Participatory decision support for agricultural management. A case study from Sri Lanka

Agricultural policy makers were helped to construct and use a decision support system (DSS) to identify problems and assess potential solutions for a river basin in Sri Lanka. Through building the DSS themselves, policy makers should reach better decisions. The main aim of the study was to test whether this could be done using a tool called a Bayesian network (BN) which is accessible to non-specialists and able to provide a generic, flexible framework for the construction of DSS. Results from a workshop indicated that the approach showed promise, providing a common framework for discussion and allowing policy makers to structure complex systems from a multi-disciplinary perspective. The need for a multi-disciplinary perspective was clearly demonstrated. The study also suggested improvements to the ways in which BNs can be used in practice. Further workshops with farmers highlighted the importance of involving them in the planning process and suggested more effective ways of doing this while using BNs.