The ultimate goal of lean is Single Piece Flow, which simply means working on one product at a time from start to finish before beginning to work on the next product. If each process step in a value stream achieves Single Piece Flow, then the entire value stream achieves Single Piece Flow.
The reality is that pure Single Piece Flow is difficult to do because of variability that exists in processes. There are 3 types of variability that exists in all value streams: the variability of demand, the variability of cycle times and the variability of waste. Lean practices either eliminate the variability or manage the variability.
Establishing Flow & Pull is the first step to achieving productivity gains. Just as the CFO oversees internal control systems to ensure that the financial numbers are valid & relevant, the Lean CFO must oversee the lean data collection system & measurement system to ensure these numbers are also valid & relevant. The numbers used to identify waste & measure improvement will drive productivity gains. Poor numbers will not achieve the necessary productivity gains.
Measuring the Variability of Waste
The variability of waste must be eliminated to improve flow. Continuous Improvement systems are used to identify & eliminate the variability of waste. The primary root cause of waste in value streams is how the company designed the process to work. Companies create processes that are wasteful and can eliminate the waste if a measurement system is designed to measure waste & flow.
Before waste can be eliminated, it first must be measured. The waste in any value stream is measured during the current state mapping process. This is a critical step in continuous improvement because if the waste is not measured properly, kaizen events will not be effective in removing the waste.
The Lean CFO does not need to become an expert in value stream mapping, but does need to ensure that the data collection process used to create the current state map is objective and valid.
The recommended method to measure waste in the current state is direct observation of the process. We’re not talking about 6 month time studies. We are talking about people observing the process as it is working and noting the time spent on all wasteful activities. For statistically valid data, about 30 observations are needed. The types and amount of waste observed will guide the continuous improvement efforts.
The end result of observing the value stream should be data such as, but not limited to:
- Scrap & rework rates
- Wait time of products between process steps
- Downtime of people or machines
- Amount of time spent getting parts, tools, and any other activity other than adding value to the product
- Changeover time of machines