Embodiments of the invention provide a method of controlling a gas-phase polymerization process. The method includes determining a difference between a control variable of the polymerization process, such as the production rate, and the desired value of the control variable; adjusting or maintaining a first manipulated variable to at least partially compensate for the difference between the control variable and the desired value; and adjusting or maintaining a second manipulated variable to at least partially compensate for the effect of adjusting or maintaining the first manipulated variable. The first and second manipulated variables can include process variables such as the fluidized bed weight, the catalyst concentration, the concentration of one or more monomers, the flow of one or more comonomers, the ratio of one comonomer to another comonomer, the activator concentration, the ratio of an activator to selectivity control agent, the concentration of a chain transfer agent, and the retardant concentration.