BACKGROUNDManaging herbicide-resistant weeds is becoming increasingly difficult. Here we adapted the weed dynamics model AlomySys to account for experimentally measured fitness costs linked to mutants of target-site resistance to acetyl-coenzyme A carboxylase (ACCase)-inhibiting herbicides in Alopecurus myosuroides. We ran simulations to test how effectively cultural practices manage resistance. RESULTSSimulations of an oilseed rape/winter wheat/winter barley rotation showed that, when replacing one of the seven applied herbicides with an ACCase-inhibiting one, resistant mutants exceeded 1 plant m(-2), with a probability of 40%, after an average of 18 years. This threshold was always exceeded when three or four ACCase-inhibiting herbicides were used, after an average of 8 and 6 years respectively. With reduced herbicide rates or suboptimal spraying conditions, resistance occurred 1-3 years earlier in 50% of simulations. Adding spring pea to the rotation or yearly mouldboard ploughing delayed resistance indefinitely in 90 and 60% of simulations respectively. Ploughing also modified the genetic composition of the resistant population by selecting a previously rare mutant that presented improved pre-emergent growth. The prevalence of the mutations was influenced more by their associated fitness cost or benefit than by the number of ACCase-inhibiting herbicides to which they conferred resistance. CONCLUSIONSimulations allowed us to rank weed management practices and suggest that pleiotropic effects are extremely important for understanding the frequency of herbicide resistance in the population.