This paper reviews one chemical company's experience in conducting Major Incident Prevention Program studies over the past 10 years in 90 bulk storage and processing units. The value and limitations of dispersion modeling, recurring findings from good practices and emergency response reviews, and dominant strategies,for consequence and frequency reduction are outlined. The challenge and benefits posed by the use of quantitative criteria in evaluating and managing residual risk of ongoing operations are also discussed. (C) 2004 American Institute of Chemical Engineers Process.