**3.4.2 Determining the type of research that needs to be done**

After defining the precise research topic it is important to determine the types of research to be conducted. Each research concludes with a model. There are several types of models:

	- *scientific models*: The phenomenon is carefully observed and the model proposes a theory to explain it. Measurements and analyses will be carried out to validate the model;
	- *engineering models*; Existing solutions are being studied, measurements are being undertaken and improvements of these solutions are proposed and implemented. These activities are repeated until no more optimisations are required;
	- *empirical models;* In this case there is no need for a theory or an explanation for the phenomenon. The model works as a black box with inputs and outputs. Some experiments are usually performed, together with a statistical research, analysis and validation of the model.

Experimental investigations are run through a few steps, namely:

 Preliminary design of the experiment. The variables are selected, as well as their level of investigation. It is necessary to settle the parameters to be measured, their characteristics, measurement units, format, etc.;


Fig. 5. E design – e research platform (problem definition, research planning, presentation of

After defining the precise research topic it is important to determine the types of research to be conducted. Each research concludes with a model. There are several types of models:

Mathematical models (analytic). In this case a formal theory is developed, and the

 Experimental models. These models require an experimental design observation of the phenomenon, data collection, data processing, and the validation of the model. There are three types of experimental research based on the level of understanding of the

 *scientific models*: The phenomenon is carefully observed and the model proposes a theory to explain it. Measurements and analyses will be carried out to validate the

 *engineering models*; Existing solutions are being studied, measurements are being undertaken and improvements of these solutions are proposed and implemented.

 *empirical models;* In this case there is no need for a theory or an explanation for the phenomenon. The model works as a black box with inputs and outputs. Some experiments are usually performed, together with a statistical research, analysis

 Preliminary design of the experiment. The variables are selected, as well as their level of investigation. It is necessary to settle the parameters to be measured, their

These activities are repeated until no more optimisations are required;

**3.4.2 Determining the type of research that needs to be done** 

results of this theory can be compared against empirical values;

the up to date information)

model:

model;

and validation of the model.

Experimental investigations are run through a few steps, namely:

characteristics, measurement units, format, etc.;

	- The model reflects the system target in an acceptable way?
	- For input data, does the simulation provide the correct output data?
	- Are there statistical methods that allow the experimental verification of the model? The mostly used method is ANOVA.

To solve this issue, the chosen option was to build a representative model that describes all the manufacturing possibilities of the drilling heads. This model consists of an imaginary body with several steps, which can do several movements and has different adjustment possibilities. It also possesses several active edges, all with different shapes and positions.

Through customization, one can get different boring heads with various functions. The need of the representative tool is not only aimed towards a particular solution, but also towards synthesising the knowledge in this area and completing the network database.
