USDA Research
The CRIS records for the two USDA-supported research projects which serve as the foundation for DairySight are duplicated below, for reference purposes.
ACCESSION NO: 0175041 SUBFILE: CRIS
PROJ NO: NYK-9700136 AGENCY: SBIR NY.K
PROJ TYPE: SMALL BUSINESS GRANT PROJ. STATUS: TERMINATED
CONTRACT/GRANT/AGREEMENT NO: 97-33610-4014
START: 15 MAY 1997 TERM: 30 NOV 1997 GRANT YR: 1997
INVESTIGATOR: Ehrlich, J. L.
PERFORMING INSTITUTION:
DAIRY VETERINARIANS GROUP
RD 1, BOX 1162
ARGYLE, NEW YORK 12809
"INTELLIGENT REPORTS" FOR DAIRY HERD MANAGEMENT, BUILT TO ORDER ON THE INTERNET
OBJECTIVES: The rapid development of the Internet, combined with the increasing power of microcomputers, provides an opportunity to improve greatly the quality of information available to the managers of dairy farms. Internet technology provides a framework upon which information can be distributed quickly, with great flexibility, and at low cost, but existing information tools for dairy management are largely optimized for a system built around mainframe computers and character-based paper records. The multi-platform browsers that have fueled growth of the World-Wide-Web (WWW) offer tremendous new flexibility in the way data may be displayed. Increasing computer power alwo makes it possible to add embedded intelligence and measures of confidence to reports. Thoughtful graphic design can make this increased sophistication unobtrusive, and accessable to dairy farmers and the consultants that serve them.
APPROACH: See objectives.
PROGRESS: 1997/05 TO 1997/11
A prototype database and WWW report server, with some experimental designs for dairy reports, is at http://www.dairydata.com. Modules import data from the DHIA Standard Transfer Format, and other data sources, and store it in a relational database from which SQL queries can retrieve data on demand. A report generator produces reports, and delivers them via HTTP to a user's WWW browser. For example, a Pregnant Cows List uses a compound SQL query to retrieve a list of the pregnant animals in a particular herd and list them in a printable HTML table. A Herd Check List shows animals to be submitted for rectal palpation, based on parameters chosen by the user. The Relative Herd Summary report shows cross-herd comparison capabilities. The user specifies a herd and date of interest, and a comparison group (for example, all herds in the same county or state). The report calculates summary statistics for various measures, and plots mean and SD for the comparison group along with the value for the herd of interest. Statistics are generated on-the-fly, and presented graphically. We are developing a method for projecting milk production, based on a re-design of Wood's curve, which will be used within reports for evaluating changes in herd and individual milk production. The Lactation Projection Demo page includes an embedded Java applet that accepts user-input data points (DIM, milk), plots them, calculates and draws a lactation curve to fit the points, and predicts total 305-day production. These, and other reports, are undergoing field testing by cooperating practitioners.
PUBLICATIONS: 1997/05 TO 1997/11
No publications reported this period
ACCESSION NO: 0180989 SUBFILE: CRIS
PROJ NO: NYK-9803212 AGENCY: SBIR NY.K
PROJ TYPE: SMALL BUSINESS GRANT PROJ. STATUS: NEW
CONTRACT/GRANT/AGREEMENT NO: 98-33610-6608
START: 01 SEP 1998 TERM: 31 AUG 2000 GRANT YR: 1998
INVESTIGATOR: Ehrlich, J.
PERFORMING INSTITUTION:
DAIRY VETERINARIANS GROUP
832 COOT HILL ROAD
ARGYLE, NEW YORK 12809
INTELLIGENT REPORTS FOR DAIRY HERD MANAGEMENT, BUILT TO ORDER ON THE INTERNET
OBJECTIVES: 9803212. We will develop a series of Internet-based information tools for use in management of dairy herds. Reports will be generated and delivered on demand to dairymen, consultants, and researchers, using standard Internet protocols. Data will be assembled from DHIA and other data sources, processed according to needs (observing privacy restraints), then formatted for delivery via the WWW. Specialized report designs may include embedded intelligence and expert knowledge in various ways. Simpler designs may mimic existing reports, but will have advantages over paper counterparts in immediacy and ad hoc configurability. In an "Internet application," we will join the flexibility and low cost of microcomputer applications to the powerful databases of the DHIA system. Adding new functionality to these legacy databases will enhance their value to the dairy industry, and encourage continued participation by dairy farms. Interconnecting systems will permit analysis, research, and benchmarking covering much of the US dairy industry. These technical objectives can best be reached by establishing a cooperative business structure where experts in various dairy specialties may develop designs, and assess royalties for the use of their intellectual property, if they so choose.
APPROACH: Tasks can be divided between those necessary to establish the sub-network (within the Internet) for generation and delivery of reports, and the design of reports themselves. Both must be developed to achieve commercial sustainability. The network implementation will use an n-tier Internet structure built around a middle-tier application server. Connecting to data sources will require development of a meta-data repository and use of various techniques for connecting to cooperating systems. Connection to clients will be through standard WWW browser software. The application server will identify users through logon passwords, and calculate charges to users, and credits to suppliers of data and intellectual property. Simple reports will include typical action lists (such as cows due to be checked for pregnancy or drying off) as well as standard summary reports, and may be calculated according to user-input parameters. Printing can be done through standard browser print functions. "Intelligent Reports" will assist as experienced user to investigate complicated interactions within a dairy herd. They may involve fitting data to a model and displaying the result, but leaving much synthesis and interpretation to the person. The emphasis will be on exploration, but (where possible) reports will also include measures of confidence.
Return to DairySight
Home