Little's Law is a simple, exact relationship that governs any stable process: the average amount of work-in-process (WIP) equals the average throughput multiplied by the average lead time. Written as WIP = throughput × lead time, it proves that if you want shorter lead times at the same output, you must carry less WIP.
It sounds almost too plain to be useful, but Little's Law is one of the few laws in operations that is always true, regardless of how a process is arranged, how variable it is, or what it makes. That is what makes it powerful on a plant floor: it lets a planner reason about lead time and inventory without a simulation, using numbers already on the board. This post gives the formula, a worked example with real numbers, the assumptions that keep it honest, and how to put it to work.
What is Little's Law?
Little's Law states that for any stable system, the long-run average number of items in the system equals the long-run average arrival rate multiplied by the average time each item spends in the system. In queueing notation that is L = λW. On a factory floor the same law reads WIP = throughput × lead time, and any one of the three can be found from the other two.
The law was proved by John D. C. Little in 1961. Before then the relationship was used but unproven; Little's paper showed it holds for essentially any queueing discipline, which is why it applies to a machining cell, a whole plant, an order backlog, or a hospital waiting room without modification. The three terms have plain-floor meanings:
- WIP the average number of units inside the boundary you drew: parts between the first and last operation, orders open in the system, jobs on the schedule.
- Throughput the average rate the system completes and releases units, in units per hour, day, or shift. This is the exit rate, not the theoretical machine rate.
- Lead time also called flow time or cycle time in the queueing sense: the average clock time a unit spends inside the boundary, from entry to exit.
By the numbers. John D. C. Little published the first general proof of the formula L = λW in Operations Research in 1961, and later work extended it to essentially any stable system regardless of arrival pattern or service order (Little's law). The Lean Enterprise Institute lists lead time, throughput, and WIP among the core measures of a value stream, and Little's Law is the identity that ties them together (Lean Enterprise Institute, Lean Thinking and Practice).
Why does cutting WIP cut lead time?
Because throughput is set by your slowest necessary step, not by how much WIP you pile in front of it. Rearrange the law to lead time = WIP ÷ throughput. If throughput is fixed by the bottleneck, then lead time rises and falls in direct proportion to WIP. Doubling the queue in front of a station does not make the station faster; it just makes every unit wait twice as long.
Take a line that finishes 500 units a day and carries 2,000 units of WIP on average. Little's Law gives a lead time of 2,000 ÷ 500 = 4 days. Now suppose you cut WIP in half to 1,000 units without touching the equipment. The line still finishes 500 a day, so the new lead time is 1,000 ÷ 500 = 2 days. You halved lead time by releasing less work, not by buying a machine. This is the arithmetic behind pull systems, kanban limits, and one-piece flow: they all work by holding WIP down so the same throughput clears faster.
How do you apply Little's Law on the floor?
You apply it as a quick check and a lever, not a one-time calculation. The steps below turn the identity into a routine you can run on any value stream in an afternoon.
- Draw the boundary. Decide exactly where the system starts and ends, a single cell, a department, or the whole plant from raw release to shipping. WIP, throughput, and lead time must all refer to the same boundary or the numbers will not agree.
- Measure the two you trust. Count average WIP inside the boundary and measure throughput as the actual exit rate over a representative window. These are usually easier to observe reliably than lead time.
- Solve for the third. Compute lead time = WIP ÷ throughput. Compare it to the lead time you quote customers. A large gap means orders are sitting in queue far longer than anyone thinks.
- Attack WIP first. To shorten lead time without adding capacity, set WIP caps, kanban limits, a kanban ceiling, or a release rule that only starts new work when a unit finishes. Lead time falls in direct proportion.
- Re-measure and hold. Recount WIP and throughput after the change. If throughput held steady while WIP dropped, lead time fell exactly as the law predicts. Lock the new WIP cap into standard work so it does not creep back up.
What counts as WIP, throughput, and lead time?
The law is exact, but only if the three terms are measured consistently against the same boundary and the same units. WIP is an average count of items in the system, not a peak or a wish. Throughput is the exit rate you actually achieve, which already includes downtime, scrap, and breaks, so it is almost always lower than a machine's nameplate rate. Lead time is total clock time inside the boundary, most of which is usually queue time, not processing time.
A worked table makes the units concrete. Notice that lead time is expressed in the same time base as throughput.
| System | Average WIP | Throughput | Lead time = WIP ÷ throughput |
|---|---|---|---|
| Machining cell | 120 parts | 40 parts/hour | 3 hours |
| Assembly line | 2,000 units | 500 units/day | 4 days |
| Order backlog | 90 open orders | 15 orders/day | 6 days |
What are the assumptions and limits of Little's Law?
Little's Law is exact only for a stable system observed over a long enough window, one where average WIP, throughput, and lead time are not trending up or down. It says nothing about variability, so two lines with identical averages can have very different worst-case lead times. It is a statement about averages, not a promise about any single order.
The practical traps are all about stability and boundaries. If you are ramping up and WIP is growing, a snapshot will mislead you, because units entering are not yet balanced by units leaving. If the boundary leaks, parts scrapped mid-stream, orders split or merged, your counts drift. And the law gives an average lead time; to promise a delivery date you still need to understand the spread, which is where statistical process control and queue-variability thinking come in. Used with those caveats, the law is a reliable compass, not a fragile estimate. A good habit is to compute lead time from the law and compare it to the lead time you actually quote customers; when the two disagree, the law is usually right and the quote is stale.
How does Little's Law connect to lean and throughput?
Little's Law is the arithmetic proof behind most of lean's inventory rules. Pull systems, one-piece flow, and small batches all lower average WIP, and the law guarantees that lower WIP at steady throughput means shorter lead time. That is why a lean line quotes shorter lead times even when its machines are no faster: it simply carries less work in queue. Cutting WIP is the cheapest lead-time lever a plant has, because it costs release discipline rather than capital. It also improves quality, because a defect made in a low-WIP line is caught within hours instead of being buried in a week's worth of finished queue.
It also frames the constraint conversation. The Theory of Constraints says throughput is set by the bottleneck, so if you cannot safely raise throughput, the only honest way to shorten lead time is to cut WIP, exactly what Little's Law predicts. Map the flow with value stream mapping to see where WIP piles up, then use the law to price the lead-time payoff of draining each queue. For the underlying metrics, see throughput and lean manufacturing. Seeing true WIP and true exit rates in real time, rather than guessing from a spreadsheet, is exactly what a live operations view gives a planner, the kind of connected floor data Harmony puts on the board, shown in the CLS case study.