Document Type

Thesis - University Access Only

Award Date

2005

Degree Name

Master of Science (MS)

Department / School

Electrical Engineering and Computer Science

Abstract

A problem with extending the usefulness of spatial data currently be collected by farmers and researchers Is that most of the spatial data Is collected at different scales and different formats. Few farmers have the complete skills to make use of data from different scales and formats. Most of the data sets also contain Incorrect records that need to be removed prior to analysis. Producers are not Interested In data per se but want to use the data to enhance their management decision making process too. To accomplish this, producers need tools that take their mass of collected data and transform It Into more usable Information. To solve this problem an Expert System (ES) can be a powerful tool that Incorporates the expertise of agricultural scientists and the farmers archived data to Improve field management. A Decision Support System (DSS) needs to be developed as Graphical User Interface (GUI) to allow a user to clean a data set, select a target area, display selected spatial data and analyze multiple years of yield data. Field Analyst 1.1 Is decision aid software that allows the user to Integrate a variety of data sources with different scales Into a common format and develop a data set that can be used by Management Zone Analyst 1.0.1 (Kitchen et al.) to develop management zones. Sequential decision models play an important role in expert system optimization if the system has noisy inputs that greatly affect decisions. Many expert systems in agriculture, engineering, business, and medicine have benefited from sequential decision technology. The noisy data collected by a yield monitor results from calibration or operation errors create anomalies that greatly affect the combine documented yield records. The focus of this research is to introduce a sequential decision model to minimize yield monitor error in a rule based expert system accomplished by sequential decisions for purposes of increasing agronomic management intensity.

Library of Congress Subject Headings

Farm management -- Computer programs
Crop yields -- Measurement -- Data processing
Expert systems (Computer science)
Decision support systems

Format

application/pdf

Number of Pages

48

Publisher

South Dakota State University

Download

Share

COinS