Risk-based, multistage stochastic energy project selection

Energy conservation through the implementation of energy efficient retrofit projects can be viewed as a series of investments with annual returns. These returns can be used to fund additional projects. However, planning for energy conservation ignoring the fluctuations in energy costs and uncertainty in the estimated savings severely impacts project selection and initial budget requests. These impacts drive returns and influence the ability to implement future projects. This paper demonstrates from the Agency perspective, how a risk-based, stochastic multi-period model with binary decision variables at each stage provides a much more accurate estimate for planning than traditional and deterministic models. This approach accounts for uncertainties while determining the proper budget request that minimizes risk of the worst outcomes. The practical application shows that agencies can adjust their risk appetites and make more cost-effective selections while considering the energy saving uncertainties. The application of stochastic optimization with the inclusion of risk to an important energy conservation problem makes the proposed model novel. Finally, while most portfolio selection and optimization problems seek to choose a proper mix of securities or projects, all must be selected here, making the timing the key selection criteria and further differentiates this proposed method.