Project Selection with Multiple Performance Objectives

Abstract

Electric utilities are on a never ending quest to attain higher levels of performance for increasingly lower costs. Often times this leads to project requests that far exceed budget and resource constraints. Many utilities have started to manage this problem through well-defined project evaluation and selection processes. At a minimum, these processes are able to rank project proposals within a given category with respect to expected cost and expected benefit. More mature systems are able to: trade-off capital, operations, inspection, and maintenance; look at marginal project value; trade-off risk versus expected performance; and manage performance over multiple years. The most common project selection approach is to rank all projects based on the ratio of benefit to cost. By forcing all projects to be assigned a benefit and a cost, projects across departments and functions can be directly compared. By ranking all projects based on the ratio of benefit to cost, projects can be selected in order until budgets are exhausted. This presentation suggests a new approach to project ranking that is designed for multiple performance targets. This allows a utility to identify a large number of benefit measures and to set performance goals related to each measure. Once metrics and targets are identified, the methodology identifies a project portfolio that achieves all performance targets for the least possible cost. This methodology has been implemented in an easy-to-use tool called AMPS (asset management project selection), which allows scenarios and sensitivities to be quickly explored