Establish Reproducibility

Establishing Reproducibility

OK, now you know how to “Define the Problem clearly.” Let us discuss how to Establish Experimental Reproducibility.

This is the simplest step in our DEPLOY process.

In this really short course, I will focus only on What to do and how to do it. I will cover the Why we should do this later, it will take a bit more time. 

Establishing reproducibility early on is simple and easy to do.

To estimate the reproducibility here is what needs to be done:

  1. Run an experiment. Ideally in the middle of the experimental region.
  2. For each experiment get samples, take measurements, and report.
  3. Do not take any shortcuts.
  4. Repeat the same experiment 4 to 8 times. 
  5. D not collect samples of all experiments and wait for analysis later. Do not run all measurements in one day!

Compute average, standard deviation and range. Report the experimental procedure, sampling and measurement method and results. That is reproducibility. It is a realistic estimate of reproducibility.  Here is an example. This output can be generated automatically for you, you do need to compute anything. Just run the lab experiments, and compile data. Here is an example:

Reproducibility Data, Cheesecake Project

Here is an example of the Reproducibility Report for our Cheesecake Data

Good reproducibility will help you develop confidence in your apparatus, test method, lab setup, procedures.

If reproducibility is bad and unknown, it could put you in a never never land. You have no clue what is going on and why. and if it persists too long, it might even hurt your reputation.

That concludes the second step of the DEPLOY process: Establish experimental reproducibility.

About Author

Mukul Mehta

Mukul Mehta has over 40 years of proven industrial experience in chemical , polymer, and plastics industry. Worked as a Sr. Manager, Statistics and Computer Aided Research for BF Goodrich Chemical, a Fortune 500 company, and then as a software entrepreneur, promoted "quantitative, predictive modeling in one minute or less as a mantra for R&D and New Product Development." Many multi-million dollar successes for dozens of Corporate R&D clients in chemical and pharma industry. Trained over 750 R&D chemists, engineers and managers to Speedup New Product Development through statistical design of experiments.

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}