Exploring the synergistic influences of labour, environmental, and plantation management practices in the rubber latex production industry
Аннотация
Major global suppliers of natural rubber and its derivatives—including rubber sheets, latex, tyres, and garments—face common challenges in rubber plantation management. This study examined spatio-temporal variation in rubber latex production across 15 divisions in Sri Lanka from 2013 to 2022 and identified principal influencing factors using random forest machine learning. Monthly data on latex extraction and predictor variables were classified into six categories: environmental, plantation management, labour-related, technological, temporal, and geographical. Initially, time series plots were employed to illustrate the spatio-temporal variation in latex output across divisions. Trend and seasonality intensity in the latex production time series were analysed to assess regional disparities in rubber latex production. The findings indicate the presence of both spatial and temporal variability in rubber latex production. The random forest technique, accompanied by various importance metrics, was employed to ascertain the most significant elements. The findings identified year, month, division, quantity of stimulant-treated trees, and labour turnover as the primary determinants. Moreover, the variable month exhibited interaction effects with multiple other components, signifying intricate temporal dynamics. These findings provide critical insights for enhancing rubber latex production through strategic management and policy measures.