Economic Impact of Pesticides on the Colombian Floriculture Industry: A Predictive Model Using Deep Neural Networks for Labor Cost Optimization
Floriculture, Neural networks, pesticides, labor costs.
AuthorsAbstractThe Colombian floriculture sector is one of the country’s most important economic sectors because of its contribution to GDP and exports. However, the use of pesticides and fungicides in flower production increases the risk of occupational diseases. This study developed a predictive model based on deep neural networks and reinforcement learning using the PPO algorithm and a 256–128–64 architecture. Using longitudinal data from 50 companies covering 2014–2024 and 22 normalized variables representing production, financial, labor, and occupationalhealth dimensions, the model achieved an accuracy of 89.3% and an R² of 0.86. The results indicate that pesticide exposure generates an estimated annual cost of USD 165,120 per company, equivalent to 4.2% of operating profit. The study concludes that pesticide exposure compromises the long-term competitiveness and sustainability of the sector.
•••••••••••••••••••••••••••••••• ejprd.org - Published by Riset Publication Services LLC
EJPRD
Copyright ©2026 by Riset Publication Services LLC