Latest update to top statistical software adds tools designed to eliminate false alarms when monitoring processes.
October 19, 2011
STATE COLLEGE, Pa. -- Minitab® 16.2 Statistical Software, an update to the most recent version of the leading software used for quality improvement and statistics education, adds new diagnostic tools and control charts that will make it easier to determine if a process is out of control.
The update is free for Minitab 16 users. The company also offers a free, fully functional, 30-day trial at http://www.minitab16.com
As any engineer knows, all processes have natural variation—but unusual variation may need attention. By graphing process data against an upper and a lower control limit, control charts help us recognize variation we should be concerned about.
But most people who’ve worked extensively with control charts, particularly attributes charts, have encountered “false alarms” where the control limits fell close to the mean and data points exceeded the limits even though the process was in statistical control.
The problem of false alarms due to overdispersion has been known since the 1950s, but a good solution didn’t exist until quality engineer David Laney proposed the P' and U' charts many decades later. Like traditional P and U charts, we use P' and U' charts to monitor defectives and defects, respectively. But where traditional charts assume your defective or defect rate remains constant over time, P' and U' charts assume that no process exhibits a truly constant rate, and factors that into its control limits.
As a result, these charts can give you a more reliable indication of whether your process is truly in or out of control.
Minitab 16.2 incorporates Laney’s P' and U' charts as well as an easy-to-use diagnostic tool that tells you whether or not these charts are appropriate for your process data. “With these tools, quality practitioners across all industries can be confident about detecting special-cause variation only when it truly exists,” says Michelle Paret, Minitab product marketing manager.
Minitab worked directly with David Laney to incorporate the charts into this update. An extensive interview with Laney about the origin and theory behind the P' and U' charts is available at http://www.minitab.com/training/articles/On-the-Charts-A-Conversation-with-David-Laney.aspx.
Control chart false alarms can be caused by too much variation in process data, or “overdispersion,” especially if you collected your data using large subgroups. This situation creates artificially tight control limits, so points on a traditional P chart look like they’re out of control even when they are not.
Too little variation, or “underdispersion,” can also be problematic. Underdispersion can make control limits on a traditional P chart or U chart too wide, so data points that should raise concerns appear to be in control.
Whether your data exhibit overdispersion or underdispersion, a P' or U' chart will help you distinguish true common cause versus special cause variation.
Minitab makes it easy to see whether a traditional P or U chart or a Laney P' or U' chart is more appropriate for your process data. The P Chart or U Chart Diagnostic in Minitab quickly tests your process data for overdispersion and underdispersion, and recommends the appropriate chart.
The P Chart Diagnostic in Minitab quickly tests your process data for overdispersion and underdispersion, and recommends the appropriate chart.
Minitab 16 was introduced in May 2010 and received rave reviews for its new functionality and features, especially the Assistant, a menu-based tool designed to guide users through their analyses and help them interpret their results with confidence.
Thousands of companies worldwide use Minitab to analyze data for quality improvement programs. More than 4,000 colleges and universities use Minitab for teaching and research.
Because it has been adopted by companies around the globe, Minitab 16 is available in English, French, German, Spanish, Simplified Chinese, Japanese, and Korean. Documentation is available in several other languages.
The company offers two other products. Quality Companion by Minitab® is process improvement software designed specifically to help professionals manage and execute quality improvement projects. Quality Trainer by Minitab® is an e-learning course that teaches statistics and serves as an ongoing resource for professionals who use statistics to implement quality improvement.
Minitab, LLC. is the leading provider of software for Lean, Six Sigma and quality improvement projects. Thousands of distinguished companies use Minitab software, including Toshiba, DuPont, Boeing, Royal Bank of Scotland, Nestlé and Pfizer.
Minitab, LLC. is headquartered in State College, Pa., operates offices in the United Kingdom, France and Australia, and has additional representatives throughout the world.
FOR MORE INFORMATION or materials, including screen shots, product packaging images and logos, and other graphics, contact Eston Martz in Minitab media relations: PublicRelations@minitab.com
免费获得每月电子新闻简报，其中包含最新 Minitab 新闻、教程、案例研究、统计提示和其他有用信息。
Minitab 发布新款 Companion by Minitab® 5.3
使用此网站，即表示您同意对数据分析和个性化内容使用 Cookie。 请阅读我们的政策