Analyzing Alerts in SAS Field Quality Analytics
This post is the next in a series of posts about SAS Field Quality Analytics. In my previous posts I talked about the different types of alerts groups within SAS Field Quality Analytics, and how to create a data selection. The purpose of this post is to show how to further analyze the data underlying an alert.
Once you have received an alert it can be helpful to dig deeper and identify what the core problem is. For example, in the alert below we know that it is for models with model code Gemini, in the United States and Primary Labor Code D-007:
Select any image to see a larger version.
Mobile users: To view the images, select the "Full" version at the bottom of the page.
Based on this information alone we don’t have enough information to make a change that can prevent future warranty issues. There are analyses that we can do to dig deeper into the alert that is shown above. For example, we know that this alert took place in the United States, one analysis could be a Geographic analysis to see if there are any trends by state. To do this I ensure that the desired alert is selected and then select the Analyze alert button:
Next, select a project to save the analysis to, and give the data selection and the analysis a name and select Save:
An Emerging Issues analysis using the new data selection is created and saved in the folder that was selected:
There are two ways that you can further analyze this alert. You can create a new analysis based on the new data selection that was created, or you can select Analyze further within the alert analysis that was created. This post will focus on the first approach and future posts will explore the Analyze further option in more detail. To create a new analysis based on the data selection that was created select the new data selection and then select the new button:
Next, move over any of the analyses that you want to create. In this example I will move over the geographic analysis, then select save:
The settings for the geographic analysis appear, I will leave all settings as their default apart from changing the reporting variable to be repair dealer state. Future posts will dive more into the specific settings of analyses. Additionally, these are covered in the SAS® Field Quality Analytics on SAS® Viya® course which is part of the SAS Internet of Things Learning Subscription. When the analysis settings are set to the desired properties, select the submit button:
The analysis is run and you can view the results:
The results indicate that the highest claim costs are in South Dakota, North Carolina, and Tennessee. If desired, I could dive into this problem in even more detail. Some examples could be conducting a pareto analysis on specific parts to see if a part is the problem. I could also see if there was an issue where the products were produced. In any case this is all possible by starting with an alert and drilling in deeper until finding the root cause. Future posts will go into more specifics on analysis types. For additional information, see the SAS® Field Quality Analytics on SAS® Viya® course which is part of the SAS Internet of Things Learning Subscription.
Find more articles from SAS Global Enablement and Learning here.
