PRODUCERS, DECISION SUPPORT SYSTEMS, AND GPFARM: LESSONS LEARNED FROM A DECADE OF DEVELOPMENT

Authors: Ascough Ii, James; Dunn, Gale; McMaster, Gregory; Ahuja, Lajpat; Andales, Allan

Submitted to: International Congress on Modeling and Simulation Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 11/10/2005
Publication Date: 12/10/2005
Citation: Ascough Ii, J.C., Dunn, G.H., Mcmaster, G.S., Ahuja, L.R., Andales, A.A. 2005. Producers, decision support systems, and gpfarm: lessons learned from a decade of development. International Congress on Modeling and Simulation Proceedings.

Interpretive Summary: In the Great Plains, there has been a recognized need for a systems modeling approach for sustainable agricultural research and development. Central to meeting the challenge of delivering viable decision support software, the USDA-ARS Great Plains Systems Research Unit (GPSRU), in a collaborative effort with Colorado State University (CSU), developed the Great Plains Framework for Agricultural Resource Management (GPFARM) DSS. The GPFARM DSS is primarily composed of six major components: 1) a Microsoft® Windows-based graphical user interface; 2) Microsoft® Access databases containing soil, crop, weed, climate, equipment, chemical, and economic parameters; 3) an object-oriented modeling framework and science simulation model; 4) a stand-alone economic analysis tool; 5) a set of analysis tools including a multicriteria decision making module, an output visualization module, and summary report tables and graphs; and 6) a web-based information system. Thus, GPFARM is unique in that it brings together the above suite of decision support tools integrated across a whole-farm/ranch system. The idea for developing GPFARM was conceptualized in the late 1980’s. Actual GPFARM development occurred from the early 1990’s to 2003, ending with the current version 2.6. Despite a reasonable level of producer involvement in the requirements analysis, development, and evaluation phases of GPFARM, it can be argued that the rate of adoption has been slow compared with the rate predicted for it over a decade ago at development initiation. The successor to GPFARM (iFARM – integrated Farm and Ranch Management) is currently under development. Can we use our experiences from the GPFARM project to improve iFARM? In this paper, we first provide a brief overview of the GPFARM DSS. We then discuss the lessons learned (e.g., successes and failures) in over a decade of agricultural DSS development. A summary of conclusions resulting from discussion and critical analysis of the GPFARM project include: 1) It is important that the DSS development process includes careful evaluation of the scope of the DSS in relation to the human and fiscal resources available; 2) Careful attention to the intended target user group(s) is needed; 3) Simpler tools or database information generated from simulation analyses of alternative management options may have been more appropriate; 4) The capability to rapidly update major components (e.g., simulation model, databases) and address current questions or problems in the system is an absolute necessity; and 5) An appropriate compromise between scientific rigor and simplicity is essential for critical DSS components to ensure overall quality of the product.

Technical Abstract: In the Great Plains, there has been a recognized need for a systems modeling approach for sustainable agricultural research and development. Central to meeting the challenge of delivering viable decision support software, the USDA-ARS Great Plains Systems Research Unit (GPSRU), in a collaborative effort with Colorado State University (CSU), developed the Great Plains Framework for Agricultural Resource Management (GPFARM) DSS. The general purpose of GPFARM is to serve as a whole-farm/ranch DSS in strategic planning across the Great Plains. GPFARM runs on a field-by-field basis (with aggregation up to the whole-farm/ranch level), and provides production, economic and environmental impact analysis and site-specific database generation, from which alternative agricultural management systems can be tested and compared. Agricultural consultants and producers (both farmers and ranchers) were targeted as the primary users of GPFARM. User requirements were identified by an ARS customer focus group comprised of eastern Colorado farmers, ranchers, agricultural consultants, and NRCS and extension professionals. The GPFARM DSS is primarily composed of six major components: 1) a Microsoft® Windows-based graphical user interface; 2) Microsoft® Access databases containing soil, crop, weed, climate, equipment, chemical, and economic parameters; 3) an object-oriented modeling framework and science simulation model; 4) a stand-alone economic analysis tool; 5) a set of analysis tools including a multicriteria decision making module, an output visualization module, and summary report tables and graphs; and 6) a web-based information system. Thus, GPFARM is unique in that it brings together the above suite of decision support tools integrated across a whole-farm/ranch system. The idea for developing GPFARM was conceptualized in the late 1980’s. Actual GPFARM development occurred from the early 1990’s to 2003, ending with the current version 2.6. Despite a reasonable level of producer involvement in the requirements analysis, development, and evaluation phases of GPFARM, it can be argued that the rate of adoption has been slow compared with the rate predicted for it over a decade ago at development initiation. The successor to GPFARM (iFARM – integrated Farm and Ranch Management) is currently under development. Can we use our experiences from the GPFARM project to improve iFARM? In this paper, we first provide a brief overview of the GPFARM DSS. We then discuss the lessons learned (e.g., successes and failures) in over a decade of agricultural DSS development. A summary of conclusions resulting from discussion and critical analysis of the GPFARM project include: 1. It is important that the DSS development process includes careful evaluation of the scope of the DSS in relation to the human and fiscal resources available (e.g., assessment of personnel available for developing, evaluating, implementing, and maintaining a DSS that matches the scope, scale, and complexity of the project). Formal project management and software engineering protocols and tools can aid in this regard. 2. Careful attention to the intended target user group(s) is needed by: 1) matching the proposed technology appropriately with the user, and 2) gathering input from a broad spectrum of potential users when performing a requirements analysis. 3. Simpler tools or database information generated from simulation analyses of alternative management options may have been more appropriate for delivery to producers and consultants at this stage in time. 4. The capability to rapidly update major components (e.g., simulation model, databases) and address current questions or problems in the system is an absolute necessity - the GPSRU has recently developed an Object Modeling System (OMS) for this purpose. In addition, an appropriate compromise between scientific rigor and simplicity