Incorporating deeply uncertain factors into the many objective search process. (March 2017)

• Record Type: Journal Article
• Authors: Watson, Abigail A.; Kasprzyk, Joseph R.
• Is Part Of: Environmental modelling & software. Volume 89(2017)
• Page Start: 159
• Page End: 171
• Publication Date: 2017-03
• Subjects: Multi-objective optimization -- Robust decision making -- Scenario analysis -- Robust optimization -- Water resources planning and management -- Water markets; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2016.12.001
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: This paper proposes an approach for including deeply uncertain factors directly into a multi-objective search procedure, to aid in incorporating divergent quantitative scenarios within the model-based decision support process. Specifically, we extend Many Objective Robust Decision Making (MORDM), a framework for finding and evaluating planning solutions under multiple objectives, to include techniques from robust optimization. Traditional MORDM first optimized a problem under a baseline scenario, then evaluated candidate solutions under an ensemble of uncertain conditions, and finally discovered scenarios under which solutions are vulnerable. In this analysis, we perform multiple multi-objective search trials that directly incorporate these discovered scenarios within the search. Through the analysis, we have created multiple problem formulations to show how methodological choices of severe scenarios affect the resulting candidate planning solutions. We demonstrate the approach through a water planning portfolio example in the Lower Rio Grande Valley of Texas. Highlights: A multi-scenario optimization technique for deep uncertainty is proposed. Including severe scenarios within the optimization yields more robust solutions. Parallel coordinates effectively show solution performance in alternate scenarios.