"Evaluation of Vision Correction Alternatives for Myopic Adults"

by G. Mark Brown in Interfaces, 27:2, Mar-Apr 1997, pp. 66-84.

Decision tree analysis provides the framework the author used on his vision correction strategy. The detailed, very readable article describes an approach suited to many similar problems.

What are the cures for myopia (near-sightedness)? The alternatives considered include:

Radial keratotomy (RK; tiny incisions are made that flatten the curvature of the cornea)
Photorefractive keratectomy (PRK; highly automated laser "sculpting" the cornea)
Eyeglasses
Contact lenses

The objective or value function is a simple additive function involving:

Summing over the decision maker's remaining lifetime:

Corrected visual utility (0-1 scale based upon the Snelling Fraction, e.g., 20/20=1)
Monetary cost of vision correction
Time spent on vision correction (e.g., cleaning contacts)
K1 = unit value trade-off between corrected visual utility and $ cost (Brown assigned a value of $8000 per year to correct from worst to best vision.)
K2 = unit value trade-off between time and cost during year (he used $25/hour)

Because the non-monetary attributes are comparable in dollars, this becomes simple expected (value) cost minimization problem. Of particular interest is how the author framed the problem and elicited his attribute preferences.

Ralph L. Keeney of the University of Souther California contributed his thoughts to the author in discussions.

A wonderful and readable book:
Keeney, Ralph L., 1992, Value-Focused Thinking: A Path to Creative Decisionmaking, Harvard University Press, Cambridge, Massachusetts, 416 p. 90% very readable and good guidelines to formulating decision problems. The 10% mathematical material provides an overview of multi-attribute decision making.

—John Schuyler, May 1997