Why Process Control Charts are a Roadmap to Improvement
Organizations that purchase the Lean or Six Sigma business methodology, et al that are dedicated to continuous improvement, often use a bunch of visual management tools to realize consistency and introduce positive change. Kanban, huddle boards, and value stream maps are all very fashionable and effective. Process control charts are another valuable visual management tool for recognizing and reacting to process variation. Here are the small print about why they're so useful.
Statistical Process Control
Statistical Process Control It is probably helpful to start with a definition of process. A process is sort of simply anything that gets done. It may well be putting gas in your car, filling out a written account, delivering ASCII text file to QA, or checking during a patient. Each of those activities ends up in some output. Sometimes it's a product, but often it's a service or a deliverable to the subsequent process. additionally to the results of the method, data is additionally generated. Statistical process control is that the act of using that data to form the method better. the information may be associated with timeliness, cost, quality, or quantity.
Visual Process Control
A process control chart is a method of visualizing a process variable over time. They are useful in almost every industry (although not for every problem). By plotting process results on a chart, you are able to recognize variation in the process and understand the nature of that variation. Like many visual management tools, control charts get you to that “Ah-Ha!” moment more quickly than simply reading a list of data points.
Variation in process variables must always be expected. Process control charts don’t eliminate variation and therefore the object isn't to form a process always produce “average” results every single time. There's always some amount of variation in every process. What process control charts do is make the sort of variation obvious. They create a stark distinction between “common cause” variation, which must be addressed within the conventional process, and “special cause” variation, which is addressed outside of the method. Here’s an example. Let’s say it always takes you between half-hour and 40 minutes to drive to figure each day. You can’t predict your exact time on any given day, but you'll be able to be pretty sure, supported this chart and therefore the calculated control limits, that it'll be between 30 and 40 minutes. These limits are calculated, not arbitrary... and that they are calculated supported the conventional day-to-day variation that we experience. The reason that you simply can’t predict the precise amount of your time it takes to urge to figure whether or not you permit at the identical time of day and take the identical route is common cause variation, maybe you miss a light-weight someday, but not the subsequent or traffic runs a touch slower from time to time. These are expected, normal events that can’t change, so you merely build them into your expectations. Now imagine someday you forget to pay your state registration fee and you get pulled over on your thanks to work. That day, it'd take you hour to urge there. Getting pulled over would be a special cause variation. therein case, a separate process (paying your registration) and an unusual event impacted the method you're tracking (driving to figure.) If you had a process control chart for driving to figure, it'd appear as if the chart at right. You don’t should stare at that chart for very long to comprehend that something unusual happened on the day you bought pulled over. That day aside, you've got a predictable process for planning to work. It may be said to be a process that's “in control," which implies performance is stable and it's predictable within a specific range.
The reason it's important to know whether a variation is common or special cause, is that the steps you're taking to boost the method are different in either case. Let’s say you opt you would like to urge to figure faster. If you still leave at the identical time of day and take the identical route, you'll be able to expect the identical average time, with the expected normal variation. to cut back the time, you would like to alter the method. Perhaps you permit earlier or invite a fan to ride along and take the carpool lane. These changes within the method itself may lead to improvement. However, they won’t address the special reason for getting pulled over that just the once. so as to resolve that issue, you would like to try and do something about your process for paying your registration, not your process for driving to figure. That’s why process control charts are so useful. They tell you where to seem for opportunities for improvement. They also allow you to know if the improvements that you simply have implemented are producing the intended result against the variable that you simply are tracking. Control charts can easily be applied to any process that produces data, leading to a transparent path to raised performance.