Industrial Statistics with Minitab

21.4 Plastic (1st Part)

Plastic parts are produced by injection and are supplied to a manufacturer of consumer electronics devices. These parts are used as covers for audio and video equipment, and among the references there is a length whose specifications are 452 mm ± 0.5 mm. This is of extreme interest to the client and has been giving problems.

As part of a project to solve this problem, it is decided to control this length by means of a - R chart. Each hour, a sample is taken consisting of 4 consecutive parts (the working schedule is from 7 a.m. to 3 p.m., the first sample is taken at 8 a.m.). The data are in file PLASTIC_1.MTW.

Which conclusions can be drawn from the - R chart? For this, using all default options, choose:

Stat > Control Charts > Variables Charts for Subgroups > Xbar–R. All observations for a chart are in one column: Length. Subgroup sizes: 4.

The process is clearly out of control. The R chart shows only one point outside the control limits (the rest does not seem to indicate signs of special causes).

The graph of the sample shows a nonrandom behaviour. During the first hours of the second day (around sample 10), there is a clear increase in the length ...