Discusses the relationships between finite schedulers which operate on output from a higher level in a conventional MRPII system, and so-called stand-alone schedulers which essentially operate directly on orders. Presents two graphical models – a demand flow model and a flowpipe model – which help to explain the differences between such schedulers and are used to illustrate the key features of each. Coupled schedulers tend to work on demand which has already been smoothed or planned in some way and can be described as planning-led, whereas stand-alone schedulers work from unfiltered data, with the operator reacting to schedule quality (due-date performance, resource activation levels, etc.) and thence adjusting resource and rerunning. Ascribes the growth of the stand-alone scheduler, at least in part, to the greatly increased computing power available quite inexpensively on modern equipment.