There are a lot of different process improvement programs on the market today and everyone has their favorite. I am sure you recognize a few of these programs. Have you ever wondered what the difference is between each of these? There are many consultants in the market that are advocating the use of one of these improvement methods to solve all of your organizational problems. But will all of these work in any situation? Or, are there specific situations that are better suited for a particular approach?
Let’s compare and contrast the first three improvement programs: Six Sigma, Lean Thinking, and the Theory of Constraints. We will explore the advantages and disadvantages of each one and the best way to use each method to realize improvement.
In statistics, a ‘sigma’ (s) refers to the standard deviation from the mean of a population. Standard deviation describes the likelihood of your next data point deviating from the mean of the whole data set. The sixth sigma refers to the likelihood that only 3.4 out of every 1 million data points will appear outside the sixth standard deviation. That translates into less than 4 errors per million transactions.
Six sigma is one of the process improvement programs that is all about variance reduction. What do we mean by variance? We are talking about the amount of control you have over your processes. Another way to look at it is how good you are at predicting or forecasting the future outcomes of a given process.
For example, if we are talking about an accounts receivable process with a goal (effectiveness criteria) of collecting all receivables within 30 days, then we would measure the time variance of each individual receivable collected against the average days outstanding (let’s say it’s 35 days) for the whole population of receivables collected. If your individual collections range from 25-90 days then we would say your process variance is high, which translates into a low sigma. The difference between the process average (35) and the goal (30) would reflect your process’ capability or your process’ ability to meet the target.
As we have seen in business metrics for results, variance is a symptom of waste. So, the higher the sigma, the greater the control you have over your process, which means greater forecasting accuracy and less error. Processes that exhibit a lot of variance mean they have a lot of waste.
The name six sigma tells us a lot about this one of the process improvement programs. You can see that the six sigma approach involves a little advanced math, some statistical tools to understand the various fluctuations of a process, and a lot of data to run the calculations.
Six sigma is very problem focused. It uses a scientific approach called DMAIC to analyze a specific problem. DMAIC stands for Define, Measure, Analyze, Improve and Control, which is also known as the learning loop or PDCA.
Process improvement programs like Six sigma are ideal for scientific or numbers based organizations including: high technology manufacturing (semiconductors), high transaction businesses (check or postal processing), fast paced research and development (electronics, pharmaceuticals), or environments where errors are extremely expensive (surgical operations, space exploration, aircraft takeoff and landings).
Motorola and General Electric are two high technology companies that have helped to popularize the six sigma concepts. Motorola developed the methods back in the 1980’s to improve manufacturing yields. However this method is not ideal for low transaction situations (most small businesses) or mature industries with simple manufacturing processes (like baking).
Six sigma requires a rigorous application of DMAIC and statistics that may not be suited to many organizations. It is also highly problem focused, which works well for big problems but can lead to sub-optimization when applied across an entire organization. That is one department or process being optimized at the expense of another.
A lot of process improvement programs are available on the market today. If you are in a high technology, high transaction, or expensive error environment then six sigma will work real well for your organization. If not, then perhaps Lean Thinking (waste reduction) or the Theory of Constraints (throughput improvement) would work better.