Carbon-aware coordinated supply chain scheduling in the net-zero era: Combining order allocation, production scheduling, and distribution routing problem

• A novel coordinated model integrates procurement, production, and distribution decisions. • Carbon emissions are estimated from energy usage in production and fuel in distribution. • Flexible job shop scheduling includes setup and intra-shop transport times. • A hybrid ENSGA-II algorithm combines NSGA-II, PSO, and VNS for optimization. • Proposed method improves solution quality in cost and emission trade-offs. As global industries race toward greener operations, sustainable manufacturing is no longer a competitive advantage but a necessity. Amid this urgency, traditional supply chain management often treats procurement, production, and distribution as independent processes, leading to suboptimal performance and increased environmental impact. To address this, we propose a novel C arbon- e fficient C oordinated P rocurement, P roduction Scheduling, and D istribution Routing S ystem ( CE-CPPDS ) that aims to minimize total supply chain costs and carbon emissions simultaneously. The model involves a multi-stage process where raw materials are sourced from suppliers, processed in a flexible job shop environment near the manufacturer, and final products are delivered using a heterogeneous fleet of vehicles. Unlike approaches focused solely on cost or profit, carbon emissions are explicitly considered on a multi-objective setup based on energy consumption across production states (processing, idle, transport, setup, auxiliary) and fuel usage in delivery. Energy-efficient production scheduling is achieved by reducing idle times and improving machine assignment. In addition, route optimization and vehicle selection support further fuel reduction. To address the computational challenge in solving, we develop an enhanced non-dominated sorting genetic algorithm (ENSGA-II) that integrates global and local search capabilities. The algorithm is benchmarked against state-of-the-art multi-objective evolutionary algorithms and an exact solver using twelve CE-CPPDS instances. Results show that the proposed method delivers a superior solution quality across these instances, with an average hypervolume (HV) of 0.7199 and an inverted generational distance plus (IGD + ) of 0.0241, as validated by nonparametric statistical tests. This study presents a practical and scalable framework for companies seeking to achieve cost-effective and environmentally sustainable supply chain operations through coordinated, carbon-aware scheduling.