Expert systems for grading hardwood lumber
Abstract
In order to facilitate the training of lumber graders, the lumber industry is eager to apply new methods of instruction for this important profession. Expert systems can serve as such instructional aids. There are six stages in building such expert systems: (1) problem analysis, (2) tool selection, (3) knowledge acquisition, (4) system structure selection, (5) prototype development and testing, and (6) system modification and verification. In the knowledge acquisition stage, in order to ensure the correctness and completeness of the knowledge base, professional graders were interviewed to extract their rules of thumb. According to the characteristics of the problem domain, a rule-based system was considered as the most natural way to represent the knowledge. The language LISP was utilized to implement this system so that an efficient inference engine could be built to control the system. Inside the inference engine a forward reasoning method was used to manipulate the knowledge to reach the final solutions. These systems are run on the VAX 11/780 computer. Finally, this computer-aided instruction tool was employed to assist in training professional lumber graders.
Degree
Ph.D.
Advisors
Sparrow, Purdue University.
Subject Area
Industrial engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.