Development and evaluation of an expert system approach to uneven-aged management of loblolly-shortleaf pine stands in the west Gulf region.

摘要

Forest managers expend significant time and effort seeking, organizing, and synthesizing information relevant to making effective forestry decisions. Oftentimes, they must rely on the knowledge and experience of human experts, a resource that is in short supply, requires many years to acquire, and is concentrated in a few individuals. This research task suggests expert systems as one viable solution to the problems of technology transfer and automating and maintaining expertise in consistent and usable form.; Expert systems are practical computer programs which solve problems that were previously considered only solvable by human expertise. The expert system developed in this research, named FOREX, was written in ProLog. FOREX is primarily a second-generation expert system for prescribing silvicultural systems. Aside from human expertise stored in its knowledge base, FOREX is linked with growth and yield and optimization models to complement the search for optimal recommendation.; A methodology was developed for transforming available literature/research information and the private knowledge of human experts into decision rules. Factors pertinent to prescribing silvicultural systems were identified. English-like decision rules were developed, and human experts were then asked to verify and confirm these rules. The process of encoding these rules into ProLog format was an important phase of the development process.; In a modified Turing test, nine human evaluators rated prescriptions from four other human experts, FOREX, and another computer-based model. FOREX's scores were found comparable to the research foresters and superior to the industrial foresters and the other computer model. These results indicate that human expertise, in uneven-aged management of loblolly-shortleaf pine stands, has been captured by an expert system. Success in this project should encourage other researchers to apply this approach for other forestry problems.