The allowable biological catch (ABC) in Japanese fishery management is currently determined by applying harvest control rules, which are categorized into two types depending on whether or not stock-size information is available. We evaluate management procedures (MPs) in data-poor situations using an operating model (OM) based on a production model. The OM incorporates uncertainties regarding its assumptions and the process and observation errors of population dynamics and fishing processes, while the values of the MP parameters are determined to avoid stock collapses and low catches. We evaluated the MP formulation ABC = δ×C_t ×(1+k(b/I)), where δ is a coefficient dependent on the stock status level on the stock size, C_t is the catch in year t, k is the weight coefficient, b reflects the trend in stock abundance index over time, and I is the mean of the stock abundance index I. This study shows that smaller values of k reduce the frequency of substantially low catches, particularly when there is significant uncertainty surrounding the stock status. In addition, the value of δ affects both the frequency of fishery collapse as well as stock and catch sizes. We conclude that more reliable stock abundance indices are necessary if the stock size and catch are to stabilize and MPs become more robust to uncertainties.