Generating a robust model for production and inventory control

In tills stud}', we generate a production and inventory control model which gives h'obust‘ solutions against demand estimation errors. This model is applied to a real production and inventory system; howe\’er, it is a general model where the demand rate is stochastic with a known probability distribution and other parameters of the system are constant. The proposed model is a bi-objective chicision making model, with two decision variables. .A ‘compromised* solution is found for the problem using the trade-off curve generated by a constrained sequential optimization technique, applied on a nonlinear programming model parametrically. Robustness against parameter estimation errors is tested by sensitivity analysis. Here a new dimension is added to sensitivity analysis methodology by including a sensitivity measure as a ‘cost of error* of parameter estimation. By so doing, the proposed model is tested against the classical EOQ model and it is shown that the proposed model ])erforms far better.