Improving farming systems in resource-poor contexts is often difficult as farmers face multiple challenges to implement the innovations developed by researchers. Viable solutions may however be present within local communities by positive deviant farmers, i.e. farmers that outperform positively compared to others. This study develops a positive deviance informed methodology to support redesign of farming systems, with the aim to improve farm productive, economic and environmental performances. We tested the methodology in Bihar, India, using survey data from 43 farms and the indicators of operating profit, soil organic matter balance, water use and dietary energy production. Positive deviant farms and practices were first identified and then recombined into a redesigned farm in consultation with farmers. The FarmDESIGN model was used to calculate current farm performance and to explore potential alternative farm configurations in the redesign. We found that outstanding performance on all indicators could only be reached by integrating high livestock density with an optimal combination of crop practices, which confirms the key role of interactions among components in mixed crop-livestock systems to improve all dimensions of farm sustainability. The redesigns outperformed all real farms on the indicators assessed. Farmers confirmed the viability of the redesigns in focus group discussions and their suggestions can serve as useful input for a next cycle of farm redesign. Since all suggestions are locally practiced and have proven to be accessible, affordable and recognizable, we conclude that our methodology based on positive deviant farms and practices yields promising results with a large potential to boost agricultural development for resource-scarce smallholder farmers.

中文翻译：

---

使用积极的偏差方法为农业系统重新设计提供信息：来自印度比哈尔邦的案例研究

由于农民在实施研究人员开发的创新方面面临多重挑战，因此在资源贫乏的情况下改善农业系统通常很困难。然而，积极的越轨农民可能会在当地社区中提出可行的解决方案，即与其他农民相比，表现优于其他农民的农民。本研究开发了一种积极的偏差知情方法，以支持重新设计农业系统，旨在提高农业生产力、经济和环境绩效。我们在印度比哈尔邦测试了该方法，使用来自 43 个农场的调查数据和营业利润、土壤有机质平衡、用水和膳食能量生产指标。积极的异常农场和做法首先被确定，然后在与农民协商后重新组合成一个重新设计的农场。FarmDESIGN 模型用于计算当前农场的绩效，并在重新设计中探索潜在的替代农场配置。我们发现，只有通过将高牲畜密度与作物实践的最佳组合相结合，才能在所有指标上实现出色的表现，这证实了混合作物-牲畜系统中各组成部分之间的相互作用在改善农场可持续性的各个方面的关键作用。在评估的指标上，重新设计的表现优于所有真实农场。农民在焦点小组讨论中确认了重新设计的可行性，他们的建议可以作为下一个农场重新设计周期的有用输入。由于所有建议都在当地实践并已被证明是可获得的、负担得起的和可识别的，