JMP platforms and statistical methods

Graph Builder:
Comparative dot plots and box plots

Distribution:
Histograms, summary statistics, one-sample inference for population mean and standard deviation, process capability analysis

Fit Y by X:
Two-sample inference comparing population means and variances

Objective

Use one- and two-sample statistical inferential methods for means and variability, along with process capability analysis to evaluate and compare the performance of adhesive materials to each other and to process specifications.

Problem statement

Engineers in a product development team for a manufacturer of medical equipment packaging are evaluating the performance of two different formulations of adhesive that's used to seal the packaging.

An important characteristic is that the adhesive is strong enough to withstand the usual sterilization and handling rigors while not so strong that it is difficult for a medical practitioner to open easily.

The engineers produce a sample of n=50 pouches sealed with each of these two new adhesive formulations (A and B). The peel strength on each specimen is measured using a device known as a universal testing machine.

The tester records the force (in Newtons) being applied across a range of peel displacement. The average force across a defined range of peel displacements is used as a single data value summarizing a test.

The specification for the adhesion is an average peel strength (m) of 11.0 N, along with +/- 3 standard deviations (3s) ≤ 1.5 N (i.e., s ≤0.5 N).