Make Everything You Do an Experiment

Lean operating systems unlock the value in business through continuous improvement that comes from continuously learning. The flood of learning driving improvement comes from an unyielding commitment to the scientific method. Two beautiful benefits of the scientific method are: You can’t lose, you always learn, and just forming a hypothesis can sometimes bring about needed change.

You Can’t Lose, You Always Learn

Thomas Edison is rumored to have known 10,000 ways to not make a light bulb. Fortunately for us, it led him to the one way to make a light bulb. That is the beauty and purpose of the scientific method. We may not get the desired result (a light bulb), but we always gain the next level of understanding by following the process.

Say there’s a hypothesis that the production process can produce 240 widgets every hour at the rate of four every minute, one every 15 seconds. If we consider each time we process a widget an experiment, we begin to learn. 

The first eight widgets process on time and exactly two minutes go by. On the ninth widget, the piece jams in the machine and is delayed 30 seconds. Suddenly our hypothesis fails. Our process cannot produce 240 widgets every hour. However, the learning organization sees this as a great opportunity, not a loss. The learning organization now knows one way the process can fail, and it begins the process of understanding the cause and solving it.

The distinguishing feature of the scientific method is not that it always gets the answer right, but that it fails forward by learning from its mistakes.

Forming a Hypothesis Brings About Change

Organizations fail when they do not have a standard in place. They fail because they lose the opportunity to learn, but they also fail because they have not set the expectation for everyone to follow. 

Personal protective equipment is a good example. There’s a hypothesis that safety glasses will prevent injuries in people working in areas with the potential for debris to get in their eyes. In order to run the experiment, the expectation has to be established that everyone must wear safety glasses. The expectation needs to be clear, binary, and as visual as possible. This means there can’t just be a sign stating that everyone has to wear safety glasses, but an audit by supervision and leadership ensuring that everyone is wearing their safety glasses.

Now take the same principle and apply it to the widgets. If we make the expectation clear and binary that our hypothesis processing 240 widgets every hour, then it is likely the people responsible for the process will work toward that goal. If the equipment is set at a rate below the expectation, the operator will increase it to the expectation. Change comes about by merely setting the standard. 

Don’t Forget, Science Requires Analysis

Classroom simulations that teach lean often miss one of the key lessons – the reflection period. A workshop may show how inventory is reduced as the participants reorganize their simulated manufacturing process to assemble Legos as a continuous flow rather then a traditional batch environment. Usually the workshop will involve several runs with time to improve the process between each run. The participants learn the value of continuous flow over traditional batch manufacturing, but they forget what got them to that improved environment was a reflection period after each run. Imagine if that were applied to business processes. Think of the value unlocked by a reflection of our hypothesis and learning after each experiment. 

Encourage your organization to make everything an experiment. Businesses that do continue to improve because they continue to learn.

What is your expectation for the day? It might be interesting to see what you can learn.

Read more at www.csi-opex.com.