Statistical Quality Control For Process Improvement

 

 

Statistical Quality Control (SQC) is a term for an approach to solving problems in production process. The term “statistical quality control” is misleading. First, the quality is not the only focus of these techniques. Identical methods are useful for improving production rates, yield, and activities in sales and other non-manufacturing areas. Second, the only universal technique for SQC is careful logical reasoning; complex statistical methods are generally helpful but not essential. So SQC is a method of management than of control in any narrow sense. It is a way of learning systematically, rather than haphazardly.  SQC is a set of statistical methods used to monitor and control processes to ensure that they operate efficiently and produce products or services that meet specific quality standards. When applied to process development, SQC plays a crucial role in identifying and addressing variations in the development process to improve the quality of the final product.

One approach to improve a process is the hit-or-miss approach. In this guess at a cause of problems, assume it is really the cause, implement something that fixed that cause and see whether problem goes away. This ad hoc approach is simple and common, but it is usually slow and costly way to improve a process.

.  

Basic Approach:

Using SQC to improve a process can be divided into five major steps:

1.      Awareness that a problem exists.

2.      Determine the specific problem to be solved

3.      Diagnose the causes of the problems

·         Formulate hypotheses about possible causes

·         Test the hypotheses, using either existing information or new experiments.

·         Continue until one or more causes are proven.

4.      Determine remedies

5.      Implement remedies

6.      Implement controls to hold the gains.

This approach is designed to find underlying problems which cause defects or other undesirable results, and fix them permanently.

Here are some key aspects of using Statistical Quality Control (SQC) in process development:

Define Quality Objectives:

Clearly define the quality objectives for the process under development. This involves specifying the critical characteristics and quality standards that the final product should meet.

 

Data Collection:

Collect data on key process parameters and quality characteristics. This may include measurements, observations, or other relevant data points that provide insights into the performance of the process. 

Statistical Analysis:

Utilize statistical tools and techniques to analyze the collected data. Common statistical methods include:

·         Descriptive statistics to summarize and describe data.

·         Control charts to monitor process stability and identify any unusual patterns or trends.

·         Capability analysis to assess whether the process is capable of meeting quality specifications.

 

Control Charts:

Implement control charts to monitor the stability of the process over time. Control charts help identify variations that may indicate special causes of variation, allowing for timely corrective action.

Process Capability Analysis:

Assess the capability of the process to meet specifications. This involves comparing the inherent variability of the process with the allowable variability specified by quality standards.

Root Cause Analysis:

When variations are detected, perform root cause analysis to identify the underlying factors contributing to the variation. This may involve investigating equipment issues, material quality, operator training, or other relevant factors.

Continuous Improvement:

Implement corrective and preventive actions based on the insights gained from the statistical analysis. Use the information to make informed decisions about process adjustments, improvements, or changes.

Documentation and Reporting:

Maintain thorough documentation of the SQC activities, including data collection procedures, analysis results, and corrective actions taken. This documentation is valuable for regulatory compliance and continuous improvement efforts.

Training and Communication:

Ensure that personnel involved in the process development are trained in SQC principles and methods. Effective communication is crucial for the successful implementation of SQC practices.

Adaptability:

SQC is not a one-size-fits-all approach. It should be adaptable to the specific needs and characteristics of the under develop process. Regularly review and update SQC methods based on changing conditions and requirements.

By incorporating Statistical Quality Control into process development, organizations can enhance product quality, reduce variability, and improve overall process efficiency. It provides a systematic approach to monitoring and improving processes, leading to more consistent and reliable outcomes.

 Seven Useful Diagrams:

There are at least seven diagrams that aid in Statistical quality control. They can be used to help an individual investigator. The charts typically help in formulating and testing hypotheses. They are: 

1.Pareto Chart:

It shows the magnitude of each of many effects. Used to distinguish critical problems (step 2)

2. Cause and Effect Diagram (Fishbone):

A method of showing possible causes of a problems. An aid to brainstorming and hypothesis generation (step 3)

3.Process Flow Chart:

Useful for showing linkages among parts of a process (step 3 and 4).

4.Run Chart:

It is a sample graph. It can be examined to see whether the process is subjected to changes or behaves consistently over time.

5.Frequency Histogram:

It shows the distribution of some variable that is thought to be important. It is useful for hypothesis testing.

6.Scatter Diagram:

It shows correlation between two variables, typically a problem and a potential causes (step 3). To examine interaction among several variables, use multiple regressions.

7.Control Chart:

This is similar to run chart but used for operational control. It shows upper and lower allowable limits for a process variable. When process exceeds those limits, action should be taken to adjust the process controls.

 Cause and Effect Analysis:

Cause and effect analysis (Fishbone analysis), also known as root cause analysis, is a technique to stimulate ideas about the causes of a problem or phenomenon. It helps to understand the relationships between various factors and their effects on a particular outcome. This analysis is often represented visually through tools like fishbone diagrams or Ishikawa diagrams. The step-by-step guide line to conduct a cause and effect analysis: 

Steps for Cause and Effect Analysis:

Define the Problem: Clearly articulate the problem or the outcome you want to analyze.

Identify the Effect: Determine the specific outcome or problem that you are trying to understand.

Create a Fishbone Diagram:

·         Draw a horizontal arrow representing the timeline of the problem.

·     Identify major categories of potential causes and draw diagonal lines (bones) stemming from the timeline, representing these categories. Common categories include People, Process, Equipment, Materials, Environment, and Management etc.

Identify Possible Causes:  Brainstorm and list potential causes within each category. Encourage input from team members and stakeholders.

Probe Deeper:  For each identified cause, ask "Why?" to probe deeper and uncover the root causes. Continue this process until reaching a fundamental cause.

Analyze Causes:  Assess the identified causes to determine their relevance and importance. Some causes may be more influential than others.

Prioritize Causes:  Prioritize the causes based on their impact and feasibility of addressing them. This helps in focusing efforts on the most critical issues.

Develop Solutions:  Once the root causes are identified, brainstorm and develop potential solutions to address each cause.

Implement Solutions:  Put the selected solutions into action. Monitor and evaluate the effectiveness of these solutions.

Monitor and Adjust:  Regularly monitor the situation to ensure that the implemented solutions are having the desired impact. Adjust strategies as needed.

 Tips for Cause and Effect Analysis:

Collaboration: Involve relevant stakeholders and team members in the analysis process to gather diverse perspectives.

Data Collection: Use data and evidence to support your analysis. This enhances the credibility of the identified causes.

Iterative Process: Cause and effect analysis may require iteration. As you implement solutions, new insights may emerge, leading to further analysis and adjustments.

Continual Improvement: Use the findings from the analysis to implement changes and improvements in processes or systems.

 Conclusion:

By systematically examining the causes and effects of a problem, organizations can make informed decisions and implement effective solutions. The managerial judgment is required to decide what to investigate and how to examine. The “right” answer depends on many tactical and strategic factors. This includes:

·         The importance of the product.

·         The maturity of the product and manufacturing process

·         The current level of understanding about the process and what affects it.

·         The availability of resources to devote to experimentation, including engineers and surplus production capacity.

·         Whether knowledge gained about this one product and process will have value and possible strategic leverage for other products and process.

·         How much effort to put into each problem and each hypothesis.

For more interesting articles click the link: Academic Learning Solutions