Assessing the sustainability of agricultural systems encompasses complex and interchanging economic, environmental and social issues, and requires multi-criteria decision-analysis approaches. Various models have been proposed to assess agricultural sustainability considering these issues, based for example on programs for multi-attribute decision making or Fuzzy Interference Systems. However, we identify a lack of comprehensive models applicable to broad agricultural conditions in different environments and socioeconomic contexts. To fill this gap, we propose a novel, indicator-based fuzzy logic model for assessing the sustainability of agricultural systems. To test the model’s suitability, we conducted twenty-two case studies over the 2018/19 cropping season in the Brazilian agricultural-forest frontier region; the farms chosen represent the three most common farming systems there: (i) pure crop farming (crop rotation only: soybean - corn), (ii) pure livestock, and (iii) integrated farming (crop - livestock and livestock - forest). Partial indicators were built to assess the economic, environmental, and social performances of those farming systems, then were further integrated in a sustainability index. The results show higher and better-balanced performance for integrated farms, which displayed the highest sustainability index values. In contrast, livestock farms performed poorly in all dimensions and showed the lowest sustainability index. Crop farms showed higher economic, but lower social and environmental performances. These results are in contrast to the oft-perceived trade-offs among different pillars of sustainability and show that integrated systems have the potential to balance multiple sustainability objectives, by leveraging multiple subsystem synergies. The innovative fuzzy inference model proposed is suitable to deal with information at the farm level, handling different types of farming systems, and applicable to different environmental or socioeconomic contexts. Moreover, the proposed indicators and associated indices offer relevant information to policy-makers to foster the sustainable intensification of farming systems, while promoting environmental protection and the coexistence of biodiversity and the agricultural sector.

评估农业系统的可持续性包括复杂且相互交织的经济、环境和社会问题，需要采用多标准决策分析方法。考虑到这些问题，已经提出了各种模型来评估农业可持续性，例如基于多属性决策程序或模糊干扰系统。然而，我们发现缺乏适用于不同环境和社会经济背景下广泛农业条件的综合模型。为了填补这一空白，我们提出了一种新颖的、基于指标的模糊逻辑模型来评估农业系统的可持续性。为了测试模型的适用性，我们在 2018/19 种植季期间在巴西农林边境地区进行了 22 个案例研究；所选农场代表那里三种最常见的耕作系统：(i) 纯作物耕作（仅作物轮作：大豆-玉米），(ii) 纯畜牧业，以及 (iii) 综合耕作（作物-畜牧业和畜牧业-森林）。建立了部分指标来评估这些农业系统的经济、环境和社会绩效，然后进一步整合到可持续性指数中。结果显示，综合农场的绩效更高、更平衡，显示出最高的可持续性指数值。相比之下，畜牧场在各个方面都表现不佳，可持续性指数最低。农作物农场表现出较高的经济效益，但社会和环境绩效较低。这些结果与人们经常认为的可持续性不同支柱之间的权衡取舍形成鲜明对比，并表明集成系统有可能通过利用多个子系统的协同作用来平衡多个可持续性目标。提出的创新模糊推理模型适用于处理农场层面的信息，处理不同类型的农业系统，适用于不同的环境或社会经济背景。此外，拟议的指标和相关指数为决策者提供了相关信息，以促进农业系统的可持续集约化，同时促进环境保护以及生物多样性与农业部门的共存。提出的创新模糊推理模型适用于处理农场层面的信息，处理不同类型的农业系统，适用于不同的环境或社会经济背景。此外，拟议的指标和相关指数为决策者提供了相关信息，以促进农业系统的可持续集约化，同时促进环境保护以及生物多样性与农业部门的共存。提出的创新模糊推理模型适用于处理农场层面的信息，处理不同类型的农业系统，适用于不同的环境或社会经济背景。此外，拟议的指标和相关指数为决策者提供了相关信息，以促进农业系统的可持续集约化，同时促进环境保护以及生物多样性与农业部门的共存。