Ceramic Coating Basics3 min read

Fleet Coating Trials: Choose the Car, Panel or Wash as the Experimental Unit

Choose the experimental unit before a fleet coating trial so repeated readings do not create an inflated impression of evidence.

Updated · NANOLAB Knowledge

Editorial illustration: A Practical Quality-Control Process for Ceramic Coating
Editorial illustration

A fleet trial can produce hundreds of measurements while still containing very few independent comparisons. The important unit is the object assigned to a treatment, not simply the number of cells in a spreadsheet. Decide that unit before choosing vehicles or writing a performance claim.

Start with the decision the trial must support

If the question is whether one complete coating programme reduces fleet maintenance work, assign programmes to comparable vehicles. If the question concerns two coatings on similar paint, independently prepared panels may be appropriate. If every vehicle receives the same coating and is washed repeatedly, the washes describe its progress; they are not independent coating installations.

Observation What it contributes What it cannot replace
Several drops on one panel Local wetting variation More independently coated panels
Several panels on one car Position and application information More vehicles under different use
Several washes of one car A maintenance time series Independent fleet exposure

A worked planning example

Suppose a fleet offers twelve comparable vans and two coating programmes. A practical proposal is six vans per programme, allocated after grouping similar routes and parking conditions. Ten inspection points on each van generate 120 readings, but the programme comparison still involves twelve assigned vehicles. Preserve each van’s identity in the analysis instead of treating all 120 readings as interchangeable.

Agree the main outcome in advance: for example, accepted cleaning time per scheduled wash. Keep paint appearance and coating maintenance events as separate outcomes. Record changes in route, driver, washing contractor and repair history because these can change the interpretation.

Make the report match the design

Report the number of vehicles, panels, visits and readings separately. If a vehicle leaves the fleet, record the reason and its last observation. Do not quietly replace it with a better-performing vehicle. A pilot can justify a larger controlled trial even when it is too small to support a broad durability claim.

The useful deliverable is a vehicle allocation sheet and a consistent follow-up record. A large reading count is valuable only when the structure behind it remains visible.

Technical references

The sources below support the stated concepts or describe their own methods. The worked scenarios are editorial examples, not NANOLAB test results.

Ask NANOLAB about your surface and intended use. For application and care, use the current instructions for the exact product.