Sensitivity Analysis Case Study

5.0 INDUSTRY CASE STUDY, SELECTION OF MULTI-PROJECT USING SENSITIVITY ANALYSIS

The selection of projects from a large list of possibilities is a common problem in Civil and Environmental Engineering. Use of sensitivity analysis and multi-attribute optimization model for project selection problems will be explained in this case study. To demonstrate the model we use the example of selecting a portfolio of projects to compute for funding within the New Zealand Department of Conservation (DoC). The results highlight the sensitivity of project selection to attribute weights to help decision makers examine the robustness of the optimal solution. The below table explains the project selection criteria used in this case study [11].

Figure

Because of this ‘‘fuzziness’’, uncertainty in the parameters must be handled differently than point of view changes. The managers and technical staff with which we were assessing the weights and single-attribute utility functions expressed considerable uncertainty about their assessments, but they were willing to have a range of estimate for their assessments between high and low. No coherent method for conducting a sensitivity analysis in a situation with bounded uncertainty in the objective function parameters of the optimization problem found in the literature. In order to address such uncertainty, a method is present here based on a combination of the ideas. The below figure 6 explains the main 7 steps used in this study

The purpose of this section is to discuss in general some of the concepts behind these display results to explain to decision-maker what it means. Two general methods were used in this study. The first was the grid of two-dimensional plots shown in Figure 7, and the second was the triangular plots used in Figures 8 and 9 [11]. FIGURE 7 Utilities of the eight sets of projects in the DoC example as a function of the weight values

FIGURE 8 Project set selection decision as a function of three of the critical weights.

Figure 8 shows how Set 3 becomes preferred when Weight 2 increases, either by decreasing Weight 3 (moving horizontally right to left), or by decreasing Weight 1 (moving diagonally from the top of the figure to the bottom of the figure along the right-hand side axis) [11].

Figure 9 shows how Project 210 is funded only when Set 3 or Set 4 is preferred. FIGURE 9 Funding decision for an individual project (here project 210) as a function of three of the critical weights. Weights for attributes four and five are fixed at zero.

6.4 STUDY