Kaizen como facilitador de la economía circular en las operaciones agroindustriales de banano: evidencia empírica de la reducción de residuos plásticos y la recuperación del valor de los materiales.

The global imperative to transition from linear to circular production models has intensified pressure on agro-industrial firms to develop operational frameworks that simultaneously reduce environmental impact and preserve resource value. This study investigates Kaizen methodology as a practical circular economy (CE) enabler in banana production operations in northern Peru, addressing a significant gap in the literature on CE implementation at the operational level in agricultural value chains. Employing a quantitative pre-experimental design (O₁XO₂) across 18 banana crops, Kaizen was operationalized through five structured phases—Define, Register, Design, Implement, and Verify—targeting material waste reduction and quality preservation across the full production chain. Post-intervention results demonstrated a statistically significant 28.4% reduction in total weekly waste (587.5 → 420.63 kg; 89Wilcoxon signed-rank test, Z = -2.666, exact p = 0.004, r = 0.89), with HDPE plastics declining 30.9%, packaging films 29.3%, and adhesive tapes 27.7%. Critically, systematic segregation and handling protocol improvements shifted the clean-to-contaminated HDPE ratio from 1:2 to 1:1.2, substantially enhancing recyclability and unlocking previously inaccessible secondary market value streams. The findings introduce the concept of the ‘Kaizen-CE nexus’—a mechanism by which incremental operational improvement generates circular value through three pathways: waste volume reduction, material quality preservation for recycling, and systematic waste-flow data generation enabling continuous optimization. This framework extends Kaizen theory beyond manufacturing into agro-industrial systems, provides practitioners with a low-cost replicable model for SDG 12 alignment, and contributes to strategic management literature on the sustainability-efficiency nexus in agricultural value chains. Given the pre-experimental, single-site design without a control group, these findings should be interpreted as evidence of association and operational effectiveness rather than definitive causal proof, and as an initial step requiring multi-site, multi-commodity validation before broader generalization.