Computational Techniques in Statistical Analysis and Exploitation of CNC Machining Experimental Data

2. Basics Of Data Gathering And Processing In Cnc Machining

Several statistical techniques are used in order to summon and evaluate the appropriate information of manufacturing processes that are controlled in terms of quality characteristics and responses. Parameter values are scouted in a pre-determined value ranges that users specify. In terms of reliability and productivity, obtained data sets must conform to predefined technological and economical constraints.

Before one proceeds on performing experimental design, it is essential to understand the philosophy of collecting data in a systematic manner. Some of the basic issues, one should see to, are the following (Faraway J., 2002):

• •

  The data category should be specifically determined, i.e. whether data is a result of observation or of actual experiments. Moreover, it should be clarified whether data sets are “convenient” samples or are obtained by a designed sample survey. The method of data collection has crucial impact on results and conclusions.

• •

  It should be checked whether there is non-response on the data. The data that cannot be observed may be just as important as the observed ones.

• •

  Investigation for missing values. This is a common problem that is troublesome and time consuming to deal with.

• •

  Quantitative representation and data encoding.

• •

  Data normalization (see also Section 2.1.2) and measurement units.

• •

  Performance of data integrity checks and estimation of errors during data entry.