You are here

DESIGN OF EXCAVATION OPERATIONS USING EXPERT SYSTEMS

Primary tabs

Nader M. GHAZAL

 

Univ.

University of Cincinnati

Spec.

Civil Engineering/ Structural

Deg.

Year

# Pages

Ph.D.

1991

204

 

 

The undertaken research studies the various elements required for designing excavation operations and selecting the optimum alternative to complete the project with the required quality at the least cost and allotted time. The thesis portrays an effort to define the essential elements for managing excavation operations and focuses on developing an expert system called Excavation Operations adviser, “EXCOP ”.

EXCOP is capable of advising the inexperienced engineer on issues relating to methods of excavation, production requirements, fleet selection criteria, means for fleet procurement, and methods for financing.

In this thesis, the domain of expert systems is examined, the state‑of‑the‑art of their application in the construction industry is investigated, and their future use for solving problems relating to the construction industry is assessed. The currently available knowledge acquisition on techniques are discussed in detail and the knowledge elicitation process for EXCOP is elaborately explained,

The required knowledge for building the knowledge base of EXCOP was elicited from literature, manufacturers' manuals, field survey, and interviews with experts in the area.

EXCOP is expected to be useful where human experts are scarce and where the novice engineer may be in a remote area or away from the home office. EXCOP introduces experts in excavation operations to new methods for solving problems. Also, it has a great potential to be used as a tutor for students in academic institutions.

Further research and development of expert systems in the construction industry are required to cover and integrate other areas of the industry. Expert system empty shell developers are encouraged to look into better means for simulating the human experts method for handling uncertainty. Better interface between expert systems and traditional programs are needed for the enrichment of both types.