Once you encounter abnormal variation, or once you detect a defect, you have to start investigating the root causes. But how do you find the root cause? In this session, we'll introduce some tools for root cause problem solving. Specifically, we'll introduce the concept of an Ishikawa diagram, as well as a Pareto chart. Also with these tools beyond to a toolbox known as Kaizen. The idea of the Toyota production system is that quality and process-improvement tasks don't have to be carried out by the management, but are carried out by the frontline employees. So our jobs is to put together a simple set of toolbox, so that we can equip with the frontline workers, to go about improving their work. Ishikawa diagrams are a structured brainstorming technique that help employees think about the root cause for a problem. Lets get back to our chocolate example. Maybe we want to figure out why the weight, that we have measured for a particular set of bags was too heavy. Let's think about the potential root causes of this. One set of root causes might relate to there being too many pieces of chocolate in the bag. Another root cause might relate to each of the pieces being too heavy. Then in turn, it might have been the bag that was too heavy, or maybe the scale was broken and the batch was also the right weight. And so we're struggling with a measurement problem. Within each of these root causes, there are potential sub-root causes. A famous technique in the Toyota production system is known as the 5 Why's. When you're root cause problem solving, you keep on asking, why, why, why? The bag was too heavy, why? Well, there were too many pieces in there. Why? Well, the machine that was putting pieces in was poorly calibrated. Why? And so on, until you get to the root cause. Now you'll notice there's no rocket science involved in these diagrams. It's just a useful brain summing technique. If you draw nicer than I do, you notice that this little diagram looks a little bit like a fish, with kind of the head of the fish up here and these being the fish bones. For this reason, this is also called a fish-bone diagram. Once you have mapped all the potential root causes, you go into the field and start collecting data about the actual defects. You'll then map out the defect occurrences, so the number of defects for each of the types here in the issue covered diagram for each of the types. And that leads to a graph that looks like this. So you sort them by number of occurrency with same too many pieces in the lead. And then all the way, maybe having here just two problems. The weight of the bag was the root cause. And then you can plot the cumulative number of problem occurrences in a curve that looks like this. Typically you find an 80-20 logic. 80% of the defects could be explained by 20% of the root causes. This is known as the Pareto principle. Allow me to use the examples of the Ishikawa diagram and the Pareto chart to make some philosophical reflections about operations management. Most of what we do with economics and most of what we have done in this course is create models of the underlying operations. Now, models are always imperfect approximation. They're hopefully good approximation, but they're still imperfect approximation of the underlying reality in the operations. Now, as academics, consultants, and managers, we are always at risk that the ideas that we come up here are just based on our thoughts and the models that we've generated in our brain. But they don't work in reality. What you've noticed is that problem solving in Kaizen, is that there's a strong emphasis on iteration between the world of reality and the world of thought. The trigger for the Improvement projects is happening in the field when there are defects. Because of Jidoka we stopped the process and start thinking about it. This reflection is happening in the world of thought. The Ishikowa diagram forces us to think about many root causes. And that probably will start triggering some improvement ideas. But we don't just jump to the idea that we like, instead we go back out in the field and collect data. That's what the Pareto chart does. It's an empirical support for which root cause we should go after first. Once we've identified the root cause that accounts was the majority of the defects, we'll take that back into the world of thought and start exploring some alternative solutions. Again, we're not going to run out and standardize these solutions too quickly. We'll take it back on the field and run some small scale experiments. Only once we have done so, can we standardize the process and say that's really a better process. So, this isolation between the world of thought and reality is the key signature of good problem solving. Rapid iteration is at the heart of creativity. Notice that both the Six Sigma production system, as well as the Toyota production system emphasize this quick iteration between the world of thought and reality.
