Life Cycle Assessment and Multiobjective Optimization for Steam Cracking Process in Ethylene Plant.

Area of Science:

• Chemical Engineering
• Environmental Science
• Industrial Ecology

Background:

• Growing emphasis on energy savings and emission reduction in industrial policies.
• Limited focus on the upstream environmental impacts (feedstock extraction, processing, transport) of industrial production.
• Need for integrated approaches to assess and mitigate industrial environmental footprints.

Purpose of the Study:

• To propose an integrated multiobjective optimization framework for steam cracking processes.
• To develop and compare economic-environmental optimization models using life cycle assessment (LCA) and data-driven methods.
• To evaluate the trade-off between economic benefits and environmental impacts in ethylene production.

Main Methods:

• Life Cycle Assessment (LCA) integrated with data-driven modeling.
• Development of a multiobjective economic-environmental optimization model.
• Comparative analysis with a standard energy cost model.
• Utilization of the nondominated sorting genetic algorithm-II (NSGA-II) for optimization.
• An industrial case study for validation.

Main Results:

• The LCA-based optimization framework effectively integrates upstream and process-specific environmental impacts.
• The proposed method demonstrates a superior representation of environmental impacts compared to the standard energy cost model.
• The nondominated sorting genetic algorithm-II successfully identified Pareto optimal solutions.
• The industrial case study confirmed the practical applicability and effectiveness of the developed framework.

Conclusions:

• The LCA-based optimization framework provides a more comprehensive assessment of environmental impacts in steam cracking.
• The method enables a better balance between economic performance and environmental sustainability in ethylene production.
• This approach offers valuable guidance for environmentally conscious industrial decision-making and policy implementation.