Published 2026-07-07 · Practical guides · Tooling Trials
Run a Fair Tooling Trial
Compare tools under controlled conditions and record the results needed for a purchasing decision.
A useful trial answers a specific question: can this tool meet the part requirement with better cost, life or cycle time? Decide that question before the supplier demonstration begins.
State the baseline and the target
Record the current tool and assembly, material condition, operation, cutting data, coolant, tool life and inspection results. Choose a target and minimum quality requirements.
Agree how tool life ends, such as an inspection limit or defined wear condition. Set inspection intervals, stop conditions and the person authorized to make process changes.
Make the test conditions comparable
Keep the material, setup and inspection method comparable. Record unavoidable changes instead of quietly assigning them to the tool.
Decide whether you are testing a drop-in replacement at existing settings or comparing separately optimized processes. Running different geometries at identical feeds and speeds may answer the first question without answering the second.
Repeat both tools. Where practical, randomize the run order so one tool does not always get the earlier or later conditions. If material lot or setup period matters, test both within each comparable group. This applies the randomization and blocking principles described in NIST's experimental-design guidance.
Capture the whole result
For each run, record tool identity, settings, accepted parts, rejects, cycle time, change time and the reason the tool was removed. Save the relevant inspection results and wear observations.
Compare the range as well as the typical result. One exceptional edge does not establish expected production life. A tool that improves cycle time but creates extra inspection or rework needs those costs included.
Make a scoped decision
Approve the tool for the job and setup actually tested, request another trial or keep the current process. Record the reason and the complete approved assembly.
Give purchasing the part numbers, quantities and any required accessories. Give the next shift the accepted settings and inspection plan. A trial is not complete until the result can be used again.
Sources
NIST/SEMATECH: randomized block designs and nuisance factors