January 2020
one or more values, rerunning the model, and repeating the entire process until you find a satisfactory solution. The approach is very tedious and time consuming, and it is often not clear how to adjust the values from one iteration to the next. Amore rigorousmethod systematically enumerates all possible alternatives. This approach guarantees optimal solutions if the model is correctly specifed. Suppose that an optimizationmodel depends on only two decision variables. If each variable has 10 possible values, trying each combination requires 100 iterations (10 × 10). If each iteration is very short (e.g., 2 seconds), then the entire process could be done in approximately 3minutes of computer time. However, instead of two decision variables, consider six, then consider that trying all combinations requires 1,000,000 iterations (10 6 alternatives). Complete enumeration can easily take weeks, months, or even years to carry out (Mun, 2015). The timing of course also depends on the complexity of the model, the number of objectives and constraints interacting, as well as exogenous impacts such as a simulation of the input variables (e.g., dynamic and stochastic optimization). Adding more complexity will generally add more computational time. (Mun [2015] shows more detailed explanations of how the computational time is computed and how it can increase exponentially.) To run the analysis, we use the Portfolio Optimization tool in the ROV PEAT (Real Options Valuation Project Economics Analysis Tool) software application (courtesy of http://www.realoptionsvaluation.com). In the Portfolio Optimization section of this tool, the individual projects can be modeled as a portfolio and optimized to determine the best combination of projects for the portfolio. The projects can be modeled as a portfolio and optimized to determine the best combination of projects for the portfolio in the Optimization Settings subtab. Analysts start by selecting the optimization method (Static or DynamicOptimization). Then they select the decision variable type Discrete Binary (choose which Project or Options to execute with a go/no-go binary 1/0 decision) or Continuous Budget Allocation (returns percentage of budget to allocate to each option or project as long as the total portfolio is 100%); select the Objective (Max Net Present Value [NPV], Min Risk, etc.); set up any Constraints (e.g., budget restrictions, number of projects’ restrictions, or create customized restrictions); select the options or projects to optimize/ allocate/choose (default selection is all options ); and when completed, click Run Optimization .
73
Defense ARJ, January 2020, Vol. 27No. 1 : 60-107
Made with FlippingBook Learn more on our blog