Throughput Time

Throughput time and lead time get used interchangeably in most shop conversations, and that is fine for casual use. When something is broken in the value stream, though, the distinction matters. Lead time is the customer's clock. Throughput time is the floor's clock. Improving one without the other leaves either the floor sluggish or the customer waiting longer than the floor is.

"The clock starts when work starts. Everything that happens before that is procurement. Everything after is the stream."

How throughput time works

Throughput time is measured by following one unit from work-start to work-done. The unit is a part, a sub-assembly, or a finished product, depending on which slice of the value stream you are studying. The clock starts at the first value-adding operation. It runs continuously, regardless of whether the unit is being worked on, sitting in a cart, waiting for inspection, or staged for ship. It stops the moment the unit is ready for the next customer in the chain.

The math underneath is Little's Law applied to a value stream: throughput time equals work-in-process divided by throughput rate. If a shop has 500 parts of work in process and ships 100 parts per day, the average throughput time is five days. That formula is the most useful diagnostic in any shop that wants to understand why things take as long as they take. WIP and throughput time move together. Cutting WIP cuts throughput time. Adding WIP, without adding throughput, lengthens it.

The thing that surprises shops new to lean is the breakdown. Most throughput time is queue or wait time, not value-added work. A part that has eight hours of actual machining might have a throughput time of forty hours, because thirty-two of those hours are queue time between operations. The ratio of value-added time to total throughput time is called process cycle efficiency, and in most non-lean shops it runs between two and ten percent. The leverage is not in cutting the eight hours of work. It is in cutting the thirty-two hours of waiting.

The trick to seeing throughput time clearly is measuring it physically, not through software. Chalking a date on a WIP cart the moment material enters the floor turns throughput time into a number anyone can read by glancing at the cart. ERP reports tend to back out throughput time from system events, which often miss the actual physical wait. The chalk method is more reliable and faster to install.

Where throughput time fits on the shop floor

Picture a small contract machine shop running precision brackets for industrial OEMs that make ground-support equipment. Total ordered work-in-process across the shop is about 800 brackets. The shop ships about 200 brackets a week, or 40 a day. By Little's Law, average throughput time is 20 days.

The owner walks the floor on a Friday with a stopwatch and a date marker. The first cut happens. The bracket goes into a cart. Three days later, the cart is finally pulled for the second operation. Three days for one transition. The shop has six operations. If the pattern is similar at each handoff, eighteen days of throughput time is spent in carts, and only two days is in actual work. That single observation drives the entire improvement plan: do not buy a faster machine, do not add a shift. Cut the cart time. A WIP limit between each operation, sized to one day, brings throughput time from 20 days down to about 8 within a quarter, without touching the machining itself.

Common mistakes with throughput time

• Averaging without looking at variation. An average hides spikes. The spikes are the flow problem.
• Confusing it with cycle time. Cycle time is one operation. Throughput time is the whole stream including all the queues.
• Measuring only through ERP events. Software events miss physical wait time. Chalk on a cart is more reliable.
• Trying to fix it with faster machines. Most throughput time is queue, not cycle. Faster machines do not move the needle if the carts still sit.
• Treating it as the customer-facing metric. Customers feel lead time, which includes order entry and material procurement. Throughput time is the floor's metric.

Throughput time and related Lean tools

Throughput time is closely tied to lead time, with the difference being where the clock starts. It is built from many cycle times chained together, with queue time between them. The dock-to-dock measurement is a close relative, scoped to the time between receiving and shipping. And process cycle efficiency, the ratio of value-added time to throughput time, is the metric that exposes how little of the stream is actual work and how much is waiting.