Abstract: This study investigates the optimal decision-making and coordination mechanisms of intelligent manufacturing systems under a carbon-quota trading policy, revealing how government subsidies, carbon-quota prices, and consumers′ low-carbon preferences affect firms′ behavior and system performance. To this end, considering both the government′s carbon-quota trading policy and consumers′ low-carbon preferences, a dynamic analytical framework is established to construct a tripartite differential-game model involving a manufacturer, a retailer, and the government. The equilibrium strategies are compared under three decision modes: decentralized, retailer-myopic, and centralized. Analytical and simulation results show as follows: 1)Under centralized decision-making, the steady-state level of low-carbon goodwill and the payoffs of all system members are significantly higher than those under decentralized or myopic modes, making it the optimal strategy for achieving long-term sustainable development; when short-term profit is prioritized, decentralized decision-making is preferred. 2)The carbon-emission coefficient has a significant positive effect on product pricing but negative effects on the subsidy level,manufacturer′s R&D intensity, retailer's marketing effort, low-carbon goodwill, product demand, and equilibrium payoffs. 3)Both carbon-quota trading prices and consumers′ low-carbon preferences reduce firms′ R&D and marketing investments while increasing product prices; moreover, the carbon-quota trading price is positively related to the level of intelligent-manufacturing subsidies but negatively related to consumers′ low-carbon preferences. Unlike previous studies that focus only on a single policy scenario, this study integrates policy and market mechanisms, providing theoretical and managerial implications for optimizing intelligent manufacturing systems under coordinated low-carbon policies.