Managing Uncertainty in the Application of Composite Sustainability Indicators to Transit Analysis

Abstract Composite Sustainability Indices (CSI) present an opportunity to understand the holistic sustainability performance of a transit system through rigorous analysis. Each CSI is based on processing data through a set of indicators, normalizing the indicators, and applying weightings to create a weighted sum index. While CSIs are useful for understanding system performance, indicators may be unreliable due to the types of uncertainty included in their development. This paper proposes that there are four key types of uncertainty embedded in CI development – indicator, weighting, normalization, and data uncertainty – and presents techniques to mitigate and manage them. First, a critical literature review of CSI and indicator development is presented. Second, a new framework for managing CSI uncertainty based on these four types is proposed. Third, a case study analysis demonstrating the use of this framework, including Monte Carlo simulation and multiple normalization techniques, is presented. Next, key conclusions from the case study are shared. Finally, the paper recommends future directions for the application of the framework and continued development of CSI techniques.