Parasol: an open source, interactive parallel coordinates library for multi-objective decision making

Abstract

This paper introduces Parasol—an open source, interactive visualization library to support the development of web applications for multi-objective decision making. Multi-objective optimization is a popular way to explore competing objectives in environmental management problems. Interactive visualizations allow stakeholders to explore and gain insights about the large, high-dimensional datasets produced by multi-objective optimization. Among visualization methods, parallel coordinates are well-suited for this task. However, current software and open source libraries have limited support for these plots. The Parasol library described in this work provides developers with the building blocks to create sharable, interactive parallel coordinates web applications. Moreover, by incorporating state of the art clutter reduction techniques—such as clustering, linking, brushing, marking, and bundling—Parasol improves upon traditional parallel coordinates visualizations. We demonstrate the benefit of such features through simple examples and by exploring a real-world water resources problem commonly used in multi-objective optimization literature.