*2.6.1.3 Plackett-Burman designs or Hadamard designs*

This design is special two-level full factorial design and generally employed for the screening of factors. Plackett-Burman designs are mainly applicable for screening a large number of factors if we want to test the effect of 7 factors then we have to put some dummy factors. The results of full factorial designs, Plackett-Burman design, and Taguchi design are interpreted by using a half-normal plot and Pareto chart. By using these designs we can detect all prime effects economically and all other interactions assumed as negligible when compared with few prime effects [46–48].

#### *2.6.1.4 Response surface methodology (RSM)*

Response surface methodology is used after the identification of the critical variables affecting a response. Response surface methods such as central composite design, Box–Behnken design, and three-level factorial designs can recognize the optimum/suitable processing parameters or conditions [49, 50]. The primary advantages of response surface methodology are as hitting a target, minimizing or maximizing a response, minimizing variations, setting a robust process, and finding multiple objectives.

#### *2.6.1.5 Central composite designs or Box-Wilson design*

This is one of the most commonly employed optimization design because this is used for 5 levels of each loaded factor with a less number of runs required when compared with 3 levels full factorial designs. Central composite designs are


**Table 4.**

*Fractional factorial design of a full factorial design, prime results can be estimated.*

generally used for nonlinear responses. In this model, a two-factor central composite design is similar to a 32 factorial design by using the experimental domain at α = ±1. Dash RN et al. successfully developed a glipizide-loaded formulation by using central composite designs [51].
