[abstract based on the first two paragraphs] In our studies of military systems, we typically look at the relationship between resources expended and expected results, the latter often given in terms of a measure of effectiveness (MOE). Frequently we have, as in Figure 1,
what Gen. Jasper Welch (Welch, 1977) referred to as a Pay More-Do Better curve. Very often, as in Figure 1, inspection of the curve leads us to select a point (x, y) where the curve's direction is changing rapidly from large slopes to small slopes. We
say that the curve has a knee at (x, y), or at least in its vicinity. Moreover, it is often inferred that x is a good limit on the resources worth expending on the given system. To the right of x lies the region of
diminishing returnswhat Alain Enthoven used to call the flat of the curve (Tucker, 1966). Rather than expend more on the given system (i.e., where x > x), it may be more rewarding to devote resources to other uses. In a general sense the above inference has
much merit, and the knee of a curve may be a useful clue, but it is incomplete in itself. We need to embed the given resource expenditure problem in a larger resource allocation problem, in order to compare returns as shown in Figure 1 with the returns available from other uses of resources.
There is nothing new in this observationit has been a generation since Charles Hitch wrote his eloquent words on suboptimizing (Hitch, 1953) and many more generations since economists first discoursed on diminishing returns.
Military Operations Research will no longer be hosted by ingentaconnect as of March 1, 2015. Subscriptions and access to issues for subscribers will be available through the MORS website Journal page here: http://www.mors.org/Publications/MOR-Journal For assistance please contact Liz Marriott email@example.com or 703-933-9071.