Abstract: Under the context of China's carbon neutrality strategy and the green transformation of the construction industry, the refined management of carbon emissions during construction processes has become a critical issue. Although academia identifies digital twins as a key direction for the low-carbon and intelligent transformation of construction, no comprehensive digital twin system currently exists to guide carbon emission management in construction projects. This study, based on the concept of digital twins, establishes a simulation and optimization control framework for carbon emissions during construction processes. By analyzing emission sources such as machinery, material transportation, and workforce activities, a data-driven integrated control system is designed to monitor emissions in real-time across various construction stages. Within this framework, methods such as discrete events, multi-agent systems, and system dynamics are proposed for simulation modeling, covering processes, equipment operations, and material usage from micro to macro levels. Simulation results are analyzed using intelligent algorithms to provide visualized insights, serving as a scientific basis for optimization strategies. The results, evaluated within a multi-objective optimization and decision-making system, incorporate construction requirements and explore bias-aware multi-objective algorithms to dynamically balance competing goals. Through visualization and interactive decision-making, construction managers can address sustainability alongside other project benefits under complex constraints. This framework offers a research methodology for digital twin-driven simulation and optimization of carbon emissions in construction, advancing intelligent carbon emission management and expanding future applications.