Proposta de modelo para avaliação de desempenho das práticas de logística verde em indústrias de transformação

Abstract

The growing pressure for sustainability requires manufacturing industries to adopt environmentally responsible practices. In this context, Green Logistics (GL) plays a strategic role, yet its performance assessment still faces a significant gap: the absence of integrated models that simultaneously address supply, production, distribution, and reverse logistics, thus hindering a systemic analysis. This theoretical shortcoming was empirically validated in this research through a diagnostic survey focused on 76 companies in the electronics sector of the Manaus Industrial Hub (PIM), which revealed that the majority (73.7%) lack a consolidated measurement system. Given this problem, the objective of this study was to develop an evaluation model to measure the performance of GL practices in this context. To this end, an applied, mixed-methods approach was adopted, organized into five phases: a literature review, the administration of questionnaires, interviews with experts, and a case study in two PIM industries to test and validate the proposal. As its main result, the study presents an evaluation model composed of 31 performance indicators, segmented by the four logistics areas, and a five-level maturity scale (Incipient, Under Development, Structured, Optimized, and Strategic). Technical validation with experts demonstrated the model's robustness, while its diagnostic application in the case study revealed greater maturity in production logistics compared to the challenges found in the supply and reverse logistics links, reinforcing its utility for identifying weaknesses and strengths. It is concluded that the objectives were fully achieved and the research question was answered. The research provides a methodological tool with theoretical validity and practical applicability, whose main contribution lies in filling the identified gap. The proposed model offers organizations an integrated instrument for diagnosis, monitoring, and decision-making, with multiple impacts: academic, by offering a validated and replicable model; economic, by enabling resource optimization and the reduction of operational costs; and socio-environmental, by promoting safer operations and fostering advances toward a circular economy, gaining amplified relevance due to its potential for application in the sensitive industrial context of the Amazon.