Metrics are a fairly common topic of discussion on the various lean manufacturing forums. One theme that comes up fairly frequently is how to determine “what counts” in this-or-that measurement.
For example, a recent post asked about measuring lead times. The way they were measuring lead time only counted the time from when the order was actually started on the shop floor. But it didn’t include the latent time between the time the customer placed the order and when processing it actually started.
A rule of thumb I like to apply in these cases is “What is the customer’s experience?”
In this case, the customer starts waiting on the order the moment he informs the company he wants something. The customer really doesn’t care whether the order is waiting in order-entry, waiting for the computer to run its batch process, or whether it is stuck in production queues. Time is time from the customer’s viewpoint.
Of course this doesn’t mean I would ignore the internal components of the lead time… but I would include all of them because that is what the customer experiences.
While I am on the topic of metrics, I want to reiterate the importance of also having a specification or standard. It is not enough to simply measure something and graph it. That does nothing but consume people’s time. Each instance must be compared against a specification. “Lead time” doesn’t tell me anything. What I want to know is “Was this order on time?” Yes or no? How late was it? What delayed this one? Why? Then launch into the problem solving process.
Once a standard is being consistently met it is appropriate to then ask whether you want to set the bar higher. But to try to simply measure your way into excellence, without regard for stability and sources of variation, is an exercise in frustration.
As you look at the various things you measure, ask yourself if your metrics are reflecting the experience of the internal or external customer. That can help reduce some of the questions about what is “in” or “out” of the measurement.