program and specific adjustments to them — are 

 preferable. Additional information on methods of 

 dealing with risk and uncertainty can be found in 

 Section 1.4.13(d) of Chapter I. 



(k) One guide to the use of the techniques dis- 

 cussed here is displayed in Table S-2. In general, 

 more complex techniques are appropriate as plan- 

 ning proceeds from the initial development and the 

 screening of alternatives to the analysis and pres- 

 entation of the final set of alternative plans. For ex- 

 ample, sensitivity analysis— testing the sensitivity of 

 the outcome of project evaluation to variation in the 

 magnitude of key parameters— may be most useful 

 and applicable in the early stages of planning, 

 when the concern is to understand single factors or 

 relatively general multiple-factor relationships. Multi- 

 ple-factor sensitivity analysis, in which the joint ef- 

 fects or correlations among underlying parameters 

 are studied in greater depth, may be more appropri- 

 ate in the detailed analytic stage than in the 

 screening stage. 



(I) Similarly, analysis of risk and uncertainty 

 based on objective or subjective probability distribu- 

 tions would be more appropriate in the detailed 

 analytic stage than in the early screening stage. Al- 

 though hydrologic and economic probabilities may 

 be used in the screening stage, the full use of inde- 

 pendent and joint probability distributions, possibly 

 developed from computer simulation methods, to 

 describe expected values and variances, is more 

 appropriately reserved for the detailed stage. 



Table S-2— Planning Task and Approaches to Risk 

 and Uncertainty 



(m) Although decisionmakers' attitudes and deci- 

 sion rules can be used to give perspective on alter- 

 native designs throughout the planning process, 

 they are more appropriate at the stage of displaying 

 alternative designs. 



(n) The differences among the underlying de- 

 grees of risk and uncertainty, the design adapta- 

 tions to them, and the preferences of decision- 

 makers should be kept clear throughout the analy- 

 sis. The first two depend primarily on technical ex- 

 pertise; the last is the set of preferences based on 

 various attitudes toward risk and uncertainty. 



S3 Report and display. . 



The assessment of risk and uncertainty in project 

 evaluation should be reported and displayed in a 

 manner that makes clear to the decisionmaker the 

 types and degrees of risk and uncertainty believed 

 to characterize the benefits and costs of the alter- 

 native plans considered. 



17 



