Conditional simulation block models enable quantification of orebody variability and uncertainty. Even when conditional simulations of mineral resource models exist, many mining engineers are reluctant to use the simulations for mine planning due to commercial software limitations and/or complexities—engineers strongly prefer having a single estimate of grade value for each block. A new mixed integer linear programming (MILP) stochastic production scheduling model, which does not rely on arbitrary penalty functions, has been developed and solved to demonstrate how mine production scheduling will benefit from using a conditionally simulated mineral resource model that properly accounts for selectivity and considers uncertainty associated with block grade estimates. An open pit gold mine case study is used to compare the improvements in production scheduling with Sequential Gaussian Simulation inputs versus less complex alternative inputs including indicator kriging (IK), ordinary kriging (OK), and localized indicator kriging (LIK). Some of the alternative inputs (OK, LIK) provide a single value per block, and some of the inputs (IK, LIK) provide a good match to the global grade-tonnage distribution. Impacts of the alternative geostatistical inputs are shown in terms of differences in recoverable reserves, mine plans, NPV and yearly profiles for cash flow and metal production. In the long-term planning case study presented, the estimators which match the variability of the global grade-tonnage distribution resulted in deterministic production schedules with similar geometry and value to the stochastic production schedules created from conditional simulation inputs. However, the conditional simulation model is still needed to quantify the uncertainty in the production schedules.