Wire Harness Labor Time Standards: Building and Tuning Your Own

Standardize at the Operation, Not the Harness

A single hours-per-harness number cannot be reused because no two harnesses are alike. Break the work into operations that scale with something you can count on the drawing: wires cut, terminations crimped, seals inserted, connectors populated, splices made, feet of sleeving, labels applied, test points. Each operation gets a run time per unit and, where it applies, a setup time per lot.

Machine operations such as an automatic cut-strip-crimp machine have their own rate, and the operator may tend more than one machine, so record machine time and labor time separately. Estimating then becomes multiplication against a bill of operations, which is auditable and explainable to a customer.

• Cut and strip per wire, by gauge range and length band
• Crimp per termination, by terminal family and applicator type
• Connector population per cavity, with extra time for seals and locks
• Tie, tape, sleeve and lace per bundle foot
• Test and inspection per circuit and per harness

Setup Time Is the Hidden Driver for Small Lots

A lot of ten harnesses can spend more time on changeover than on production. Applicator swaps, strip-length and crimp-height adjustment, first-piece pull test, reel changes, formboard layout and test program loading all happen once per lot. Capture setup as its own standard, split by operation, and divide it across the lot quantity in the quote.

If you blend setup into run time, short runs look cheap and long runs look expensive, and your quote ladder will be backwards.

Add Complexity Factors for Hand Assembly

Hand assembly on the formboard does not scale linearly with circuit count. Shielded cable termination, drain wire handling, heat shrink with solder sleeves, tight-pitch connectors, overbraid, backshell assembly and tight bend radii all add time. Define a small set of complexity multipliers or add-on operations with clear criteria (for example, shield termination as its own operation) rather than letting each estimator apply a personal fudge factor.

Three or four well-defined factors outperform a dozen vague ones.

Tune Standards From Actuals, Not Opinion

Your ERP or shop data collection already holds actual hours by job and operation if operators clock against routings. Compare quoted to actual by operation, not by job total, since a job can come in on hours while two operations are badly wrong and offsetting. Look for operations where actuals consistently run above standard, then ask whether the cause is a bad standard, a recurring quality rework, or an unlogged wait.

Revisit at a fixed cadence and keep a change log so a revised standard does not silently alter old quotes.

• Variance by operation, over the last 20 or more jobs
• Rework hours separated from first-pass hours
• Operator and shift spread, to see learning effects
• Prototype vs production results kept apart

What Not to Do

Do not import a vendor's or association's published times without checking them against your own build observations, and do not set standards from your best operator on a good day. Do not use the same standard for a first-article build and a mature production run: the learning curve on a new harness is real and often shows up in the first several units. If your work is low-mix and repeat, the actual routing from the last order may be a better estimate than any standard.

Where AI Fits

A language model can read the drawing, produce the operation counts, and pull actual hours from comparable past harnesses in your ERP so the estimator checks a draft instead of building one. It can also point out operations where your standard disagrees sharply with recent actuals. The standards themselves, and the sign-off on any change, should stay with a person.