A novel decision analysis and risk assessment framework for improving agro-ecosystem interventions

 Designing and implementing agro-ecosystem intervention decisions is a complex undertaking, outcomes are uncertain and risk of project failure and cost overruns are large.  Sub-optimal intervention decisions are resulting from failure to consider uncertainty, omission of important variables that are difficult to measure, and a lack of frameworks for combining diverse types of information.  We aim to develop and promote a Bayesian analysis framework and set of tools that will help agricultural scientists and development specialists easily apply sound probabilistic decision analysis and risk assessment to improve agro-ecosystem intervention design, monitoring and evaluation.  We plan to pilot the tools in regional development programmes across Africa and Asia and train a cadre of practitioners in their use.  The most critical decision facing development planners is how to approve and prioritize projects. A small investment in improving decision quality can benefit livelihoods and ecosystems throughout the developing world, and save billions of dollars in sub-optimal investments.