数据模型与决策(运筹学)课后习题和案例答案017s.docVIP

CD17s-PAGE 2 CD17s-PAGE 19 CD Supplement to Chapter 17More About the Simplex Method Review Questions 17s.1-1 No. 17s.1-2 The adjacent corner points that are better than the current corner point are candidates to be the next one. 17s.1-3 The best adjacent corner point criteria and best rate of improvement criteria. 17s.1-4 The simplex method starts by selecting some corner point as the initial corner point. 17s.1-5 If none of the adjacent corner points are better (as measured by the value of the objective function) than the current corner point, then the current corner point is an optimal solution. 17s.1-6 Choose the best adjacent corner point. 17s.1-7 Choose the next corner point by picking the adjacent corner point has the lowest objective function value rather than highest when getting started. There may only be one adjacent corner point if the feasible region is unbounded. 17s.2-1 It is analagous to standing in the middle of a room and looking toward one corner where two walls and the floor meet. 17s.2-2 There are three (at most) adjacent corner points. 17s.2-3 Yes. 17s.2-4 With three decision variables, the constraint boundaries are planes. 17s.2-5 A system of n variables and n equations must be solved. 17s.3-1 The name derives from the fact that the slack variable for a ≤ constraint represents the slack (gap) between the two sides of the inequality. 17s.3-2 A nonnegative slack variable implies that the left-hand side is not larger than the right-hand side. 17s.3-3 For the Wyndor problem, the slack variables represent unused production times in the various plants. 17s.3-4 It is much simpler for an algebraic procedure to deal with equations than with inequalities. 17s.3-5 A nonbasic variable has a value of zero. 17s.3-6 A basic feasible solution is simply a corner point that has been augmented by including the values of the slack variables. 17s.3-7 A surplus variable gives the amount by which the left-hand side of a ≥ constraint exceeds the right-hand