At myDIALS, we tend to think of ourselves of providing business intelligence and analytic capabilities through a very intuitive, interactive user experience. Lately as I’ve been talking to various people in the industry, I’ve heard our capabilities described more frequently as “visual analytics”. I’ve been thinking more about this as it does describe what we do, particularly as it relates to the way we utilize analytics to help people draw more information from the data we present. We have enhanced this significantly with our upcoming myDIALS 3.4 release, and here are some examples of the sort of visual analytic workflow we now make possible.
Let’s start with a fairly standard trend dial showing overall product production:
Then by using the information menu, we can begin the process of extracting more useful information, perhaps starting with basic analysis such as min, max, averages etc as shown below:
Or we can begin to apply more detailed analysis such as trend lines, perhaps with forecast projections:
To give the following result to show how production might continue to trend during the month. As shown below, the average daily production is projecting slightly upwards but not in a dramatic way.
We might want to look at a variance analysis against the budget or against the prior period or prior comparable period as shown below:
This makes it very clear when production is better or worse on a daily basis against a target, prior period etc. In this case we are under performing against our budget on most days although we have had a couple of good days. This might lead us to investigate further if daily production is relatively consistent from one day to the next and to do so, we might want to use more advanced statistical analysis such as a histogram or a control chart as shown in the sequence below, starting with a histogram:
And here are the results of the Control Chart:
We can see from the control chart that we are operating in a fairly tight range around the mean and certainly well within the Upper and Lower control limits. This tells us the production process is consistent from one day to the next, but we are under performing against budget so we should look a little deeper into whether there is some way to isolate where the issue lies. We might look again at the variance against budget, but expand along a relative dimension such as location:
When we do this, we can immediately see our problem is in Europe as shown below:
Now the story doesn’t end here, as we would once again look at trends, projections, control charts, break Europe production further by the various locations within Europe or perhaps by individual product lines etc. However, this process of applying visual analytics is quick, easy and follows a natural path of question / answer that we are all very familiar with. In a very short period of time a operations or business person is using analytic capabilities themselves to answer their questions, analyze a situation and make decisions on how to improve production to meet or exceed targets – all without requiring the assistance of IT, statisticians or business analysts.
{ 0 comments… add one now }