In case of incidental confinement failure, mixed oxide (MOX) fuel preparation may expose workers to plutonium aerosols. Due to its potential toxicity, occupational exposure to plutonium compounds should be kept as low as reasonably achievable. To ensure the absence of significant intake of radionuclides, workers at risk of internal contamination are monitored by periodic bioassay planned in a routine monitoring programme. From bioassay results, internal dose may be estimated. However, accurate dose calculation relies on known exposure conditions, which are rarely available when the exposure is demonstrated by routine monitoring only. Therefore, internal dose calculation is subject to uncertainty from unknown exposure conditions and from activity measurement variability. The present study calculates the minimum detectable dose (MDD) for a routine monitoring programme by considering all plausible conditions of exposure and measurement uncertainty. The MDD evaluates the monitoring quality and can be used for optimization. Here, MDDs were calculated for the monitoring of workers preparing MOX fuel. Uncertain parameters were modelled by probability distributions defined according to information provided by experts of routine monitoring, of workplace radiological protection and of bioassay analysis. Results show that the current monitoring is well adapted to potential exposure. A sensitivity study of MDD highlights high dependence on exposure condition modelling. Integrating all expert knowledge is therefore crucial to obtain reliable MDD estimates, stressing the value of a holistic approach to worker monitoring.