Setup Reduction for Small Manufacturers: How to Measure Changeover Time by Machine and Cut Lost Capacity Without New Equipment

Long setups quietly steal capacity from small manufacturers. A machine can look busy all day while producing fewer sellable hours than expected because too much time disappears between the last good part of one job and the first good part of the next. Many shops know setups are a problem, but they do not measure them consistently enough to see where the biggest losses really are. For a complete overview, see our manufacturing execution system software guide.

The good news is that setup reduction does not have to start with new equipment, automation, or a major lean initiative. It starts with a clear definition of changeover time, a simple way to track it by machine and part family, and a practical method for turning those minutes into scheduling and capacity decisions. Once you can see where setup time is going, small improvements often free up meaningful capacity with very little capital.

What changeover time should include

If your shop does not define setup the same way every time, your data will be noisy and hard to trust. The simplest useful definition is this:

Changeover time = time from the last good part of the previous job to the first good part of the next job at standard production conditions.

That definition matters because it avoids two common errors:

• Starting too late. If you start the clock only when the operator begins touching tools, you miss waiting, cleanup, material movement, paperwork, and line clearance.
• Ending too early. If you stop the clock at first machine cycle instead of first good part, you hide trial cuts, offsets, first-article approval, and startup scrap.

For most small shops, setup should include whatever prevents the machine from producing the next job correctly at normal pace. Depending on the process, that can include:

• Removing prior tooling, jaws, fixtures, or material
• Cleaning chips, coolant, dust, residue, or debris
• Fetching tools, gages, fixtures, programs, and raw material
• Loading the next program, recipe, or traveler
• Installing tooling, fixtures, or workholding
• Adjusting offsets, stops, sensors, and parameters
• First-piece inspection and approval
• Startup scrap and trial runs until the first good part

This is closely aligned with the logic behind SMED: separate work that can be done while the machine is still running from work that can only happen when the machine is stopped. You do not need a formal lean program to use that idea. You just need a consistent stopwatch definition.

Define start and end points by machine type

The exact setup boundary should be written down by machine type so operators and supervisors record the same event the same way.

CNC mills and lathes

• Start: last good piece completed on Job A, including any final unload tied to that piece
• End: first good piece approved on Job B with production-ready offsets and tooling

Press brakes, stamping presses, and punch equipment

• Start: last good part from prior job
• End: first good part of next job after die/tool setup and required checks

Injection molding or similar process equipment

• Start: last acceptable shot or part from previous run
• End: first good part after mold change, parameter adjustment, purge, and approval

Assembly or packaging cells

• Start: last acceptable unit of prior order
• End: first acceptable unit of next order after line clearance, material swap, and verification

If first-article inspection happens away from the machine, decide one rule and use it every time. In many small shops, it is best to keep that time inside setup because the machine is still effectively unavailable until approval is complete.

Track setup minutes by machine and part family

Most small manufacturers do not need a complex data collection project to start. A simple spreadsheet, traveler field, whiteboard log, or shop floor app is enough if the entries are disciplined. The key is to track at a level that reveals repeatable patterns.

At minimum, capture these fields for every changeover:

Part family is especially important. If you only track by part number, the data becomes too fragmented. Family-level analysis makes improvement practical. Examples include:

• Aluminum housing family using common jaws and tool package
• Small stainless shaft family on the same lathe platform
• Press brake jobs using the same material thickness range
• Mold changes between same resin family versus full resin change

If you already track downtime, setup can become one of your downtime categories. That makes it easier to see whether lost capacity comes more from breakdowns, waiting, or changeovers. For a good starting point, see this guide to using a simple downtime code system.

How to make setup data reliable in a small shop

Bad setup data usually comes from vague rules, not bad people. Keep the process simple:

1. Use one definition. Post the start and end rule at each work center type.
2. Record in real time. Do not ask operators to reconstruct setup time at the end of the shift.
3. Capture one main delay reason. Too many codes create guesswork. Start with 6 to 10.
4. Audit a few setups each week. A supervisor can compare actual events to what was logged.
5. Review median, not just average. One extreme event can distort averages.

Useful delay reasons might include:

• Tool search
• Fixture change
• Program issue
• Material not ready
• First-article wait
• Cleaning
• Adjustment / trial cuts
• Operator handoff

If setup errors are causing repeated trips for prints, tooling notes, or inspection criteria, standardizing job information can be one of the fastest wins. Related reading: how digital traveler packets reduce setup errors and first-article delays.

Calculate the true scheduling cost of long setups

Setup reduction matters because setup time consumes capacity at the bottleneck and pushes work into overtime, queue time, or missed due dates. The cost is not only labor. It is also the production you could have run instead.

Step 1: Convert setup minutes into lost production hours

Example: a machining center averages 4 changeovers per day at 45 minutes each.

That equals 180 minutes, or 3 hours per day, where the machine is not making saleable parts.

Across a 5-day week, that is 15 hours. Across a 50-week year, that is 750 hours of machine capacity tied up in setup.

Even if your actual numbers are smaller, this calculation quickly shows why setup deserves management attention.

Step 2: Look at the bottleneck first

One hour saved on a lightly loaded machine is not equal to one hour saved on your bottleneck. If the bottleneck is overloaded, setup reduction there directly improves throughput and lead time performance. If you have not mapped available hours by work center, start with this capacity planning guide.

Step 3: Estimate scheduling impact

Long setups force planners into bad tradeoffs:

• Run larger batches than demand really needs
• Delay urgent jobs because changing over is too painful
• Build ahead and increase WIP
• Accept overtime to recover lost machine time
• Quote longer lead times to protect the schedule

In other words, setup time changes scheduling behavior. It does not just reduce run time; it often increases inventory and complexity too. If your planners struggle with constant reprioritizing, this article on production scheduling for small job shops is a useful companion.

Step 4: Put a dollar lens on the worst offenders

You do not need perfect accounting to prioritize. Use a simple illustrative estimate:

• Machine burden rate or internal shop rate per hour
• Average setup hours per week on that machine
• Overtime premium or outside processing risk created by the lost hours

For example, if a bottleneck machine loses 10 setup hours per week and your internal burden estimate is $100 per hour, that is roughly $1,000 per week of consumed capacity before considering late shipments or overtime. Use examples like this carefully as planning signals, not audited financial statements.

If you want a quick way to frame the capacity impact of non-running time, the downtime cost calculator can help translate lost machine hours into a more visible business discussion.

Prioritize setup reduction opportunities that are actually worth doing

Once you have 4 to 8 weeks of data, rank opportunities using three filters:

1. Total minutes lost by machine
2. Frequency of the changeover
3. Ease of improvement without capital spending

This prevents teams from chasing the single ugliest setup if it only happens twice a quarter.

A simple prioritization matrix

Start with high-impact, high-ease items. Those are usually basic organizational fixes rather than engineering projects.

Simple improvements that reduce setup time without new equipment

Move external work out of machine downtime

This is the classic first step. Anything that can happen before the machine stops should happen before the machine stops:

• Stage tools, inserts, holders, fixtures, gages, and material
• Print or preload setup sheets and programs
• Preset tool lengths or offsets offline where possible
• Verify raw material and job packet before the previous run finishes

Standardize setup kits

For recurring part families, create repeatable setup kits with labeled tooling, jaws, hardware, inspection tools, and instructions. This reduces searching and variation between operators.

Improve fixture and workholding organization

Many long setups are really storage and retrieval problems. Assign fixed locations, shadow boards, and clear labeling for high-use setup items. A five-minute search repeated several times a day adds up fast.

Reduce first-article approval delays

If inspection is a frequent bottleneck, define response expectations, prioritize first-piece checks, and clarify who can approve what. In some shops, setup time falls more by fixing approval flow than by changing tooling.

Build setup sheets around part families, not tribal memory

Short, practical setup sheets with fixture photos, tool lists, critical dimensions, and common offset notes can significantly reduce trial-and-error. If one experienced operator carries the whole process in memory, your setup time will be inconsistent by shift and by person.

Sequence jobs to reduce hard changeovers

Scheduling can lower setup time even before process changes do. Group jobs by material, fixture family, resin, color, thickness, or tooling package where due dates allow. This is one reason setup data by part family is so valuable: it shows which transitions are naturally easier.

Connect setup reduction with quality

A fast setup that creates startup scrap is not a good setup. Track first-pass quality on the first pieces after changeover. If you are not already measuring it, read how to measure first-pass yield by work order. Setup improvements should reduce both lost time and startup defects.

Build a weekly review that drives action

Do not let setup logs become another report nobody uses. Once a week, review:

• Top 5 machines by total setup minutes
• Top 5 part families by changeover frequency
• Median setup time versus target by machine type
• Top delay reasons
• Startup scrap tied to changeovers
• One improvement action owner and due date

Keep the meeting operational, not theoretical. Ask:

• What setup happened most often?
• What delayed it?
• Could that task have been external?
• What should be staged next time?
• What standard needs to be documented?

If equipment reliability is blurring the picture, separate true setup from maintenance-related stoppages. Planned upkeep should not be hidden inside changeover data. See this preventive maintenance system for small manufacturers for a simple way to keep those categories clean.

What a good first 30 days looks like

You do not need a perfect rollout. A practical 30-day plan is enough:

1. Week 1: define setup start/end rules for each machine type and choose 6 to 10 delay codes.
2. Week 2: start logging setup minutes on the top 3 to 5 critical machines.
3. Week 3: group jobs into part families and review the first data for obvious causes.
4. Week 4: implement 2 or 3 no-capex improvements such as staging kits, setup sheets, or inspection response changes.

At the end of the month, compare median setup time on those machines before and after. Even modest reductions can free up enough weekly capacity to absorb rush work or reduce overtime pressure.

For additional guidance on lean manufacturing practices for smaller operations, the National Institute of Standards and Technology Manufacturing resources are a credible place to explore practical improvement frameworks.

Conclusion

Setup reduction starts with measurement, not guesswork. When you define changeover the same way on every machine, track minutes by machine and part family, and translate those minutes into lost capacity, the biggest opportunities become obvious. In many small shops, the best gains come from staging, standardization, scheduling discipline, and faster first-article flow, not from buying more equipment.

If you want a simpler way to track setups, downtime, work orders, and machine capacity in one place, start a free FactoryOS trial and see where your lost capacity is really going.