In this paper, we present a new method for dynamic optimization to achieve maximum technical benefit to the insurer using genetic algorithms. The objective of this work is to select, from a database containing forms of reinsurance, pricing models and credit ratings parameters (risk measurement methods, such as Value at Risk (VaR), Conditional Value at Risk (CVaR) or ruin probability) a form of reinsurance, a way of pricing and a solvency parameter to maximize technical benefits of the insurance company.