Use an adaptive method to tune your parameters

see

See http://www.cpdee.ufmg.br/~jramirez/publicacoes/p2001-1.pdf

Probability of mutation and probability of crossover can be linked to genetic diversity as in the paper above.

see

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=875398&url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F7012%2F18933%2F00875398

Which suggests a population size of 16

See http://sarmady.com/siamak/papers/genetic-algorithm.pdf

As a general rule, population size depends on number of genes. So for 9 genes need  16 chromosomes, 16 genes need 32 chromosomes.

I normally start off by choosing population size 1.5-2 times number of genes, to a maximum population size of 100.

You can try having the algorithm self-adapt on the fly the values of those parameters. See for example:

https://www.researchgate.net/publication/277289544_Self_Adaptation_in_Evolutionary_Algorithms

https://www.researchgate.net/publication/274263659_Parameter_Control_in_Evolutionary_Algorithms_Trends_and_Challenges

Article Self Adaptation in Evolutionary Algorithms

Article Parameter Control in Evolutionary Algorithms: Trends and Challenges

The only factors that I consider variable are the crossover and mutation probabilities and the number of individuals (which should be greater than 100). I want to know if I can use classical parameter tuning methods such as DOE, EVOP, RSM etc but in a small number of generations and find the best levels of the parameters and run the algorithm till convergence. Is it a reasonable method?

The recursive least squares algorithm (RLS) allows for (real-time) dynamical application of least squares (LS) regression to a time series of time-stamped continuously acquired data points. As with LS, there may be several correlation equations with the corresponding set of dependent (observed) variables. RLS is the recursive application of the well-known LS regression algorithm, so that each new data point is taken in account to modify (correct) a previous estimate of the parameters from some linear (or linearized) correlation thought to model the observed system. For RLS with forgetting factor (RLS-FF), acquired data is weighted according to its age, with increased weight given to the most recent data. This is often convenient for adaptive control and/or real-time optimization purposes.

Application example ― While investigating adaptive control and energetic optimization of aerobic fermenters, I have applied the RLS-FF algorithm to estimate the parameters from the KLa correlation, used to predict the O2 gas-liquid mass-transfer, while giving increased weight to most recent data. Estimates were improved by imposing sinusoidal disturbance to air flow and agitation speed (manipulated variables). The proposed (adaptive) control algorithm compared favourably with PID. Simulations assessed the effect of numerically generated white Gaussian noise (2-sigma truncated) and of first order delay. This investigation was reported at (MSc Thesis):

Thesis Controlo do Oxigénio Dissolvido em Fermentadores para Minimi...