Sensitivity Analysis in Linear Programs
Kenneth R. Baker · 2011
This chapter covers three approaches to sensitivity analysis: the parameter analysis report, the sensitivity report, and the interpretation of optimal patterns. The sensitivity report is the most canned approach. It is a well defined report that complements the use of Solver. Most importantly, it automatically reveals shadow prices and allowable ranges. The parameter analysis report allows the user quite a lot of flexibility, and provides for linear programs the same kind of capability that Excel’s Data Table tool provides for basic spreadsheet models. The recognition of patterns in linear programming solutions is a way of looking beyond the specific numbers in the result and toward a broader economic imperative. By focusing on positive decision variables and binding constraints, this interpretation emphasizes the key factors in the model that drive the form of the solution. The chapter ends with illustrative examples and followed by a set of practice exercises. Controlled Vocabulary Terms linear programming; sensitivity analysis; shadow price