Risk-Based ROI, Capital Budgeting, and Portfolio Optimization in the Department of Defense https://www.dau.edu
Figure 1 illustrates the Optimization Results , which returns the results from the portfolio optimization analysis. The main results are provided in the data grid, showing the fnal Objective Function results, fnal Optimized Constraints , and the allocation, selection, or optimization across all individual options or projects within this optimized portfolio. For instance, with the 10 independent projects (each with its own value metrics, cost, and risk parameters), we can see that if the budget is set at $2.5 million, projects 3, 7, 9, and 10 would be selected in the portfolio, constituting the best combination possible given the budgetary constraint (of course, additional constraints can be added as required). If additional funds are now available, such that the budget is $3.5 million, the program can now aford to add project 5 to the portfolio, and so forth. The top left portion of the screen shows the textual details and results of the optimization algorithms applied, and the chart illustrates the fnal objective function (the y-axis is the objective, which, in this case, is to be maximized; whereas the x-axis is the budgetary constraint, with a graduated step of $2.5 million, $3.5 million, $4.5 million, and $5.5 million). The chart shows the investment efcient frontier curve. Figures 1 and 2 are critical results for decision makers as they allow them fexibility in designing their own portfolio of options. For instance, Figure 1 shows an efcient frontier of portfolios, where each of the points along the curve is an optimized portfolio subject to a certain set of constraints. In this example, the constraints were the number of options that can be selected in a ship and the total cost of obtaining these options, which is subject to a budget constraint. The colored columns on the right in Figure 1 show the various combinations of budget limits and maximum number of options allowed. For instance, if a program ofce in the Navy only allocates $2.5 million (see the Frontier Variable located on the second row) and no more than four options per ship, then only options 3, 7, 9, and 10 are feasible; and this portfolio combination would generate the biggest bang for the buck while simultaneously satisfying the budgetary and number-of-options constraints. If the constraints were relaxed to, say, fve options and a $3.5 million budget, then option 5 is added to the mix. Finally, at $4.5 million and nomore than seven options per ship, options 1 and 2 should be added to the mix. Interestingly, even with a higher budget of $5.5 million, the same portfolio of options is selected. In fact, the Optimized Constraint 2 shows that only $4.1 million is used. Therefore, as a decision-making tool for the budget-setting ofcials, the maximum budget that should be set for this portfolio of options should be $4.1 million. Similarly, the decision maker can move backwards, where, say, if the original budget of $4.5 million
74
Defense ARJ, January 2020, Vol. 27No. 1 : 60-107
Made with FlippingBook Learn more on our blog