Using lean methods to find process improvements
Much of what I know about lean and process improvements comes from two of my favorite books on the topic: “The Goal” by Eliyahu M. Goldratt and “The Toyota Way” by Jeffrey K. Liker. Both helped me visualize how to use constraint theory and lean thinking to find the right problems to solve and find the right steps to fix them, rather than just randomly improving things as I go. Below I will explain my basic approach to process improvement and explain the reasoning on each step.
My main goal with any improvement is simple: Make it better. More specifically though, I aim to gradually eliminate waste until I eventually reach a point where there is none. Ideally, work would be completed seamlessly from step to step, in an obvious direction, with little effort (one-piece flow). This usually takes more than one iteration to achieve.
 Step one: Value–stream mapping is a very important starting point before brainstorming, feedback or even root–cause analysis. In my opinion, it is also the most efficient because it is often reliable factual information that you already have or can easily access. At this point, a few facts can disqualify many theories.
From a process perspective, your department or business is made up of tasks, relationships and people that perform and oversee those things. Understanding each step, each relationship and each person in the process is your roadmap for continuous improvement. It will help you to visualize problems in your overall workflow.
How to value-stream map
- Make a sticky note with the name of every task and transfer that happens in your business.
- Place them in sequential order from left to right on a whiteboard.
- Draw a line horizontally on the board under the sticky notes.
- For each sticky note, if the task doesn’t change the finished product, put it below the line.
- Once complete, the notes above the line are tasks that create value, and the tasks below the line do not create value (even if necessary).
 Step two: Pay close attention to where work “pools” or backs up, and the rate of output. Measure the overall rate of speed to completion as well as the rate of completion for each task. Once you identify an area where work is slowing down, you can use a root–cause analysis to see if there is an underlying source of the issue. One option at this point is to create a second value–stream map of your ideal workflow and then compare the two to see if there are any obvious improvements to make.
As you test different things in your value–stream map, check to see if improving a point in the process will improve the overall output. Not every well-intentioned improvement will actually save time or money in the end.
How to do a root-cause analysis
- Start with an assumption of a problem.
- Ask what causes the problem.
- Continue to look for the underlying source of each new problem.
- Once you cannot find any more causes, you have found your root cause.
 Step three: If at this point the problem isn’t clear or you are unsure of your solution, it is a good time to try out an A3. I like using A3s for more complex solutions because they enforce an organized way of thinking about problems. During this process, it is helpful to interview people at the source of the problem and even try out the processes firsthand before coming up with a solution. Once a solution is implemented, follow up and adjust regularly.
If you don’t find the right solution, don’t push the wrong one. Go back to the value stream map and think harder.
Simpler solutions seem to be more effective than complex ones.
Creating a basic A3
On an 11-inch by 17-inch sheet of paper:
Draw lines to make six equal spaces (Landscape, two columns, three rows).
Fill out the following items, one in each space:
- Background: What is the history of the problem?
- Problem statement: State the problem in the clearest and simplest possible way.
- Goal or future state: What is the ideal resolution to the problem?
- Analysis: This can be your root–cause analysis or, even better, hard data that can be measured to prove the problem statement.
- Proposal: What can be done to move toward the goal or future state?
- Implementation: Who is your team, what do they do, how do they do it and how do we know we’re winning?
 Step four: Lastly, put measures in place to make sure that your plan will work when it is needed. In software development, it is called error trapping, because we try to account for anything and everything that could go wrong and provide either a solution or an instruction. This can be something as small and simple as a quality controls checklist or a good writeup that ensures quality and consistency in your new process.
My advice: Do what you can, with what you have, right where you are. Be involved in improving the things around you. If you are just starting out, set the bar relatively low and just make something better than it is now. If you are ready to take a more systematic approach, the above approach might be a good place to start. If you don’t know what to improve or where to start, then improve yourself. Take classes, read articles and ask questions.