Author
IMAM B - UNIV. OF ARIZ. | |
YAKOWITZ D S - 5342-45-00 | |
LANE L J - 5342-45-00 |
Submitted to: Multiobjective Decision Support System Conference
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 9/18/1996 Publication Date: N/A Citation: N/A Interpretive Summary: The effects of annual and long term variations of climatologic processes on the decision recommendations of a USDA-ARS Multiple Objective Water Quality Decision Support System and the effect that the application point of certain subprocesses of the system have on the outcome are investigated. The system consists of a model which simulates a farm management system and a unique decision model for evaluating the output of the simulations. The objective of the system is to rank farm management alternatives in preference order consistent with their environmental and economical impacts. The decision making process is divided into three sub processes that convert the values from the simulation to scores, aggregate the individual criterion scores and rank the alternative management systems based on their best and worst possible composite scores. In this study, four different ranking schemes are designed by changing the point at which each decision making sub-process is performed. The four ranking schemes were tested for two different management priorities. In the first case, the decision process emphasizes nutrients in runoff, sediment and percolation. In the second case, economical concerns indicated by net farm returns supersedes other management factors. In both cases, all four experiments produced the same results and thus are supportive of the current implementation of the decision model. Technical Abstract: The effects of annual and long term variations of climatologic processes on the decision recommendations of a USDA-ARS Multiple Objective Water Quality Decision Support System and the effect that the point at which aggregation of information occurs in the sytems subprocesses are investigated. The system consists of an embedded simulation model and a unique multiobjective decision model. The objective of the system is to rank farm management alternatives in preference order consistent with their environmental and economical impacts. The decision making process is divided into three sub processes that include: 1) conversion of the values of the decision criteria into the unitless [0,1] domain through scoring, 2) aggregation of the individual criterion scores into best and worst composite scores, and 3) ranking the alternative management systems based on their composite scores. In this study, four different ranking schemes are designed by changing the point at which each decision making sub-process is performed with respect to a stochastic ensemble of model input. The four ranking schemes were tested for two different management priorities: nutrient loading and net farm income. In both cases, the results were insensitive to the point at which the aggregation was performed and thus are supportive of the currently implemented choice to aggregate the simulation output prior to the first decision model sub-process. |