Scholarlink

<p>Agricultural production faces increasing pressure to meet global food demand while ensuring sustainable resource use and effective territorial planning. Although geospatial technologies and agricultural cadastres have improved land monitoring and mapping, their integration with optimisation approaches for parcel-level decision-making remains limited. This study develops an integrated cadastral–optimisation framework that combines geomatic tools and Genetic Algorithms (GA) to enhance agricultural potential and support rural planning. The methodology was applied in the Cerezal–Bellavista commune (Ecuador) and structured into four phases: (i) data collection and territorial characterisation using field surveys, Global Navigation Satellite Systems (GNSS), Unmanned Aerial Vehicles (UAV) imagery, and Sentinel-2 data; (ii) development of an agricultural cadastre through Land use/Land cover (LULC) and crop-type classification; (iii) agricultural potential modelling using a GA based on biophysical, environmental, and socioeconomic variables; and (iv) formulation of strategic guidelines using SWOT–TOWS analysis. The optimisation process generated three Agricultural Potential Models (APMs) representing economic–crop interactions, biophysical–environmental suitability, and cost-efficiency dynamics. Findings reveal a highly fragmented cadastral structure, with 95.24% of plots smaller than 5.50&#xa0;ha. The GA demonstrated robust convergence behaviour, with the economic–crop model achieving the highest performance (90.4% fit), followed by the environmental model (75.81%) and the efficiency model (68.72%). The findings indicate that strongly correlated economic and crop variables drive optimisation performance, while environmental variables increase system complexity. The spatial implementation of the APMs enabled the identification of high-performing plots and priority intervention areas, highlighting the framework’s operational value for decision-making. This study advances the literature by operationalising cadastral data within an optimisation framework. The approach provides a scalable, data-driven tool for agricultural management, land-use planning, and sustainable rural development.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A cadastral-based optimisation framework using genetic algorithms for agricultural production and rural planning

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月15日

查看全文
  • Paulo Escandón-Panchana,
  • Sandra Martínez Cuevas,
  • Lourdes Ortega Maldonado,
  • Andrés Velastegui-Montoya,
  • Gricelda Herrera-Franco

摘要

Agricultural production faces increasing pressure to meet global food demand while ensuring sustainable resource use and effective territorial planning. Although geospatial technologies and agricultural cadastres have improved land monitoring and mapping, their integration with optimisation approaches for parcel-level decision-making remains limited. This study develops an integrated cadastral–optimisation framework that combines geomatic tools and Genetic Algorithms (GA) to enhance agricultural potential and support rural planning. The methodology was applied in the Cerezal–Bellavista commune (Ecuador) and structured into four phases: (i) data collection and territorial characterisation using field surveys, Global Navigation Satellite Systems (GNSS), Unmanned Aerial Vehicles (UAV) imagery, and Sentinel-2 data; (ii) development of an agricultural cadastre through Land use/Land cover (LULC) and crop-type classification; (iii) agricultural potential modelling using a GA based on biophysical, environmental, and socioeconomic variables; and (iv) formulation of strategic guidelines using SWOT–TOWS analysis. The optimisation process generated three Agricultural Potential Models (APMs) representing economic–crop interactions, biophysical–environmental suitability, and cost-efficiency dynamics. Findings reveal a highly fragmented cadastral structure, with 95.24% of plots smaller than 5.50 ha. The GA demonstrated robust convergence behaviour, with the economic–crop model achieving the highest performance (90.4% fit), followed by the environmental model (75.81%) and the efficiency model (68.72%). The findings indicate that strongly correlated economic and crop variables drive optimisation performance, while environmental variables increase system complexity. The spatial implementation of the APMs enabled the identification of high-performing plots and priority intervention areas, highlighting the framework’s operational value for decision-making. This study advances the literature by operationalising cadastral data within an optimisation framework. The approach provides a scalable, data-driven tool for agricultural management, land-use planning, and sustainable rural development.