Statisitical_Process_Control

Statistical Process Control: Morning Routine By Monique T. Guillory-Jackson University of Phoenix Operations Management OPS/571 Quality control is important in any organization that wants to remain competitive in order to be profitable, whether it is a manufacturing or a service industry. Process control procedures are tools that can be applied to any process in an effort to monitor the quality of a product or service and improve that quality. It is important to remember that “process control is concerned with monitoring quality while the product or service is being produced (Chase, Jacobs, & Aquilano 2006 pg 354).” The reason for this is that the information of whether or not a product or service is meeting quality specifications is much more useful if it is received during the process when something can be done about it. Finding out this information after delivering an inferior product or service is not nearly as helpful to an organization. Statistical process control (SPC) is a method of quality control that uses “random samples of output from a process to determine whether the process is producing (items) within a selected range (Chase, Jacobs & Aquilano 2006).” Unlike other methods of quality control which are usually based on variables of various measurements, SPC is based upon “yes or no” specifications. Another way said SPC basically tests whether a product or service is “good or bad.” While there are still measurements involved they are much simpler equations. Let’s take a veritable simple process and use the SPC method to evaluate where quality control and detect where in the process improvements can be made to ensure the output is within a selected range. Before this method can be applied however, the variations in the output must be identified and control limits must be set. The authors of Operations management for a competitive advantage tell us that “it is generally accepted that as variation is reduced, quality is improved (Chase, Jacobs, & Aquilano 2006 p321).” This being said it is also generally accepted that it is impossible not to have some variations in a process thus it is easier to identify them and determine if and how they can be controlled. We learn from the same text quoted above that variation in process fall into one of two categories Common or Assignable. Common variation is basically inherent to the product or service and therefore is considered “normal.” Assignable variation is usually clearly identified in a process and in many cases can be controlled or at least managed. In examining my morning process one of the common variations is traffic that I can not control and can only forecast to a certain degree. For example I know if it is raining or dark outside traffic will undoubtedly be moving at a slower pace and where there are only two lanes in my direction it will be even slower. If there is road work going on or an accident this will delay traffic even more. During the summer months my town experiences rolling black outs to conserve energy, even though I know this variation can occur, I do not really know when so it is for the most part uncontrollable. I will discuss more common variations later. Assignable variations could be forgetting to set the alarm clock, hitting the snooze button too many times, time spent on tasks that could be done the night before to use less time on the process, and others to be discussed further in this paper. Over the past five weeks I have collected data in regards to my morning routine or process. The data collected included the amount of minutes it takes to perform each of the eight tasks I must perform in order to get to work. The eight tasks were timed individually and cumulative. Once this information was gathered in a spreadsheet format the parts of the process with the most variation could easily be seen and plotted on control charts. The control charts includes specifications such as upper and lower control limits, a mean or average and the standard deviation from that mean. Another measurement used in statistical calculations is the confidence interval which is expressed as a percentage and is usually between 90-99%. This percentage is basically a measurement of how accurate the data collected represents population parameters or how reliable the data is. Using the confidence interval allows me to take into account that some of the data collected may fall outside of parameters set but I can still use the data to estimate the statistical measurements for quality control. My data is still useful even taking into account possible errors since I have so many data points.