In this paper we have posited an alternative methodology for the least-cost diet selection problem subject to a given set of nutritional constraints. In order to better address the implicit compliance problem, we have re-cast the classical diet selection problem using a two-stage approach whereby in the fi rst stage the dietician decides on the total number of least-cost combinations which satisfy the nutritional requirements and in the second stage, the dietician finds the optimal number of diet plans that have to be formulated. The second stage basically involves finding a reduced search space n* out of which the dietician will sequentially select one least-cost diet plan after another till a "match" is secured with the patient type to ensure compliance, which we have argued based on prior literature to be needing more than simply finding the least-cost diet plan. Our posited approach addresses the compliance problem by making it endogenous to the optimal diet selection problem via consideration of different metabolic types.
Available at: http://works.bepress.com/kuldeep_kumar/51/