Design of experiments, results and operational observations

It was decided that four parameters or factors would cover the range of service conditions. Upper dwell time and temperatures were tested at three different values, lower dwell time at two different values and with two different environments. The number of combinations of conditions is therefore 36. A quarter fractional nine trial design necessarily has some confounding and loss of information compared to a full factorial design, but retains the analysis of main effects and some synergies or interactions between factors. 

An experiment, consisting of nine trials, was designed according to statistical criteria. The aim was to ensure that the results of the trials could be analysed statistically, to test the main linear and quadratic effects of upper dwell temperature and hot time and the main effects of lower dwell time (cold time) and environment. 

The design can be written in two ways as a Taguchi style experiment (see, for example, Breyfogle, 2000) as shown, in Table 18.1 and as a modified Graeco Latin square (see, for example, Cochran and Cox, 1957) as shown in Table 18.2. Notice that each row and column in Table 18.2 has equal numbers of 2 h and 4 h lower dwell time and wet and dry environments. Other designs (see, for example Metcalfe, 1994, and Wu and Hamada, 2000), such as central composites and star designs exist but would require more trials to obtain the same balance (see Appendix III). 

Trial Temperature(°C) Upper dwell (h) Lower dwell (h) Environment  

Order Order Order Order Order Order Order Order Order 

---