Risk-Based ROI, Capital Budgeting, and Portfolio Optimization in the Department of Defense https://www.dau.edu
The author recommends that any follow-on research incorporate the following items: • Apply the methods to actual programs with real-life data and assumptions, with SME or subject matter estimates. • Create new or evaluate existing concepts of military value. These will incorporate: ° Data validity tests using applied statistical tests (from basic linear and nonlinear correlations to econometric models and nonparametric hypothesis tests). These are applied over time to identify if the collected data are valid and actually describewhat the researcher wants or expects the data to describe. In other words, are the collected data valid, accurate, and precise? ° Big data analysis—trying to fnd patterns and analytical relationships in large data sets. ° Historical data to perform backcasting (backtesting historical data to known historical events). ° Tweaking and creating lighthouse events and programs in the past, assigning critical value metrics to these events and programs, and using these as guideposts for generating future SME estimates. ° Creating more exact definitions and methods for SME assumptions that allow for collecting amore objective and defensible data set. • Utilize multiobjective optimization. Interdependencies and competing stakeholder needs (e.g., Congress versus Ofce of the Secretary of Defense and other external stakeholders) need to be considered. These competing objectives need to be reconciled to determine a Pareto optimal portfolio. • Evaluate analytical hierarchical processes, multiobjective optimization, and other algorithms and compare the results. • Within the portfolio, model and account for risks of cost and budget overruns as well as delivery delays using risk-based simulations.
98
Defense ARJ, January 2020, Vol. 27No. 1 : 60-107
Made with FlippingBook Learn more on our blog