is to be used, a variable must be selected for monitoring.
- The variable is a leading indicator of special causes - one that detects special causes before others do.
- The selection of the variable(s) to be charted depends on experience and judgement.

Rational Sub grouping

- The method used to select samples for the control chart must be logical.
- For example, suppose a candy making process uses 40 pistons to deposit 40 gobs of chocolate on a moving sheet of waxed paper in a 5 x 8 array, as shown in fig 35.2, page 361.
- A good choice would be to pick upper left hand chocolate in 5 consecutive arrays as shown in fig 35.3 because they are all formed by the same piston.
- The choice of sample size depends on resources available – the larger the sample size, the more sensitive the chart.

Variables Chart

- The X bar – R chart is called a variables chart because the data results from measurements on a variable or continuous scale (R = range).
- The X bar – R chart is typically used when sample subgroup size is less than 10. See example 35.1 on page 363.
- The X bar – s chart is another variables chart. With this chart the sample standard deviation s is used to indicate dispersion.
- The standard deviation is a better measure of spread when the sample size is large. See example 35.2 on page 365.
- Control Limits: They are set at +- 3 σ. The upper control limit is designated UCL and lower control limit LCL.

Control Chart Selection

- Individual and Moving Range Chart: Sample size of one.
- For slow processes or destructive tests. Example 35.3, page 366.
- Attribute Charts: Attribute charts are used for count data where each data element is classified in one of two categories, such as good or bad. P charts (example 35.4, page 368) and np charts (example 35.5, page 371)are used to plot percentage or proportion defective, and c charts (example 35.6, page 373) and u charts (example 35.7, page 375) are used to plot counts of defects.

Control Chart Selection

- Short run Control Charts: Short run control charts should be considered when data are collected infrequently. Example 35.8, page 381.
- Moving Average and Moving Range Control Charts (MAMR): the MAMR charts may be suitable when data are collected periodically or when it takes time to produce a single item. Example 35.9, page 385.

Control Chart Analysis

- The control chart limits set at +- 3σ cover approximately 99.73% of the data.
- Points falling within the limits are due to common cause variations.
- Points falling outside the limits are due to special or assignable cause variations. Such points should be investigated immediately.

Analyzing Control Chart Behavior

- A process log may be a separate document or may be a part of the control chart.
- Entries in the process log should be identified by date and time and include all changes to the process and process environment.
- Such changes might include the occurrence of preventive or corrective maintenance, tool replacement, new parts, etc.

Analyzing Control Chart Behavior

- Some rules to test out of control conditions:
- 1. One point more than 3σ from the center line (either side).
- 2. Nine points in a row on the same side of a center line.
- 3. Six points in a row, all increasing or decreasing.
- Possible reasons for out of control conditions:
- Table 35.10, page 391.
- The six sigma professional might make additional tests for particular situations.
- For example if an increase in values represents a safety hazard, it is not necessary to wait of 6 points to take action.

Summary

- Every process has variation.
- The principal purpose of control charts is to help the operator recognize the presence of special causes so that corrective action can be taken in a timely manner.
- The method used to select samples for the control chart must be logical.
- The control chart limits set at +- 3σ cover approximately 99.73% of the data.
- Points falling within the limits are due to common cause variations.
- Points falling outside the limits are due to special or assignable cause variations. Such points should be investigated immediately.
- Entries in the process log should be identified by date and time and include all changes to the process and process environment.
- Such changes might include the occurrence of preventive or corrective maintenance, tool replacement, new parts, etc.

Home Work

- 1. What is the principle purpose of control charts?
- 2. What variable is selected for monitoring in a control chart?
- 3. Why do we need to investigate points falling outside the control limits of +- 3σ?
- 4. How should process log entries be identified and what types of entries must be recorded?

Download Presentation

Connecting to Server..