Smallholder Social Enterprise Development and Value Chain Inclusivity: A Review of Cooperative Models, Market Access, and Equitable Employment in Palm Oil Supply Chain

This research is motivated by the increasing need for a more consistent and technology-based quality control system in the palm oil industry, particularly in determining the ripeness level of fresh fruit bunches (FFB), which has so far relied on subjective manual assessments. This study aims to synthesize the development of artificial intelligence (AI)-based methods, especially computer vision and deep learning, in the classification of FFB ripeness. The methodology used is a Systematic Literature Review of 37 Scopus-indexed articles from the period 2020–2026. The research results indicate a significant shift from conventional machine learning methods to CNN-based deep learning models with accuracy generally above 90%. The implementation of this technology is also beginning to move towards real-time systems based on mobile and industrial devices. From a policy perspective, these findings underscore the potential of AI to enhance the efficiency, consistency, and sustainability of palm oil industry management through the digitization of harvesting and processing processes.