Lot quality assurance sampling (LQAS) is a widely used method in manufacturing and health care. LQAS is a form of acceptance sampling, which involves judging whether a batch of items is of acceptable quality based on inspection of a smaller allotment. LQAS makes no assumptions about prior beliefs in quality, and assumes that the quality of successive batches is independent, however these assumptions are unlikely to hold in many cases. Exponentially weighted moving average (EWMA) models address the latter but not the former assumption. EWMA is another often used method for the problem of LQAS. Bayesian state space models (B-SSMs) can incorporate both assumptions, and thereby potentially improve acceptance sampling decision making, thereby reducing waste and saving costs. A simulation study was done to investigate whether B-SSMs outperform LQAS and EWMA, in a factorial design where the strength of dependency between batches and prevalence of poor quality batches was varied. The results showed that B-SSMs generally outperformed LQAS and EWMA in terms of statistical accuracy, especially when dependency was strong and prevalence was high.