Anomaly Detection

The project is based on a set of data collected from a manufacturing process, in which both in-control and out-of-control data are present. You are asked to develop a method or a procedure to identify the data falling in the respective categories, i.e., which ones are in-control and which ones out of control. This is amounted to a Phase I analysis, whose purpose to isolate the in-control data for estimating the in-control distribution parameters so that a monitoring scheme can be set up for future missions.

Description of the dataset

(a) This is a manufacturing related dataset, and it has a total of 552 data records.

(b) Each row is a data record. Each data record contains 209 values.

(c) The physical meaning of each value is omitted.