• Home
  • Publications
  • Blog
  • A Definite Problem: Without a good problem definition, you may not be solving the right problem.

A Definite Problem: Without a good problem definition, you may not be solving the right problem.

21 Feb 2024 10:31 AM | Ali Kucukozyigit (Administrator)

by Phillip Power MEM, CPEM

A Definite Problem

Firms seem to put a lot of resources into problem solving methodologies without emphasizing the importance of a well defined problem. Without a good problem definition, you can have a set of problem solving skills even the most adept consultants would envy, but you may not be solving the right problem.

Personal Experience

To illustrate my point, I will use a case study from my own professional experience. While working for Large Company XYZ, we had an issue. Our product was not able to pass a specification, let’s say it was the concentration of a chemical used in medical applications. The concentration would consistently test over or under the given limits. The process to make this chemical was old (no!) and the equipment was validated over 20 years ago (was it still in the validated state, you bet!). The process controls were not reliable and the time it would take to complete each processing step changed batch to batch.

I called a kaizen and amassed the subject matter experts of the plant. Immediately people began explaining what was causing the inconsistent concentration in the chemical. The operators assured the group that it was the raw materials, which had the supplier quality team on defense. The engineers were fairly confident it was the jacketed reactor, probably scaling or a leak. This had maintenance on defense. Maintenance assured the team that they perform their PMs dutifully and have the records to prove it. Action items are put on the board to look into the raw materials and perform a trend analysis on the certificate of analysis data and to inspect the jacketed reactor at the next product changeover. Oh, and by the way, that won’t be for a couple of weeks, and even then, do we really want to run the risk of upsetting the process more by taking things apart and putting them back together? We had better start a bidding process on a new reactor in the meantime and see what the lead time is on getting a new one on site; if the reactor is the problem, we can’t afford to lose time and put more product at risk. Ok team, break!

What do you see here?

Already you can see how a problem-solving exercise can run away; I’m sure you’ve all experienced something similar. It can be easy to let it happen when you sink into a routine and you’re familiar with your co-workers. The strict structure of a kaizen or 8D problem solving are fine but why waste the time when we already know what the problem is and are halfway to a solution? In this case, had the team put effort into properly defining the problem, all of this work could have been avoided. What was the problem? Was it that the concentration was inconsistent or was it that the product was failing to meet specifications? Maybe those are the same thing, maybe they aren’t. In this case, for the application of the chemical, the range of concentration that the product was experiencing did not pose a risk to the patient. In fact, the chemical was safe up to 20x the concentrations we were experiencing. So where did the specifications come from? When the product’s initial validation protocol was pulled from the dusty catacombs of Record Archives, we discovered that the specifications were set arbitrarily using data from the 3 process qualification runs.


So now what does the solution look like? Work with quality on updating the specifications using a fresh risk assessment that documents
the safety and efficacy of the chemical at these concentrations.
Work through the change control process to implement the change and
you’re done. Now, if you’re the fellow performing the risk assessment or managing the change control, you might not think this is the best solution. But from a company perspective, you aren’t using resources trending raw material data, you aren’t taking up maintenance’s time pulling apart the reactor, you aren’t wasting purchasing’s time with bidding, and the solution requires no capital, no validation, and no additional down time. Sounds like a win to me.

About the Author

Phillip Power MEM, CPEM is a Pharmaceutical Technical Specialist for Zoetis and a Lecturer for the University of Nebraska- Lincoln's MEM program. At Zoetis, Phillip manages investigations and CAPAs, operational improvement projects, and risk assessments to ensure the market has access to the highest quality medicines for companion animals and livestock. He earned his B.Sc. in Chemical Engineering and his M.E. in Engineering Management from the University of Nebraska- Lincoln.

Phillip lives in Lincoln, Nebraska, with his wife and two sons.

BLOG DISCLAIMER: The opinions, views and content presented in this online blog solely represent the  original creator and has no association with ASEM activities, products or services. The content made available for the community for educational and informational purposes only. All blog post are created voluntarily and is not sold- but can be used, shared and distributed free of charge. The blogs are not academic pieces, and does not go through a peer-review process or fact check. All errors belong to the creators.

CLICK LOGO BELOW to find out more about ASEM's Promoted Organizations / Sponsors:


Dr. Bill Daughton

Professor Emeritus

Missouri S&T

Powered by Wild Apricot Membership Software