Abraham Maslow once said, "When the only tool you have is a hammer, all problems begin to resemble nails." In today's bleak economic environment, many people are armed only with a hammer and are tempted to start swinging away at the myriad of problems that pop up, like a game of "Whack-a-Mole." Unfortunately, running a business is not a game, and the moles never stop popping up. Swinging the hammer, then, becomes an exercise in futility.
In the last issue I cautioned you to resist another temptation, the temptation to cut costs by cutting personnel. Instead, I urged you to utilize the aggregated brainpower of your employees in a relentless pursuit of waste in order to control your costs and to position your business to be stronger than it has ever been. You and your employees will need to learn and use problem solving skills; fortunately for you, there is more than just one tool in the problem solving toolkit. These tools are easy to master and can provide a wealth of valuable data for you to guide your improvement efforts. Here are four of the most popular quality tools.
In a cause-and-effect diagram, above, the problem statement (the effect) is listed at the right side. Possible causes are shown as branches or "bones" leading into the effect, grouped together by similarity in six categories, known as the Six M's:
-Material-Mother Nature (environment)
Causes are developed during brainstorming sessions and then grouped together along one of the branches. Causes should be specific, measurable and controllable.
A cause-and-effect diagram is a quick and effective way to organize what could be a very long list of potential causes. Once potential causes are identified, hard data must be developed to determine which of these potential causes is contributing to the problem or effect. Another benefit is that you can quickly set aside causes that are minor, allowing you to concentrate on the "big fish," puns aside.
Vilfredo Federico Pareto was a 19th Century economist who postulated that 80 percent of Italy's wealth was owned by 20 percent of the population. After researching centuries' worth of economic data he discovered that this distribution was common throughout the civilized world.
The theory here is that, like the Ishikawa diagram, one can quickly set aside causes that are minor, allowing more time and resources to be spent on the few causes that are having the greatest effect on product quality.
The example above lists 10 causes of customer complaints and rejections. Two of these, deep diecuts and print register, account for 80 percent of them. While missing labels and broken copy may be serious errors, deep diecuts and print registration are the two issues causing the most waste and more than likely the highest cost of poor quality.
Control charts are sometimes shunned by companies because they appear to be too sophisticated for the average shop worker. They are, in fact, simple to use. Control charts began to appear in the early 20th Century, long before the invention of the computer.
Walter Shewhart was a statistician with the Western Electric Company who theorized that all instances of variation in a process could be put into one of two general categories: assignable causes and chance or random causes. Shewhart stated that by identifying and eliminating known or assignable causes a process could be controlled and a repeatable, reliable product could be produced within what could be called "control."
In the chart above, reading number 5 is above 0.02, which has been predetermined to be the upper control limit for this process. By using colors, specifically red, yellow, and green, we can easily show any operator that measurements in the green area are acceptable, measurements in the yellow areas require attention, usually an adjustment in the process, and measurements falling in the red areas are unacceptable and corrective action must be taken immediately.
Scatter diagrams are used to show whether or not there is a connection between two otherwise unrelated conditions. This does not necessarily imply a direct cause and effect relationship, but if the existence of one condition can be predicted based on the measurements or values of another condition, a correlation or connection exists.
For example, customer X calls on a cold February morning to register a complaint that a label that you've produced is constantly jamming in their labeler. They have tried several rolls from different lots and the same problem keeps happening. You immediately pull your records for the lots that they've identified and cannot find anything that would lead you to believe that anything related to the production of your labels is behind the jamming. You produced the labels at different times, on different presses, using differenet dies, and different lots of label stock.
Troubled by this, you immediately dispatch your quality manager to the customer to see this problem and determine the cause. During the course of the investigation your quality manager gathers data that shows that the problems occur almost immediately upon start up of the labeling line and then quickly disappear after a few hours, only to reappear again the next day when the line starts up again. Being a good quality manager and therefore a good detective (and problem solver), he or she asks to see the temperature and humidity readings for the labeling area during the time that the jams were occuring and when they ceased. The number of times that labels have jammed are plotted along with the recorded relative humidity at the time of the label jams. It is noted that the relative humidity is very low, as it can be during the winter, and gradually increases during the morning as the heating and air conditioning equipment gradually humidifies the ambient air. At night the HVAC is shut off and the humidity levels drop. When humidity levels are low, the frequency of label jams increases significantly. When the humidity increases the frequency of label jams begins to drop.
Armed with this data your quality manager surmises that low humidity is somehow related to the problem and asks your customer not to turn the HVAC off that night. The following morning the line starts up and, with humidity levels within normal range, the labels that had been jamming now run without incident.
It hasn't been shown to be a direct cause and effect relationship, but the data suggests that low humidity is somehow contributing to the label jams, perhaps due to a slight static charge that is causing the leading edge of the label to curl back toward the peel bar.
These are just a few of many tools that you can use to improve your processes. Like any tool, they're useless if they're not used and can be dangerous if used improperly. Master these few tools and you can build a stronger foundation for your business.