6S DMAIC: Measure

After we define the problem, we need evidence to support that it wasn’t just a freak accident. If odds are that it’ll never happen again, we won’t need to anything, right?

In daily life, we are conditioned by habit; in diet, sleep, work, stress, and so on. When something unusual happens, it raises alarms that something might be wrong.

In manufacturing engineering, data can be tracked using Statistical Process Control (SPC). Operators and equipment records process data of test, measurement, and yield. When the process is automated, a large database of “normal” data can be determined. Once a baseline is established, there are still allowable tolerances that are expected from noise and variation.

Image result for yield chart spc
An example SPC chart. When data exceeds the UCL (3 sigma above mean), there needs to be action!

As expected from Six Sigma: any data point that crosses beyond the Lower Control Limit (LCL) or Upper Control Limit (UCL) should be measured.

With these deviations compared against a long history of data, we can begin investigating what variables occurred at that specific time point

  • When did the problem state began?
  • What conditions were different from the previous states?
  • How often does the problem recur?
  • How large is the deviation?

With a standardized measurement process in place, it allows for comparison between problem and ideal statements.

But how reliable is the data that is collected? Is the full picture being captured? Analysis must be taken to ensure that data is being captured properly, by both method and equipment.

This is why there can be entire companies dedicated to calibrating measurement instruments. These companies rely on their instruments to provide reliable measurements to make data-driven decisions.

Many apps and services offer automated data measurement for our daily metrics, such as money spent, distance walked, or time slept. It is quite interesting to see these daily habits laid out like this, and helps provide more awareness on how we can improve.

Comments

Popular posts from this blog

Asia Trip Overview

Asia Trip Day 17: Chill and Eat (1/4/18)