The hydrothermal production of chitosan from the carapaces of gray shrimp was carried out, and the obtained material was characterized via X-ray diffraction analysis, infrared spectroscopy, and pH zero-charge point, giving the expected results. Orange G dye adsorption onto synthetized chitosan was investigated in a batch system, the kinetic study was well-described by a nonlinearized pseudo-second-order model, and the equilibrium data indicated that the nonlinear Langmuir form was appropriate to describe the adsorption system with a maximum adsorption capacity of 34.63 mg/g compared with that found experimentally of 31.9 mg/g. The influences of most of the operating parameters, such as pH, adsorbent concentration, temperature, initial dye concentration, and contact time, were studied. These five independent variables acting on the adsorption performance of Orange G were selected for optimization and modeling processes through a central rotating composite design using response surface methodology (RSM). The percentage of removal of Orange G by chitosan prepared from shrimp shells was predicted with a second-degree polynomial equation, and the postulated model was valid and represented well the phenomenon studied in the experimental domain, with an R2 = 0.98 and an RAdj = 0.95. An initial Orange G concentration of 10 mg/L, a pH of 6.5, a chitosan amount of 0.3 g/L, a temperature of 25 °C, and an adsorption time of 450 min were found to be the optimum conditions in batch mode for the maximum uptake of Orange G (removal of 97.43%).