Existing sustainability assessment tools are often unsuitable for smallholder farmers in developing countries, where resource constraints and economic pressures outweigh environmental and social considerations. This study develops and validates an SDG-aligned index that operationalizes Cleaner Production (CP) at the farm level to measure the sustainable performance of smallholder dairy farmers in Sri Lanka. Drawing on triple bottom line (TBL) and sustainable supplier development (SSD) concepts, we test a conceptual model using survey data from 324 farmers. Partial least squares structural equation modelling (PLS-SEM) estimates construct weights that are translated into a transparent 0–1000 scoring system. The index comprises five dimensions—farmer capability (FC), farmer–processor relationship (FPR), and farmer economic (ES), social (SS), and environmental (EnvS) sustainability. Results show FC is the strongest contributor (236 points), while ES lags (170 points). By embedding CP principles—resource efficiency, waste minimization and circularity, and pollution prevention—the index enables processors to classify farmers, tailor interventions, and integrate verified outcomes into sustainability reporting. The framework contributes to SSD and CP practice, offers a replicable, low-burden measurement tool for agri-food supply chains, and outlines a tiered linkage to life-cycle assessment and carbon accounting for sector dashboards. The paper's primary contribution is methodological and practical: it operationalizes established TBL, SSD, and CP perspectives into an empirically weighted (0–1000) tool for smallholder supplier management and SDG reporting. • Developed a 0–1000 index for smallholder dairy farmer sustainability. • Index integrates enablers, TBL outcomes, Cleaner Production principles. • Validated with PLS-SEM using data from 324 Sri Lankan dairy farmers. • Farmer capability is the strongest contributor, economic sustainability lags. • Tool supports SDG reporting and scaling sustainability in agri-food chains.
DeSilva et al. (Thu,) studied this question.