Value stream mapping example: 40 minutes of work, 10 days of waiting

Orders take nearly two working weeks to get through the operation. The first request is for faster preparation equipment. It is the most visible place where people are busy.

Follow an order and the picture changes. Approval takes 12 minutes, preparation takes 18, and checking and dispatch take 10. Forty minutes of active work, spread across days of waiting.

That does not settle which investment to make. It tells the team what to find out before making it: where does the order wait, and what allows it to move?

Choose one route and a clear boundary

A value stream map shows the work and the information that controls its movement. The Lean Enterprise Institute distinguishes the current state, what happens now, from the future state, how the team proposes the flow should work.

For this fictional example, choose one family of repeat orders with the same route. Start when a complete order enters release and approval. Finish when it is dispatched. Supplier replenishment and time in transit sit outside this boundary.

The operation completes 50 orders per working day. A working day contains 480 minutes. Several people or benches can handle different orders at once, so an 18-minute processing step does not imply that one person completes all 50 orders.

Before drawing, agree what an order is. Mixing individual items, pallets and customer orders in one calculation makes the result meaningless, even if every number came from a real system.

Collect the waiting as carefully as the work

Walk the route with the people who do it. Record an order's arrival, start and finish at each stage. Separate active processing from time waiting for a decision, material, a batch or a free resource.

Here are the fictional observations used in our map. Queue counts are averages over a representative, reasonably stable period, rather than one snapshot taken on a quiet morning.

Stage	Average orders waiting before it	Active processing per order	What tells the team to start
Release and approval	100	12 min	The planner releases a list
Prepare the order	150	18 min	The preparation team selects from its list
Check and dispatch	250	10 min	A ready list is reviewed for checking and dispatch

The start rules are part of the investigation. How often are those lists updated? Do all teams use the same priorities? Does an order wait after it is ready because nobody can see that fact?

A route of boxes alone would miss those questions. Draw the instructions as well as the order movement, and label the decisions that are still unclear.

Build the timeline in consistent units

Little's Law relates average work in a system, throughput and time in that system. Applied to a waiting queue with consistent boundaries and stable flow, average queue count divided by its flow rate estimates the average wait.

Here, all orders pass through each queue, so the stated flow rate is 50 a working day at each stage. Split routes, rework and changing throughput would need separate treatment.

The queue counts exclude orders being processed, so add the 40 minutes of active processing. The simplified estimate is 10 working days plus 40 minutes, or 4,840 working minutes.

If your WIP count already includes orders being processed, dividing it by throughput estimates their total time inside that boundary. Adding processing again would count it twice.

Check the estimate against actual order timestamps. A growing backlog or a period with unusual demand can make a tidy average a poor description of the current experience. Look at the spread as well as the mean, especially for late orders.

Working days also need a calendar. Ten working days are not necessarily ten calendar days. The conversion here uses 480 minutes per working day for comparison; it does not promise a dispatch date.

Compare the proposed changes

Suppose faster equipment reduces preparation from 18 minutes to eight. Hold throughput and the queues fixed for this first sensitivity test.

Active processing falls from 40 to 30 minutes. Total time falls from 4,840 to 4,830 working minutes, a reduction of about 0.21%.

Now test a different assumption: the final wait falls from five working days to one, with other waits and processing unchanged. Total waiting becomes six days, so total time is 2,920 working minutes. That is about 39.67% below the current estimate.

Simplified scenario	Waiting	Active work	Total in working minutes
Current state	10 days	40 min	4,840
Faster preparation only	10 days	30 min	4,830
Shorter final wait only	6 days	40 min	2,920

This comparison is a reason to investigate the final wait. It is not evidence that the team can remove four days just by asking it to work differently.

Faster preparation may improve throughput if that stage constrains the operation. A large downstream queue may reflect a slow check, a release rule, a shipment schedule or several causes. Queue size alone cannot distinguish them.

Turn the future map into a test

Take a sample of orders waiting before the final stage. Record when they became ready, what held them, and what finally triggered the next action. If the ready-list routine causes a delay, test a change to that routine. If checking lacks capacity, changing the list will not solve it.

Write down the proposed rule, its owner and the conditions for the trial. For example, a more frequent review of ready orders needs a person able to check them and a dispatch arrangement able to take them. Moving the queue outside the chosen boundary would merely improve the reported number.

Compare actual order lead times, completed orders, late deliveries and errors under comparable conditions. Track queue counts throughout the trial, including any queue that grows elsewhere.

Keep necessary checks. Ten minutes spent confirming the order is correct may protect the customer; it is not automatically ten minutes of value-adding work, nor automatically ten minutes to remove. Assess the purpose of each task before classifying it.

Investigate the explanation behind the wait →Use evidence to test the cause before choosing an action. The largest queue points to a question, not a completed diagnosis.

The map earns its place when it changes the next question. Before ordering faster equipment, find out whether the order is waiting for a machine, a decision or permission to move.

Sources

• Lean Enterprise Institute. Value stream mapping. Material and information flow, current and future states, process and lead time.
• Little, J. D. C. (2011). Little's Law as viewed on its 50th anniversary. The relationship between average work, throughput and flow time.

The route, queue counts, work times and proposed changes are fictional. The example assumes a stable common flow through all three stages. It illustrates which assumptions to investigate, not a demonstrated lead-time reduction.