Using AHP and Stochastic TOPSIS for Carbon Storage Screening and Ranking with Uncertainty Analysis

Abstract A methodology is presented for screening and ranking potential carbon dioxide (CO2) geological storage sites, based on the classical carbon sequestration performance factors - capacity, injectivity, and containment. Other nontechnical criteria, such as cost and public acceptance, are also considered. The various criteria are not always quantitative, sometimes conflict, and can address different risks. Their relative importance may depend on the context. The Analytic Hierarchy Process (AHP) is used to determine the relative weights of different decision criteria. It has the ability for quantitative and qualitative evaluation of attributes, enabling simple and effective determination of relative weights by comparing attributes in pairs. Next, the intuitive Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is applied to calculate site scores and rank sites by comparing each weighted criterion simultaneously for all sites. Monte Carlo simulation is used for uncertainty analysis of results. Finally, sensitivity analysis is provided to show how the ranking results will change when inputs vary. The workflow was applied to field data from an existing project for carbon capture and storage (CCS) site screening and selection, which had ranked nine sites, using Excel-based score cards and 31 criteria that were chosen based on data availability. After weighting the criteria via AHP, a stochastic version of TOPSIS was used to score and rank the sites, as well as provide an uncertainty analysis. Comparison with the previous unweighted user-dependent approach revealed that the new workflow improved site ranking because the various criteria are assigned weights, and site properties are evaluated based on precise numerical values instead of a range of values. Moreover, the impact of each site property on different aspects of the project (e.g., economics, safety) can be examined. Unlike Excel score cards, where confidence and score are not correlated, uncertainty and sensitivity analyses in the new workflow indicate overall uncertainty of the results. This novel integration of AHP for criteria weighting and TOPSIS for sites scoring improves the accuracy and efficiency of carbon storage sites selection. Introduction of Monte Carlo simulation assists the analysis of site properties uncertainty influence on final ranking results, which increases confidence in the final choice. Sensitivity analysis provides information on how ranking results will change when inputs vary, consequently guiding data collection next steps to reduce uncertainty and risk.