It is reasonably common for mineral processing plants to perform poorly compared to the expectations of the people who designed and built them. However, such failures are seldom discussed in public. This paper argues using a mythical example that such failures are often due to inadequate consideration of the likely amount of variation in the material which will flow through the plant. A suggested way of considering the likely amount of variation is given in the form of a few key questions with some ideas about how these questions might be answered in practice. Crown Copyright (C) 2003 Published by Elsevier Science Ltd. All rights reserved.