In many small job shops, the biggest driver of long lead times is not cycle time. It is the time work orders spend sitting still between operations. Parts wait for a press brake opening, sit in front of welding, stall in inspection, or get parked because a traveler is missing, material is short, or the next machine is tied up. That waiting time is queue time, and if you do not measure it, it quietly becomes your schedule. For a complete overview, see our manufacturing execution system software guide.
The good news is that queue time can be tracked with simple timestamps and a disciplined routing process. Once you can see wait time at each step, you can rank the worst delays by work center and by job, separate true capacity constraints from scheduling problems, and shorten lead times without buying another machine or hiring another operator. For many shops, the first win is simply making invisible waiting visible.
What queue time means in a job shop
Queue time is the elapsed time a job spends waiting between the completion of one routing step and the start of the next. It is not setup time, run time, move time, or inspection time unless those steps are specifically defined as separate routing operations. It is the gap between operations.
For example, if operation 20 on a work order finishes Tuesday at 1:10 p.m. and operation 30 starts Wednesday at 9:00 a.m., the queue time before operation 30 is 19 hours and 50 minutes of elapsed time. If your shop runs one shift, you may also want to calculate a second number in scheduled hours only, which strips out nights and weekends for cleaner internal comparisons.
This distinction matters because queue time is often where quoted lead times are lost. A part may only need 90 minutes of actual touch time across four operations, but still ship three weeks late because it waited in six different places.
If you already track completion against due dates, queue time gives you the missing layer underneath lateness. It pairs especially well with order aging tracking and schedule adherence analysis.
Why small shops should track queue time separately from machine utilization
Shops often assume a busy machine is the bottleneck. Sometimes that is true. Often it is not. A work center can look busy while jobs pile up there because of poor release timing, oversized batch logic, dispatching habits, missing tooling, or rework loops. On the other hand, a machine can look underutilized because upstream jobs are not arriving in a ready state.
Queue time tracking answers a different question than utilization:
- Utilization asks how much a machine ran.
- Queue time asks how long jobs waited to be worked on.
- Lead time asks how long the customer waited for the order.
Those are related, but they are not interchangeable. If you want a fuller picture of available capacity versus actual waiting, pair queue data with machine utilization by shift. If you want to understand whether quality issues are creating hidden recirculation and extra waiting, connect it to rework tracking.
The idea also aligns with the broader lean view that elapsed time and work-in-process matter as much as direct processing time. For foundational context, Wikipedia’s overview of lead time is a useful plain-language reference, and the National Institute of Standards and Technology provides practical manufacturing improvement resources for small and midsize firms.
The minimum data you need to start
You do not need a complex MES rollout to begin. For each work order and operation, start by capturing four basic fields:
- Work order or job number
- Operation sequence such as 10, 20, 30
- Operation complete timestamp
- Next operation start timestamp
From those two timestamps, you can calculate queue time between steps. Over time, add a few more fields that make the data much more useful:
- Part number
- Work center code for the next operation
- Quantity
- Promise date or due date
- Planner or scheduler
- Queue reason code when the wait exceeds a threshold
The most important design choice is to define the queue against the next work center. If a part finishes turning and then waits 18 hours before milling starts, that queue belongs to the milling work center as a demand signal. That lets you rank where jobs are waiting to get in.
Simple reason codes that fit a small shop
Do not create 40 reason codes. Use a short list that operators and supervisors will actually apply:
- Machine busy
- Operator unavailable
- Setup/tooling not ready
- Material shortage
- Program or print issue
- Inspection hold
- Batching decision
- Hot job preemption
- Outside process delay
- Unknown
These codes help you avoid treating every queue as a raw capacity problem when many are caused by release discipline, readiness, or priority churn. If shortages are a recurring source of waiting, you should also review material shortage tracking.
How to capture queue time at each routing step
The cleanest method is event-based tracking inside your production system. Every time an operation is completed, the system records the timestamp. Every time the next operation starts, the system records the next timestamp. Queue time is the difference.
If you are still working from paper travelers or spreadsheets, you can start with a manual version:
- When an operator completes an operation, mark the actual completion time.
- When the next work center begins work, mark the actual start time.
- Calculate the elapsed time between the two events.
- If the wait exceeds your threshold, record a reason code.
Use a threshold that fits your routing and shift pattern. For example:
- Less than 4 scheduled hours: normal flow
- 4 to 8 scheduled hours: monitor
- More than 8 scheduled hours: queue exception
- More than 24 scheduled hours: escalation
Those are only examples. A one-off mold shop, a laser-cutting shop, and a precision machining job shop will set different thresholds.
Decide whether to use elapsed time or scheduled time
There is no single right answer, but be consistent.
- Elapsed time is best for customer-facing lead time analysis because customers experience nights and weekends too.
- Scheduled time is often better for internal improvement because it avoids overstating waits that occur during off-shift hours.
Many shops benefit from tracking both. Use elapsed queue time for lead time management and scheduled queue time for work center comparisons.
The reports that actually uncover bottlenecks
Once queue time is captured, do not stop at averages. Averages hide the real problem. You want a short list of where lead time is being consumed.
1. Queue time by work center
Sum queue hours by the receiving work center over the past 30 days. Then also calculate:
- Average queue hours per job
- Median queue hours per job
- Maximum queue hours
- Number of queued jobs
- Percent of jobs exceeding threshold
This tells you which areas are absorbing the most waiting and whether the problem is broad or caused by a few extreme jobs.
2. Queue time by job
Rank active and recently completed work orders by total queue time across all steps. This immediately identifies which orders suffered the most waiting and which are at risk now.
A simple rule works well: if total queue time is greater than total run time, the job likely has a flow problem, not a machining problem.
3. Queue time by operation pair
Some delays are not tied to one work center in general, but to one handoff. For example:
- Saw to mill
- Mill to deburr
- Weld to inspection
- Paint to pack
Ranking queue by operation pair shows weak transitions where readiness, transportation, batching, or approval flow is breaking down.
4. Queue reason Pareto
Once your reason-code data is usable, make a Pareto chart or simple ranked table by queue hours. If the top two reasons account for most queue time, that tells you where to focus first. The same prioritization logic is useful in scrap analysis, as shown in scrap Pareto analysis.
A practical scorecard for ranking the worst bottlenecks
One of the most useful views for a small shop is a bottleneck scorecard that combines volume, severity, and customer impact. Here is a simple example:
| Work Center | Queued Jobs | Total Queue Hrs | Avg Queue Hrs | % Over Threshold | Late Jobs Affected |
|---|---|---|---|---|---|
| CNC Mill | 18 | 96 | 5.3 | 44% | 7 |
| Inspection | 10 | 72 | 7.2 | 60% | 6 |
| Press Brake | 9 | 41 | 4.6 | 33% | 3 |
In this example, inspection may be the first target even though CNC mill has more total queue hours, because inspection has a higher average wait, a higher exception rate, and nearly as many late jobs tied to it.
That is the point of ranking. You are not just finding the busiest area. You are finding the delay point causing the most lead-time damage.
How to shorten queue time without adding machines or labor
Once the data shows where waiting occurs, the fixes are usually more operational than capital-intensive.
Tighten release discipline
Do not launch work into the shop unless the next few steps are truly ready. Releasing jobs with missing material, incomplete prints, or unresolved programming issues creates fake load and blocks real priorities.
Reduce batch size where queues are inflated by lot logic
Large batches can improve local efficiency while destroying flow. If one work center waits to accumulate a full day of similar work before running, every job behind that logic absorbs queue time. Test smaller transfer batches or partial-release moves between operations.
Protect constrained support steps
Many shops focus on obvious production machines while ignoring shared support constraints such as first-article approval, inspection, deburr, or packing. If your queue report shows waiting stacked up before one of these areas, elevate it in the schedule instead of treating it as overhead.
Stop hot-job priority churn
Frequent expedites can make every queue worse. Each interruption pushes existing jobs back and increases setup losses. If you log a high share of queue time under hot job preemption, tighten your expedite approval process and make the tradeoff visible.
Use finite promises at the bottleneck work center
If one work center repeatedly carries the largest queue, stop scheduling it with unlimited optimism. Promise jobs based on realistic available slots there, not just material readiness or front-office due dates.
Separate capacity issues from readiness issues
If the top queue reason is machine busy, you may have a true load problem. If the top reasons are tooling not ready, inspection hold, or program issue, more machine time will not solve it. This is why reason codes matter.
Build a weekly queue review into your production meeting
Queue time only helps if it changes decisions. A short weekly review is enough to start:
- Review top 10 jobs by current queue time.
- Review top 5 work centers by total queue hours last week.
- Check which late jobs were delayed mainly by queue, not run time.
- Review top queue reasons and assign actions.
- Track whether last week’s bottleneck improved or just moved.
This last point is important. When you unclog one work center, the queue may shift downstream. That is normal. Flow improvement is an ongoing process, not a one-time fix.
Common mistakes to avoid
- Using only averages. Always look at counts, medians, and exceptions too.
- Blaming the previous operation. Queue should usually be attributed to the next work center waiting to receive the job.
- Ignoring support departments. Inspection, shipping, and programming can be major queue creators.
- Tracking too many reasons. Keep coding simple enough to use consistently.
- Measuring without acting. Queue reports need owners, thresholds, and escalation rules.
What good looks like after 30 to 60 days
After a month or two of consistent tracking, most shops can answer questions they could not answer before:
- Which work centers create the most waiting before jobs can move?
- Which customers or part families experience the longest internal queues?
- Which queue reasons are structural versus occasional?
- Which delayed jobs were hurt more by waiting than by actual processing?
- Where can lead time be reduced without any capital spend?
That visibility often changes scheduling behavior quickly. Supervisors stop arguing from gut feel, planners stop flooding the floor with unreleasable work, and management can target the true lead-time constraint instead of reacting to whichever machine looks busiest.
If you also want to quantify the financial effect of lost capacity from blocked flow, tools like our downtime cost calculator can help frame the impact. And if queue is being amplified by quality losses, review your scrap cost through the scrap cost calculator.
Conclusion
Queue time is one of the biggest hidden causes of long lead times in small job shops. When you measure the wait between routing steps, rank the worst bottlenecks by work center and job, and tie delays to simple reason codes, you can improve flow without adding machines or labor. The goal is not more data for its own sake. It is faster movement of real jobs through the shop.
If you want a practical way to track queue time, bottlenecks, and work order flow in one place, start a free FactoryOS trial and see where your lead time is really being lost.