Minnesota's Nutrition Coordinating Center Uses Mathematical Optimization to Estimate Food Nutrient Values

We developed, evaluated, and deployed computer software that uses mathematical optimization rather than trial-and-error methods to estimate the nutrient content of commercial food products. Linear programming (LP) and quadratic programming (QP) optimization methods were evaluated by comparing nutrient values estimated for 31 commercial products with known values for these nutrients. The optimization software was able to estimate nutrient values four times faster than conventional methods with the same degree of accuracy, although the QP method was slightly slower than LP. Nutritionists found the software easy to learn and use. The Nutrition Coordinating Center at the University of Minnesota now uses the LP version of the software as a decision support tool to maintain a food composition database.

[1]  J M Holden,et al.  Carotenoid content of fruits and vegetables: an evaluation of analytic data. , 1993, Journal of the American Dietetic Association.

[2]  Richard J. Hanson,et al.  Algorithm 587: Two Algorithms for the Linearly Constrained Least Squares Problem , 1982, TOMS.

[3]  Hoover Lw,et al.  Calculating the nutrient composition of recipes with computers. , 1989 .

[4]  E. W. Murphy,et al.  Comparisons of methods for calculating retention of nutrients in cooked foods. , 1975, Journal of agricultural and food chemistry.

[5]  D. Haytowitz,et al.  Dietary intake methodology. II. USDA's Nutrient Data Base for Nationwide Dietary Intake Surveys. , 1990, The Journal of nutrition.

[6]  D. Haytowitz,et al.  Estimating nutrient values of mixed dishes from label information : method accurately calculates the nutrient value of a food without requiring complete analytical data , 1993 .

[7]  Robert P. O'Hara,et al.  Modern Programming Using Rexx , 1985 .

[8]  Brian J. Westrich,et al.  Accuracy and Efficiency of Estimating Nutrient Values in Commercial Food Products Using Mathematical Optimization , 1994 .

[9]  Ian Barrodale,et al.  Algorithm 552: Solution of the Constrained I1 Linear Approximation Problem [F4] , 1980, TOMS.

[10]  L E Cleveland,et al.  Method for identifying differences between existing food intake patterns and patterns that meet nutrition recommendations. , 1993, Journal of the American Dietetic Association.

[11]  Linus E Schrage.  User's manual for LINDO , 1981 .

[12]  J. Douglas Faires,et al.  Numerical Analysis , 1981 .