Are you quality obsessed? 7 steps to an effective quality system

How to establish an effective quality system: Statistical process control (SPC) can help.

By Steve Wise July 12, 2012

Manufacturers are under increasing pressure to ensure product quality, especially given the growing number of strict industry regulations. If the smallest part or ingredient is out of spec and a recall occurs, it is not only the manufacturer but the entire supply chain that is at fault. All it takes is one negative headline about a defective engine or a contaminated package of spinach to jeopardize a brand’s reputation.

Companies that avoid negative press are the ones that truly embrace quality as a business function and recognize the value of an enterprise quality system. These manufacturers exert a tremendous amount of effort to secure their respective industry standards—whether Six Sigma, the Good Housekeeping Seal of Approval, or positive reviews on CNET—and interestingly, all demonstrate the same habits when it comes to ensuring the quality of the products they produce.

Emulating the habits of quality-obsessed manufacturers can ensure that a quality product runs through a facility to reach the consumer, while making the manufacturing organization more effective along the way. Seven tips follow.

1.  Brag about quality. Customer satisfaction can make or break a manufacturer. Therefore, it is imperative to give upper management the data they need to build customers’ confidence in products offered. Quality claims can be made, but sometimes words aren’t good enough. Buyers want to see data that is meaningful to them, not just the required Cp (variation measurement) and Cpk (center tendency measurement). A statistical process control (SPC) system offers upper management data that quantifies quality in clear terms.

Most importantly, do not hide data from top brass; transparency is vital. To begin effectively bragging about quality, create a list of metrics and divide them into two groups: metrics that are impressive now, and metrics that, if improved, will help achieve higher organizational goals.

2. Do what counts. Now that you know the importance of data, keep in mind that more does not necessarily mean better. Data collected must have value and should be concise. Consider the following when determining if collected data has meaning. If the data values significantly change from the norm during production, would the change lead to a corrective action? Also, if a corrective action is needed, is there a procedure in place to deal with it?

Prior to monitoring a process, ensure you have an effective sampling strategy and systems in place to take corrective action. Decide which employees can take action based on real-time data intelligence and provide them with the necessary reports to do their job the best they can.

Data consumers: Who needs data? In the case where data is already being collected and reported, be bold and challenge the status quo. At one large airframe manufacturer, a new manager wanted to find out who needed, or was even reading, numerous scheduled reports from his department. He stopped all publications and waited for the phone to ring. All answers arrived in just a couple of weeks. Using the feedback from the few that contacted him, he completely revamped reporting content and schedules.

3.  Give the process a leading role. True SPC involves three components: the process, the test characteristics being monitored, and the part being produced. When collecting data, the most important of these factors is the process, as it controls the consistency of the final product and influences manufacturing as a whole.

The process is needed to produce test characteristics, and test characteristics are needed to produce parts. Therefore, it is vital to include processes in data collection and analysis. You will achieve new insights by monitoring even the seemingly smallest pieces of the process, such as which nozzle filled a particular container.

Remember to identify the machines (processes) in your plant that are most critical to quality, and make sure you have a system that can measure their performance.

4. Keep it simple. With the right SPC software, capturing data should be a simple process. If data collection is difficult, an organization risks capturing inordinate amounts of meaningless data. Select an SPC platform that displays only what is helpful to the user. Visualizations, charts, and even user-friendly spreadsheets are ideal. The software should also automate calculations and prompt users when specific quality checks are due. It’s important to ensure that shop-floor systems are optimized for the current shop-floor environment so data can be accurately collected.

5. Expect a value chain reaction. Suppliers are an extension of the factories they feed, and the quality of the suppliers’ products directly affects final output.

For example, what happens if an automotive manufacturer unknowingly assembles a car with a supplier’s defective transmission? With cloud-based SPC, manufacturers can extend quality throughout the supply chain—all the way down to the suppliers—so the faulty transmission, for instance, never even makes it to the production line.

The transparency provided by cloud-based SPC will ultimately increase profitability for the supplier and for the manufacturer by reducing scrap. If a supply chain-wide, cloud-based SPC solution is under consideration, begin by discussing the value of sharing real-time data with customers and suppliers.

6. Always be vigilant. Control chart plot points will send one of two messages: Do something, or do nothing. As the “first life of a data point,” both are equally important.  When you see the “do something” message, you should decide on a course of action simply by comparing the data point with the previous plot point.

Understanding natural process variations can help you know when to avoid taking action. Don’t tamper with the process if signals tell you to “do nothing.” Make the control charts more meaningful by finding the earliest possible point to capture data and be vigilant with responding to those messages.

7. Always dig deeper. What happens to all the real-time data you’ve collected? A process capability database houses the once real-time data. You can use this database to gain insight on how to improve processes in the future. Even the simplest data, such as lot numbers and raw material suppliers, can provide value and help pinpoint their effects on a process’s output.

Furthermore, the process capability database can make additional calculations that can lead to more accurate business decisions on a variety of levels, including make/buy, scheduling, and raw material usage. You can improve your organization’s ability to use data analysis to “predict the future” by identifying attributes that affect process outputs.

These seven simple steps will increase your organization’s understanding of the impact quality has on operational efficiency and the bottom line. Data is your greatest asset for gaining visibility into causes of quality issues, and quick analysis often equals quick resolution. The correct approach to quality control yields benefits ranging from reduced scrap, rework, and warranty claims to audit and recall management; from supplier benchmarks to customer satisfaction.

Perhaps more importantly, these seven steps lay the framework for making your company more data driven. By working smarter, you can eliminate day-to-day headaches caused by fighting fires and replace them with a balanced, systematic approach to quality control.

– Steve Wise is vice president, statistical methods, InfinityQS, and author of an ebook called, “7 Habits of Quality Obsessed Manufacturers,” available at This article was edited by Mark T. Hoske, content manager CFE Media, Control Engineering, Plant Engineering, and Consulting-Specifying Engineer,