Toward tailored interventions in plantain (Musa paradisiaca L.) industry: Insights from heterogeneity and constraints to plantain-based cropping systems in South-Benin

Context: Plantain plays a key role in food security, income diversification and poverty reduction. Its development is hampered by various factors, which analysis by accounting for cropping systems, could lead to effective targeted interventions.OBJECTIVE: The study aims at characterizing the plantain-based cropping systems and analyzing the constraints to its production in South-Benin.Methods: Quantitative and qualitative data were collected from 584 randomly selected farmers operating on 724 plots in south-Benin. To undertake the typology, the Mixed Data Factor Analysis (MDFA) combined with combined with the Calinski-Harabasz pseudo-F index and bootstrapping were applied. Kendall rank correlation was used to prioritize the constraints. Content analysis was completed to analyze the qualitative data related to the constraints.Results And Conclusions: Five types of plantain-based cropping systems (CS) were identified: «Pure CS», «Mixed CS with staggered seedling», « Irrigated mixed CS with high productivity », «Small fields in mixed CS with natural mat expansion», «Mixed CS with cereals and row seedling». «Type3» characterized by additional water supply and high crop diversity is the most productive (7794±4210 kg ha-1). Drought, wind and low prices are the most important constraints that all «types» agree on. However, flood, lack of measurement standards, financing constraints, pests and diseases, and difficult market access are among the constraints whose importance is specific to each «type». Farmers lack appropriate technical knowledge on production techniques and strategies to overcome these constraints. More targeted interventions by type on water management, market access, financing facilities, training and technical support could help improve production and develop the sector.SIGNIFICANCE: This study emphasizes the need to integrate typology into farm analysis in order to propose specific technical recommendation for each «type». Better levels of adoption and more impactful interventions could then be achieved.