ABSTRACT. Many complex decision problems require value judgments, and it is often useful to build formal models of these value judgments. Several models and assessment procedures exist for this task ranging from simple rating and weighting techniques to sophisticated multiattribute utility models that incorporate tradeoffs and risk attitudes. In this chapter, we argue that the choice of a value model and assessment procedure depends on the decision context, the purpose of the model, the quality of available value judgments, time, and other factors. We propose the notion of a practical value model that is useful for a given decision context and purpose, using simplifications and approximations, when appropriate. We then discuss practical considerations when choosing a value model, for example, when deciding whether a value model is additive over attributes or not. We also consider practical issues when choosing an assessment procedure, for example, when selecting equivalent-cost procedures versus rating and weighting procedures. We discuss several topics related to interacting with decision makers when constructing value models, and we conclude with the description of two generic practical value models that we have found useful in many past applications.