Power take-off entanglement risk factor analysis for grain augers
Abstract
This study was completed to develop an expert system model to assist PTO driven grain auger manufacturers in identifying the risk factors associated with their equipment and to educate farmers on their risk to be involved in a PTO entanglement. The development of an expert system is a five step process which includes: (1) Identification of the problem, (2) Conceptualization of the problem, (3) Formalization of the knowledge, (4) Implementation, and (5) Testing of the model. A review of literature and the summarization of a data base containing 159 PTO entanglement investigations that have been completed by Purdue researchers since 1982 identified grain augers as being most commonly involved in PTO entanglements. A summarization of 53 auger and elevator PTO entanglements from the Purdue data base was completed to identify the risk factors associated with the use of this equipment. These risk factors were used in developing the computer model. A method was developed and utilized for collecting grain auger operator exposure measurements and the documentation of auger operator work patterns. Information obtained from 23 on-site observations was used in the computer model. Formalization of the knowledge involved refining the risk factors and utilizing a panel of experts to assign values to them. These risk factors were included in the model. The implementation of the knowledge involved the use of an IBM compatible computer and software. Knowledge rules and risk factors were placed into computer code that would allow the model to mimic the power take-off entanglement risk factor analysis by an expert. The program allows a manufacturer to evaluate the hazard level of a grain auger and risk factors of auger operators. In addition, farmers can use the program to evaluate their personal risk to be involved in a PTO entanglement on a grain auger. A trial validation, or testing, of the model was completed to determine if the computer output matched that of an expert. The testing of the model showed that refinement of the knowledge in the model is needed so the computer model output will match the opinion of the experts.
Degree
Ph.D.
Advisors
Field, Purdue University.
Subject Area
Agricultural engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.