A Method of Positional Quality Control Testing for 2D and 3D Line Strings

摘要

This article presents a positional quality acceptance control method for 2D and 3D line strings based on a statistical hypothesis test. Two statistical models are applied together: a Binomial Model is applied over a Base Model. By means of the Base Model the method can be applied to any parametric or non-parametric error model. The Base Model represents the hypothesis about the error behavior. The Binomial Model is fixed and consists of counting the number F of fail events in a sample of a determined size. The parameter of the Binomial Model is derived from the Base Model by means of a desired tolerance. By comparing the probabilities associated to F and a statistical acceptance/rejection decision is achieved. This method allows us to know and control the user's and producer's risk of acceptance/rejection. An example using a 2D line string data set from a commercial product is presented. The extension of the method to the 3D line string case is also presented. In order to facilitate the application of the method, some tables linking with F and the control sample sizes are presented.