Changing systems for supporting farmers' decisions: Problems, paradigms, and prospects

Decision support systems (DSS), like other information systems (IS) before them, were designed to serve functions deemed by ‘management scientists’ to be potentially useful to managers. But the unwelcome fact is that the use of agricultural DSSs by managers of farms has been low. This paper probes possible reasons for this through interpretation of agricultural DSS case histories and several strands of relevant social theory. From nine cases of DSS development effort and 14 products interpreted comparatively, a number of generalisations are made that serve as reference points in the following search for explanation in theory. First, the nature of management practice of family farms is explored and differences between the internal structure governing personal action and the scientific approach to practice are contrasted. Next, the interaction between the nature of the particular action/practice and the nature of the DSS is explored. A DSS designed to provide integrated, optimal recommendations for management typifies the DSS as a proxy for a manager's decision process. Examples of elaborate expert systems that simply were not used dramatically illustrate the resistance of family farmers to have their decision processes by-passed. On the other hand, the DSS designed to serve as a tool in a modified decision process is shown to have experienced higher use, by deriving and exploiting ‘deep’, abstract information about the system, by introducing a powerful ‘logic’, or a combination of both. A number of the referenced case stories demonstrate the resurgence of the decision support mode whereby the simulator is in the hands of an expert intermediary as an alternative to easy-to-use software in the hands of a farmer. This is the mode of operational research/management science, which preceded the DSS. In comparision with hierarchical organisations, available options for overcoming the persistent ‘problem of implementation’ of the DSS in family farms are inherently weak. This focuses attention on the importance of the relationship between the DSS developer and the potential user. Drawing on a classic typology of possible configurations of ‘understanding’ between the scientist and the manager, four approaches to intervention are discussed. Three entail a degee of engagement that qualified them as ‘participative’. But one of these constitutes a departure from the DSS and broader IS traditions that places it in another paradigm. In this ‘mutal understanding’ relationship, intervention intent shifts from educating and persuading to recognition of and respect for other ways of viewing the world. This opens up the opportunities for co-creating information systems that utilise the comparative advantages of both practical and scientific knowledge. Intervention emphasis shifts from prescribing action to facilitating learning in actions. Although the DSS has fallen far short of expectations in its influence on farm management, the experience has been instructive in multiple ways to both farmers and professionals in agriculture. In many cases, farmers learned from the DSS and could then jettison it without loss. From disappointments scientists have sometimes learned what was needed to achieve a better outcome. From collated DSS experiences, important lessons for the future can be drawn. The paper concludes by conjecturing that the future of the DSS and related ISs, while more limited than once imagined, holds promise in four directions: A ‘small’ tool for aiding farmers' tactical decisions; a versatile simulator as a consultant's tool; a versatile simulator as the core of a facilitated ‘learning laboratory’, and a formal framework that supports regulatory objectives in constraining and documenting farming practice.