Nice work this week on your Cause & Effect (CE) diagrams (also known as "fishbone" diagrams) and Pareto charts.
The three brave souls -- Ashlyn, Karson & Shaina -- who took to MS Word to draw their diagrams did nice work not only with the drawing tools, but brainstorming potential causes and contributors related to their identified problems. These charts are very useful in starting the root cause analysis conversation; it sets the table with all the possible causes and contributors organized in such a way that you can begin to address each in turn.
The Pareto chart provides a tool that helps prioritize the causes and contributors identified in the CE diagram. Thanks to all who used the data in question 11 to create their charts. There was unanimity on which set of errors should be addressed first (shipping delays). Andrew, James & Quin argue that both shipping delays and shipping errors should be dealt with, as these two errors combined account for 80% of the problems. This is the value of the Pareto chart. It quickly stratifies a set of problems identifying which small percentage of your total list of problems are causing most of the pain. Using this tool, then, we found that 2 of problems account for 80% of the error. As a manager, we want to focus first on addressing those two issues; that's where we get the most bang for our buck.
All of our Pareto people worked to create their own charts using Excel. Congratulations! You can never have enough Excel practice. One of the features of a Pareto chart that makes life much easier is the cumulative total curve. I've attached a link above to a Pareto chart template from the American Society for Quality (ASQ), a leading quality assurance professional society. This template is easy to use and will provide both a frequency histogram and a cumulative total curve in your chart.
Just for fun, I've also attached a link to the ASQ fishbone diagram template. Another easy-to-use template for creating CE diagrams.
Thanks for your experimentation with the factor weighting method for determining a potential Save-A-Lot store location. My general feedback is contained in the attached video.
See Ashlyn, Ethan and Quin's posts for some good discussion about the factors to be considered and their weightings. Also note my points in the video regarding the connection between factors selected, their weightings, and operations strategy/CBP).
Here's one of the articles I referred to in the video. It provides some good background on Save-A-Lot's business strategy.
Here's an excellent radio story on Save-A-Lot from the customer perspective. Its worth a listen.
Today's online case presented us with a data set comprised of external and internal data, asking us to find the relationship (if any) between these data, and then develop recommendations based on the relationship found.
Many of you identified the importance of the external data in this data set: customer satisfaction. This is a crucial external measure; if we don't have satisfied customers, we don't have a business. The key in this case was to compare the customer satisfaction data with both the new applicant processing time and the plastic production turnaround time, which most of you did. Our goal is to understand the relationship between customer satisfaction and these internal metrics. Is there a relationship? In other words, does the speed at which we process new applicants and/or produce a new credit card have an impact of customer satisfaction? And if so, how much?
Using Excel to calculate the correlation coefficient between customer satisfaction and new application processing time, you found r = -0.41. The correlation between customer satisfaction and plastic production turnaround time was -0.21. Both of these suggests a negative relationship: customer satisfaction goes down as both of these times go up. But the relative strengths of these relationships is relatively weak. You visually confirmed this in your scattergrams.
Ashlyn, Andrew, James and Korey agree that better information is needed before the business can make any decisions about how to improve customer service. Better customer satisfaction surveys could be a part of this. Ethan gets us to the heart of the matter: our correlation coefficients suggest that while new application processing time and plastic production turnaround time are somewhat related to customer satisfaction, there are other drivers of customer satisfaction not captured in these metrics. To James' point, the departments need to work more effectively together to identify other potential variables (causes) and identify relevant metrics.
Another statistic that can help us understand the strength of the relationship between variables is the coefficient of determination, or r squared. The coefficient of determination tells us how much of the variance in a dependent variable is due to the variance in an independent variable, expressed as a percentage. See the excellent 3-minute video below for more background and discussion.
Take a minute to calculate the coefficients of determination for our case r-values. How much of the variance in customer satisfaction is accounted for by new applicant processing time and plastic production turn around time? What other internal company operations factors might be drivers of customer satisfaction?
