Quality Improvement

Since W. Edwards Deming emphasized the role of statistical thinking in understanding and solving problems, quality improvement has been a major theme in worldwide industry. Many companies have adopted the use of statistical techniques, and the application of these methods has become well established in aerospace, automotive, electronics, pharmaceutical, semiconductor and other manufacturing industries.

In recent years, statistical methods have also been explored by banks, insurance companies, government agencies and health care organizations interested in improving the quality of their services to customers.

While there are many quality improvement tools on the market today, only SAS provides a complete, comprehensive and integrated platform for data analysis. With SAS, you can easily access data from any source, perform data management, carry out statistical analysis and then present your findings in a variety of reports and graphs -- all within a single, easily managed software environment.

SAS/QC software, an integrated component of the SAS analytic platform, provides a wide range of specialized tools for all quality improvement efforts within the entire organization, from designing experiments and assessing product reliability to monitoring process stability and determining process capability. And since SAS remains committed to its long tradition of constantly enriching its statistical offerings, you know that you will have access to the most up-to-date quality improvement techniques not just today, but well into the future.

Understand processes and pinpoint critical problems. The dynamic graphics environment in SAS/QC makes it easy to prioritize quality improvement activities. Pareto charts based on a single set of quality problems, or classified by multiple variables of interest allow for quick identification of those causes that require focused improvement efforts. Analysis of means techniques allow the graphical comparison of response measurements from a number of groups to determine which are different.
Establish control and reduce variation. You can monitor process data with a variety of control charts, including Shewhart charts -- the most popular method for studying process variation. You can reuse control limits created during a previous analysis, automatically adjust control limits for varying sample sizes and perform tests for special causes (Western Electric rules, runs tests). You can also create cumulative sum control charts for means or individual measurements, generate control charts for uniformly or exponentially weighted moving averages and produce historical control charts to display the evolution of a process over time.
Determine process capability. After establishing statistical control, calculate capability indices and use histograms (optionally superimposed with specification limits and fitted curves), quantile-quantile plots and probability plots to determine how well your product meets design specifications.
Design experiments to improve products or processes. The ADX Interface guides you through the entire process of designing and analyzing statistical experiments. You can generate factorial, fractional factorial, and mixed-level designs, with or without blocking. And, for situations where standard designs are not appropriate, you can construct A-, G-, and D-optimal designs.
Assess product reliability. Understanding the risk of product or component failure allows for the formulation of warranty plans for future products and the schedule maintenance actions to assure a quality customer experience. Reliability engineers and statisticians can construct probability plots and fit life distributions with right and interval censored data, fit regression models, including accelerated life test models and analyze recurrence data from repairable systems.

Powerful and versatile with a wide range of statistical and graphical methods, SAS provides software for quality improvement activities across organizations.

Quality Improvement

Case Studies

Technical Support

Our Alliance