Change language

Problem solving through root cause analysis

18/03/2020 | | Quality

In the last blog post, I discussed Key Performance Indicators (KPI’s) for optimal quality management. But it’s not enough just to calculate the KPI’s and the trends, it’s also important to understand the root cause of certain problems and resolve these causes.

To do this, data must be analysed. Often companies collect data, but don’t do anything with it. This results in many missed opportunities.

Problem statement

In March 2017, I spent one week with 33 other entrepreneurs on a business training in Dubai. Here, a business owner trading in glass doors, tablets, tables, walls, etc. came up to me and said “I want to grow my business, but before I can do this, I need to control my waste. If I don’t and my business doubles, my waste will also double and the cost will be huge. So I need to better control this, but I don’t know how. Can you help me?”


One month later, I had a meeting with him and the woman responsible for the operational side of the business.

While discussing the issue with them, they came up with only one word for glass they need to throw away: waste. To better understand and help them, we had to be more precise and dig into the waste file and pictures of glass plates that had to be reworked or be thrown away.

Research and root cause analysis

Just by looking at the pictures, we were able to establish different categories: from small chips missing on the side of a glass plate to whole corners missing and incorrect cutting (e.g. wrong size).

The pile with the chipped products was growing so rapidly that it was clear that this would be the focus of the exercise of that day. This is the first step in problem solving: break down the bigger issue into smaller ones, define what you will focus on and think about what is causing you the most pain, whether it is in time, money or frustration.

The chipped products needed to be reworked: the side where the chip was had to be removed and the glass removed could not be used anymore and was therefore defined as waste. Depending on where the chip originated, it was possible a whole finished product must be thrown away. Products were custom-made and if the chip occured at the end of the process, the whole product was considered waste.

That’s where the second refining took place: the pile with chipped products was broken down again to where in the process the chip is first established. Luckily, the pile with chipped end products was very small and as we got more to the start of the process, we saw the pile growing.

We didn’t jump to conclusions by going directly to the biggest pile. We wanted to reduce the cost of waste, which meant we had to take financials into account: what was the cost of reworking, not selling or reproducing a product? How often did what deficiency occur? Although the chipped final products were a huge cost to the company, it didn’t turn out to be the biggest one.

Fishbone (Ishikawa) diagram

We decided to start with the biggest pile after all: at the start of the process. Then I started drawing a fishbone or Ishikawa diagram. On the right side, we put the chipped glass plates when they enter the production process. Then we added the six possible cause categories: machine, method, man, material, measurement and mother nature.

Next, we delved into the data again and identified the 2 main categories: method and man within the unloading process of the glass plates coming from the supplier. We decided to start with the method of unloading “because we have good, capable people working here and it is very unlikely that that would be the main root cause”.

Then, we added the next fishbone to the diagram, with glass being chipped during unloading as the main problem. The determining variation seemed to be the man category. We further analysed this category according to the ‘5 why methodology’: why has one of the two operators have more defects than the other? They have slightly different ways of unloading, but why? Turns out, there was no instructions available on how to unload.

Process flow

We drew out the process flow of the best performing operator and even slightly improved the way of working. Next, we recorded a video of the final unloading process and used it as training material for the operators. This way, they could look back on how to unload the glass plates from the truck as often as needed.


One year later, the amount of waste was reduced by 76%.

Albert Einstein was once asked what he would do if he had one hour to solve a life-threatening problem. To this, he answered that he would:

  • take 50 minutes to define the problem and to determine its causes;
  • need 9 minutes to find a solution;
  • need 1 minute to implement the solution.

This is an example of a structured and effective way of approaching a problem, a mindset that might help you when tackling certain problems.

Could you use some assistance in your problem solving process? Don’t hesitate to contact QuontinuIm.

Stay informed