cover
Contact Name
Pegi Sugiartini
Contact Email
journaljedvb@gmail.com
Phone
+6281311603106
Journal Mail Official
journaljedvb@gmail.com
Editorial Address
https://jedvb.polteksci.ac.id/index.php/jedvb/editorialTeam
Location
Kab. cirebon,
Jawa barat
INDONESIA
Journal of Economic Development and Village Building
ISSN : -     EISSN : 29864666     DOI : https://doi.org/10.59261
This journal contains articles and research results. The scope of the research includes: general economics and teaching, public economic, law and economics, building village, economic village, Business admiration and business economics: marketing, accounting; personel economics
Articles 51 Documents
AI-DRIVEN VILLAGE PLANNING: PREDICTIVE MODELS FOR ENHANCING RURAL ECONOMIC RESILIENCE IN EMERGING REGIONS Sugiartini, Pegi
Journal of Economic Development and Village Building Vol. 3 No. 1 (2025): Journal of Economic Development and Village Building
Publisher : Politeknik Siber Cerdika Internasional

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59261/jedvb.v3i1.52

Abstract

Rural communities in emerging regions face mounting challenges, including economic volatility, climate variability, and limited access to infrastructure. However, traditional planning approaches often rely on intuition-based priority-setting and lack systematic analytical frameworks for identifying optimal intervention pathways. The integration of artificial intelligence into development planning offers potential to enhance evidence quality and allocative efficiency, though implementation feasibility and effectiveness in resource-constrained contexts remain underexplored. This research developed and validated an AI-based predictive framework to assess village economic resilience and support participatory development planning, examining model accuracy, key resilience determinants, and its practical integration with existing governance processes. The study employed mixed methods across five rural villages in Central Java Province, Indonesia, over six months (March-August 2024), combining machine learning approaches (Random Forest, Gradient Boosting) for resilience prediction with qualitative stakeholder engagement. Data collection encompassed household surveys (n=180), administrative records, spatial analysis, and participatory planning forums, with systematic comparison against conventional planning approaches. Ensemble models achieved strong predictive accuracy (R²=0.84), with Gradient Boosting demonstrating the highest performance (R²=0.89) and Random Forest (R²=0.86), substantially outperforming linear regression (R²=0.48). Digital infrastructure (24% importance), income diversification (21%), and financial service access (17%) emerged as dominant resilience determinants. AI-supported villages demonstrated enhanced planning processes, including improved evidence utilization, broader stakeholder participation, and strategic realignment of priorities toward empirically identified leverage factors. Scenario analysis projected 18-point gains in resilience over five years under integrated intervention strategies. The research demonstrates that appropriately designed AI systems can enhance the effectiveness of rural development planning while preserving participatory values.