A cognitive systems framework to inform delivery of analytic support for farmers’ intuitive management under seasonal climatic variability

The idea of the decision support system (DSS) for farmers remains an enigma. Clever technology to bridge the gap between agricultural science and farming practice still seems appropriate. Many more of the conditions for success appear to exist today than ever before. Yet the DSS has yet to significantly colonise farm decision making practice. This paper comes late in a long program of research conducted to see if, and under what conditions, computer simulation of farming scenarios, on which a DSS generally depends, can be valued by farmers.The research approach used an unconventional prototypic information system (IS), comprising local measurements, models, and facilitated discussions that evolved in an action research program. The aim has been to elucidate the means by which successful simulation-based decision support intervention can take place and why it usually does not. This required a significant expansion of the researchers’ concept of the farm as a system to include the farmer’s internal system of practical knowing and learning. This paper reports on a cognitive framework model with transactions at interfaces with both the production system and the analytical IS. Its coarse structure is the classical perception-action cycle influenced by goals and outcome feedback. In the highly uncertain production environment of Australian dryland farming, personal judgement plays a significant mediating role between perception and action, and theory of a continuum between the judgement modes of intuition and analysis adds to framework structure. Further structure comes from the theoretical distinctions between holistic and arbitrary intuition, and between causal and probabilistic analysis. Analytic interventions influence: (a) awareness of current situation conditions and (b) expectations of future conditions and action outcomes, and these serve as primary cognitive resources for evaluation of possible actions in planning and decision making. A theory that matches our research experience in bridging the gap between analytic intervention and intuitive practice posits that virtual situations simulated with analytic models and outputs represented graphically can facilitate vicarious experiential learning. This dovetails with theory concerning the education of intuition.The paper concludes by applying criteria from the field of cognitive engineering to test whether the framework presents a concept of mind that is workable for informing practical model-based research and development aimed at supporting farmers’ judgments and decisions.     

Analysing agricultural users’ patterns of behaviour: The case of OPTIRas™, a decision support system for starch crop selection

Redesigning IT systems for specific user groups encompasses a lot of effort with respect to analysing and understanding user behaviour. The goal of this paper is to provide insights into patterns of behaviour of agricultural users, during the usage of a decision support system called OPTIRas™. This system aids agricultural users in their cultivar selection activities. We analyse logs resulting from OPTIRas™, and we get insights into user’s navigational patterns. We claim that the results of our analysis can be used to support the redesign of decision support systems in order to address specific agricultural users’ characteristics.     

Pasture and forage crop systems for non-irrigated dairy farms in southern Australia. 1. Physical production and economic performance

The dairy industry in southern Australia relies on perennial ryegrass pasture to supply 60–70% of the diet of lactating cows. Improvements in the amount and quality of home-grown forage used for dairy cow feeding are critical for further productivity gains in the industry. A modeling approach was used to estimate the effects of changing the forage system on farm business profit. Base models (using 100% of farm area in perennial ryegrass pasture) were constructed for above-average (Top 40%) and high performing (Top 10%) farm types typical of two locations: Terang in southwest Victoria and Ellinbank in Gippsland, eastern Victoria. These models were then re-simulated using different forage base options such as: oversowing annual ryegrass, winter crops (annual ryegrass monoculture, winter cereal grown for whole crop silage), summer crops (grazing brassicas, maize), combinations of these (double cropping), or summer shoulder pasture (notionally based on tall fescue) on between 10% and 100% of farm area.