Things to Think About and Do in 2011: Data

Here is this week's installment of ReliabilityWeb.com's e-book, Things to Think About and Do in 2011. Can't wait for next week's topic? Thumb through the entire e-book for free at ReliabilityWeb.com, Happy Reading!

 

Data

How much valuable information are you losing because of inconsistencies in the information that was entered into your data management system? We have all heard the saying, “garbage in, garbage out,” but what does it actually mean?

                FALK 1040-EZ

                FALK 1040EZ

                FALK  1040EZ

To the human eye, the above examples shows three gearboxes that are the same make and model, but to a computer that is processing data for the generation of failure models, maintenance expenses tracking, and spare parts stock lists, there are three different make and models of gearboxes. 

The differences in the descriptions are subtle—an extra space or a dash—but these extra characters create different gearbox units. The same applies to a mere typographical error.

Now, imagine that you have 500 gearboxes in your plant. All of them are a Falk 1040EZ. And ALL of them have slightly different entry for their make, model or description. How much time is wasted scrubbing the data before a failure model can be created? Running a simple failure model might take an hour or two, but the data preparation can add weeks to that time line. Are you tracking maintenance costs collectively by make and model? You could be seeing costs spread over multiple categories that should be grouped into one, and this gives an artificially low result for problem component types. This can increase the space required for parts storage through unnecessary duplication of parts, based on the components categories. 

Data entry is often overlooked as a menial task. However, good data entry is the backbone of a reliable maintenance program. To increase the consistency of data entry, knowledge and understanding of the information being entered is needed, as well as the know-how to research and make corrections when discrepancies are found.

—Stacy Heston