Hydrogen has high energy content known to date, which produces water vapor due to combustion and is converted to electricity by fuel cells with high efficiency, therefore is considered as a candidate of future energy source. Production of bio-hydrogen through dark fermentation can ferment sustainable renewable substances such as C5 and C6, wastewater and converted waste, compared to photosynthetic methods requires no light, has higher production rate and yields. Fermentative hydrogen production is a complex process and is influenced by several factors, statistical methods of optimization offer relatively understandable interaction among the factors. One of the approach to increase the rate and yield of reaction is to screen, model, and optimize factors which have strong influence on the response comprising inucolum, operating temperature, pH, type and concentration of substrates etc by employing appropriate method. For screening factors and performing regression ANOVA is used. The experimental design method used for screening and understanding the effect of factors are one-factor-at-a-time, full factorial, fractional factorial, Taguchi, Plackett– Burman, central composite and Box– Behnken design, steepest inclined or declined, neural network ANN, genetic algorithms and, for optimization response surface methodology is frequently applied. The overview, presents appropriate advantages and disadvantages of analyzing abilities of modeling of experiments and addresses few statistical methods of optimization to understand complex effect of factors on fermentative dark hydrogen production. Such deep knowldge leads in understanding more in details the interactions of complex reactions involved and to select sound statistical method of optimization, resulting in reasonable reduced number of experiments to achieve higher rate and yield of production at lower cost of operation.