IATF 16949:2016 Clause 9.1.1.2 Identification of statistical tools and Clause 9.1.1.3 Application of statistical concepts

In the automotive industry, statistical tools and concepts are widely used to ensure product quality, process control, and continuous improvement. Various clauses o IATF 16949 arequire the application of these tools to achieve compliance and meet customer requirements. Here are some key statistical tools and concepts commonly used in the automotive industry, along with their application:

1. Control Charts (Clause 9.1.3.3 – IATF 16949): Control charts are fundamental tools for monitoring process stability and variation. They help identify special cause variation and prevent over-adjustment. Control charts are applied to various process parameters in the automotive industry to ensure that critical characteristics remain within specified control limits.
2. Process Capability Analysis (Clause 8.5.1.1 – IATF 16949): Process capability indices (e.g., Cp, Cpk) are used to assess the capability of manufacturing processes to meet customer specifications. This analysis ensures that the process is capable of consistently producing products within defined tolerances.
3. Design of Experiments (DOE) (Clause 8.5.1.2 – IATF 16949): DOE is applied during product and process design to optimize parameters and identify critical factors affecting product performance. It helps understand the relationships between process inputs and outputs.
4. Failure Mode and Effects Analysis (FMEA) (Clause 8.5.3.1 – IATF 16949): FMEA is used to analyze potential failure modes and their consequences on product performance. It aids in prioritizing risks and implementing appropriate preventive measures.
5. Statistical Process Control (SPC) (Clause 8.4.1.2 – IATF 16949): SPC involves using statistical techniques to monitor process performance and identify trends or deviations. Control charts, histograms, and other SPC tools are employed to keep processes in control and detect any signs of nonconformity.
6. Measurement System Analysis (MSA) (Clause 7.1.5.1 – IATF 16949): MSA assesses the accuracy, precision, and repeatability of measurement systems. This ensures that measurement data is reliable and trustworthy.
7. Sampling Plans (Clause 7.5.1.1 – IATF 16949): Sampling plans are applied during inspection and testing to determine the sample size and acceptance criteria. They help ensure representative data while reducing inspection costs.
8. Process Audits (Clause 9.2.2 – IATF 16949): Statistical concepts are used in process audits to assess process capability, stability, and adherence to defined control limits.
9. Continuous Improvement (Clause 10.2.1 – IATF 16949): Statistical tools and concepts underpin continuous improvement initiatives in the automotive industry. Analysis of historical data and performance metrics guides improvement efforts.

The automotive industry relies heavily on statistical tools and concepts to achieve consistent product quality, meet regulatory requirements, and drive process excellence. The effective application of these tools helps automotive companies deliver safe, reliable, and high-quality vehicles to customers worldwide.

You must plan and implement processes that measure, analyze and improve the health of your QMS. The focus of these processes must be on product and process conformity and improving QMS effectiveness. Consider using a variety of methods including statistical techniques. In planning what to track and measure, we should review the quality objectives we established and the performance indicators we established for our QMS processes and activities. You have to be careful not to overwhelm your organization with objectives as this may cause more frustration than positive results. Prioritize objectives and performance indicators to focus on meeting customer requirements and key or risk prone processes. Planning of measurement and data analyses processes must consider the methods and resources (time, manpower, computer, software, statistical tool, etc) needed to collect, organize and analyze product and QMS performance data. Use your APQP reference manual to determine what statistical methods to use for products and QMS processes and to what extent to use them. Include these methods in your control plan. Statistical methods for product development may include – variation analysis; regression analysis; dependability analysis; and prediction. Statistical methods for purchased product may include – histograms and stratification; Pareto fault analysis; sampling plans; criteria for acceptance statistics. Statistical methods to verify product characteristics and process parameters include – process capability studies; control charts; Pareto analysis; variation analysis (special cause, common cause). Statistical methods for field analysis include – dependability assessment; Pareto analysis; traceability analysis; Shainin techniques. Statistical methods for monitoring and measuring devices – refer to various techniques in your customer reference manuals – e.g. Measurement Systems Analysis (MSA) reference manual. Define and implement appropriate training and competency requirements for all personnel using statistical methods, tools and analysis.

9.1.1.2    Identification of statistical tools

The organization shall determine the appropriate use of statistical tools. The organization shall verify that appropriate statistical tools are included as part of the advanced product quality planning (or equivalent) process and included in the design risk analysis (such as DFMEA) (where applicable), the process risk analysis (such as PFMEA), and the control plan.  

The standard require statistical tools to be identified for each process during the advanced quality planning phase and included in the control plan. Within your process you will therefore need a means of determining when statistical techniques will be needed to determine product characteristics and process capability. One way of doing this is to use checklists when preparing customer specifications, design specifications, and verification specifications and procedures. These checklists need to prompt the user to state whether the product characteristics or process
capability will be determined using statistical techniques and if so which techniques are to be used.Techniques for establishing and controlling process capability are essentially the same -the difference lies in what you do with the results. Firstly you need to know if you can make the product or deliver the service in compliance with the agreed specification. For this you need to know if the process is capable of yielding conforming product. Statistical Process Control techniques (SPC) will give you this information. Secondly you need to know if the product or service produced by the process actually meets the requirements. SPC will also provide this information. However, having obtained the results you need the ability to change the process in order that all product or service remains within specified limits and this requires either real-time or off-line process monitoring to detect and correct variance. To verify process capability you rerun the analysis periodically using sampling techniques by measuring output product characteristics and establishing that the results demonstrate that the process remains capable. There are many uses for statistical techniques in establishing and controlling product characteristics.

• Receipt inspection – a technique for verifying product characteristics where sampling can be used on large quantities to reduce inspection costs and improve throughput.
• SPC – a technique for controlling product characteristics as well as controlling processes.
• Reliability prediction – a technique for establishing product characteristics where the reliability targets cannot be measured without testing many hundreds of products over many thousands of hours. (On long production runs of low value items, reliability testing is possible but with one-off systems of high value it is not cost effective; hence reliability has to be predicted using statistical techniques.)
• Market analysis – a technique for establishing product characteristics where the customer requirements are revealed by market survey and determined by statistical techniques for inclusion in specifications.
• Design by experiment – a technique where product characteristics are established by conducting experiments on samples or by mathematical modeling to simulate the effects of certain characteristics and hence determine suitable parameters and limits.

When carrying out quality planning you will be examining intended product characteristics and it is at this stage that you will need to consider how achievement is to be measured and what tool or technique is to be used to perform the measurement. When you have chosen the tool, you need to describe its use in the control plan. Statistical tools are powerful techniques that aid in analyzing data, identifying patterns, and making informed decisions in manufacturing and other processes. The organization must assess when and how to use these tools effectively to improve process understanding, control, and performance. Here are some key considerations for determining the appropriate use of statistical tools:

Process Complexity: Consider the complexity of the manufacturing process. Statistical tools are particularly useful when dealing with complex processes with multiple variables and interactions. Simpler processes may not require extensive statistical analysis.

1. Data Availability: Statistical tools require sufficient data for meaningful analysis. Evaluate whether enough data is available to support the use of statistical methods.
2. Process Stability: Determine if the process is stable before applying statistical tools. It’s essential to have a stable process to get accurate and reliable results from statistical analysis.
3. Process Capability: Assess if the process is capable of meeting the desired specifications. Statistical tools, like process capability analysis, help evaluate whether the process is capable of producing within specified tolerances.
4. Variability: Identify sources of variability in the process and determine if statistical tools can help understand and reduce this variability.
5. Root Cause Analysis: Statistical tools can aid in root cause analysis by identifying factors contributing to non-conformances or inefficiencies in the process.
6. Continuous Improvement: Statistical tools play a crucial role in continuous improvement initiatives, helping monitor process performance and identify opportunities for enhancement.
7. Understanding and Expertise: Evaluate the organization’s level of understanding and expertise in using statistical tools. Proper training and knowledge are essential for applying statistical methods accurately.
8. Resource Availability: Consider the availability of skilled personnel and resources to conduct statistical analysis effectively. Regulatory and Customer Requirements: Determine if statistical analysis is mandated by regulatory bodies or required by customers for certain processes or products.
9. Risk Assessment: Conduct a risk assessment to determine if the potential benefits of using statistical tools outweigh the associated costs and efforts.

Some commonly used statistical tools in manufacturing processes include control charts, process capability analysis, hypothesis testing, design of experiments (DOE), regression analysis, and Pareto analysis, among others. By carefully assessing the appropriateness of statistical tools, organizations can make data-driven decisions, optimize processes, and continuously improve product quality, thus enhancing overall performance and customer satisfaction.

The standard emphasizes the importance of including appropriate statistical tools in various phases of the quality planning process, such as Advanced Product Quality Planning (APQP) or its equivalent. These statistical tools are instrumental in conducting design and process risk analysis, as well as developing control plans to ensure product quality. Here’s how statistical tools are integrated into different aspects of the quality planning process:

1. Advanced Product Quality Planning (APQP): APQP is a structured approach used in the automotive industry (and other sectors) to ensure that products meet customer requirements and are launched successfully. Statistical tools are essential components of APQP to support data-driven decision-making and risk management.
2. Design Risk Analysis (DFMEA – Design Failure Mode and Effects Analysis): DFMEA is a method used during the design phase to identify potential failure modes and their effects on the product’s performance. Appropriate statistical tools can be applied in DFMEA to assess the severity, occurrence, and detection ratings of failure modes, helping prioritize and address critical risks.
3. Process Risk Analysis (PFMEA – Process Failure Mode and Effects Analysis): PFMEA is used to assess potential failure modes in the manufacturing process and their impacts on product quality. Statistical tools can assist in analyzing historical process data, identifying key process parameters, and quantifying the potential risks associated with failure modes.
4. Control Plan: The Control Plan outlines the specific control measures and activities to ensure that the manufacturing process remains in control and capable of meeting quality requirements. Statistical tools play a vital role in developing control plans by defining appropriate control limits, inspection frequencies, and acceptance criteria based on data analysis.

Examples of statistical tools commonly used in quality planning include:

• Control Charts: For monitoring process stability and identifying trends or deviations.
• Process Capability Analysis: To assess the capability of a process to meet specified tolerances.
• Design of Experiments (DOE): To optimize process parameters and identify significant factors affecting product quality.
• Pareto Analysis: To prioritize issues based on their frequency and impact.
• Failure Mode and Effects Analysis (FMEA): For risk assessment and mitigation.

By incorporating appropriate statistical tools throughout the quality planning process, organizations can ensure a systematic and data-driven approach to managing risks, enhancing product quality, and meeting customer expectations. These tools help organizations make informed decisions, continuously improve processes, and deliver products that consistently meet high-quality standards.

9.1.1.3    Application of statistical concepts

Statistical concepts, such as variation, control (stability), process capability, and the consequences of over-adjustment, shall be understood and used by employees involved in the collection, analysis, and management of statistical data.

The organization is to establish and maintain process to implement and control the application of statistical techniques. Where statistical techniques are used for establishing, controlling, and verifying process capability and product characteristics, procedures need to be produced for each application. You might for instance need a Processes for Process Control , Process Capability Analysis , Receipt Inspection , Reliability Prediction , Market Analysis , etc. The process need to specify when and under what circumstances the techniques should be used and provide detailed instruction on the sample size, collection, sorting, and validation of input data, the plotting of results and application of limits. Guidance will also need to be provided to enable staff to analyze
and interpret data, convert data, and plot the relevant charts as well as make the correct decisions from the evidence they have acquired. Where computer programs are employed, they will need to be validated to demonstrate that the results being plotted are accurate. You may be relying on what the computer tells you rather than on any direct measurement of the product.

The standard emphasizes the importance of employees understanding and using key statistical concepts in the context of data collection, analysis, and management. Statistical concepts are essential tools for ensuring quality, process control, and continuous improvement in various industries. Here are the key statistical concepts that employees involved in statistical data should understand and use:

1. Variation: Variation refers to the natural differences or fluctuations that exist in any process or data. Understanding variation is crucial because it helps identify the normal range of performance and distinguish between common cause variation (inherent to the process) and special cause variation (due to specific factors).
2. Control (Stability): Process control, also known as stability, refers to the state when a process is consistent and predictable over time. Control charts are commonly used to monitor process stability and detect any unusual patterns or trends that might indicate a process going out of control.
3. Process Capability: Process capability analysis assesses a process’s ability to produce output within specified tolerances. Employees should understand how to calculate and interpret process capability indices like Cp, Cpk, Pp, and Ppk, which provide insights into process performance and its ability to meet customer requirements.
4. Over-Adjustment (Tampering): Over-adjustment, also known as tampering, occurs when personnel intervene in a stable process in response to random variation. This can lead to increased variability and poorer process performance. Employees should be aware of the consequences of over-adjustment and the importance of letting a stable process remain in control.
5. Data Collection and Sampling: Understanding proper data collection methods and sampling techniques is crucial to ensure that the data collected accurately represents the process or population being studied.
6. Statistical Analysis Techniques: Employees should be familiar with various statistical analysis techniques, such as hypothesis testing, regression analysis, design of experiments (DOE), and analysis of variance (ANOVA). These tools help extract insights from data and make informed decisions.
7. Interpreting Graphs and Charts: The ability to interpret graphs and control charts is essential for understanding process performance and identifying trends or anomalies in the data.
8. Statistical Software: Proficiency in using statistical software, such as Microsoft Excel, Minitab, or other specialized tools, can facilitate data analysis and reporting.

Having a workforce that understands these statistical concepts enables organizations to make data-driven decisions, identify process improvements, and implement effective quality management practices. It fosters a culture of continuous improvement and empowers employees to play an active role in ensuring the overall quality and efficiency of the organization’s processes. It is not sufficient to train staff solely in the techniques they need to use – a wider appreciation of the concepts will facilitate improved application. The staff assigned to quality planning need an even wider appreciation of statistical concepts and it is probably useful to have an expert in your company upon whom staff can call from time to time. If the primary technique is SPC then you should appoint an SPC coordinator who can act as mentor and coach to the other operators of SPC techniques. All managers need a basic appreciation but those in production ought to be able to apply the techniques their staff use in order to be able to detect when they are not being applied correctly. Auditors need to be able to determine whether the right techniques are being applied and whether the techniques are being applied as directed.