Building Groundwater Optimizers That Respond to the Needs of Decision-Makersby David E. Dougherty, Univ of Vermont, Burlington, United States,
Abstract: Optimization methods are being sought and employed to design systems that remediate contaminated groundwater regions (and, in fact, the subsurface in general) with increasing frequency. This leads to a small but very active group of researchers creating, understanding, developing, and eventually testing novel schemes for optimization. Because many of the design problems involve nonlinear biological, chemical, or physical processes, this research is intrinsically quite challenging. Another aspect of the challenge in groundwater remediation design is the use of such optimization and the requirements placed upon them by users and their problems, rather than the mathematical and computational problems that attract and sustain so many academic researchers (myself included). If the 'applicable optimizers' that researchers build are in fact going to be applied and have a real impact, then these computer codes must be designed and constructed for decision-makers and not just for 'academic types'. In this paper, I describe research that is sensitive to, and responsive to, the questions decision-makers have asked. I also describe the difficulty of determining what decision-makers want, how apparently arbitrary but nonetheless real nontechnical constraints can impact optimization research, and the challenge of assimilating natural language descriptions of objectives and constraints in realistic situations. Examples from current research illustrate these points.
Subject Headings: Decision making | Groundwater pollution | Groundwater | Optimization models | Remediation | Biological processes | Chemical processes | Computing in civil engineering
Services: Buy this book/Buy this article
Return to search