To evaluate walleye pollock stock management procedures in the northern waters of the Sea of Japan, 30-year population dynamics, including uncertainties, were forecast. Errors in current stock size estimation, variability in future recruitment and changes in future fishing mortalities were incorporated. Results of virtual population analysis (VPA) from resampled catch-at-age data with bootstrap methods was used as the current stock size estimation with uncertainty. Performances of each scenario were evaluated using conservation, utilization, stability and reliability factors. Twenty-two management scenarios and continuing the current fishing mortality were evaluated. Scenarios with minor regulation changes and continuing the current fishing mortality showed poor stock conservation performances. Scenarios with minor regulation changes produced good short-term but poor long-term utilization. Stabilities were poor in continuing the current fishing mortality and fishing ban scenarios. Reliability in all scenarios after 30 years was smaller than in continuing the current fishing mortality; however, small differences among scenarios were observed. The simulation results indicated that multilateral assessment is needed to evaluate the management candidates. Uncertainty caused by recruitment variability mostly affected future population dynamics. The role of simulations in the production of effective scientific advice is discussed.