Vinyl Chloride Polymerization is known to be auto accelerating. The reaction rate increase with conversion. Because of these phenomena, substantial reactor productivity at early conversion can be lost because the heat-removal capacity of the reactor is not fully utilized until near the end of the polymerization. For this reason it is desirable to speed up the polymerization at the beginning and slow it down near the end. This rate adjustment can be achieved by running the polymerization hotter in the beginning and then cooling. Optimization methods exploit the repetitive nature of batch processes to adapt the optimal operating policy in the presence of uncertainty. For problems where terminal constraints play a dominant role in optimization, system can be operate close to the optimum simply by satisfying terminal constraints. The Genetic Algorithm optimization scheme is illustrated in simulation for the minimization of the batch time of a suspension polymerization process with terminal constraints on number of average molecular weight, rate of reaction, heat removal capacity. We have written a scientifically based computer model for the polymerization designed specifically to simulate such temperature-programmed reactions. The model has a molecular weight predictor. By using single initiator and a very simple straight-line temperature-programmed reaction, the time to 85% conversion can be reduced 20 percent. This is a substantial increase in productivity. Keywords: Batch reactor, Suspension Polymerization, Dynamic Optimization, GA Optimization, batch to batch optimization.
