Sustainable business practices through AI-Enhanced management and marketing

Sustainability has become a critical concern in contemporary business management and marketing, driven by environmental challenges and stakeholder expectations. Conventional decision-support systems, typically consisting of XGBoost predictive analytics and linear programming (LP) optimization have the advantage of resource allocation but lacks personalization, superior reasoning, and integrated sustainability scoring. This study develops an AI-enhanced multi-Agent framework integrating large language models (LLMs) with retrieval-augmented generation (RAG) to deliver personalized and context-sensitive decision support that is aligned with sustainability objectives. This paper contrasts the current MultiAgent LLM + RAG framework with the traditional XGBoost + LP model in terms of marketing optimization, operational efficiency, sustainability, and interpretability on real-world marketing and operations sustainability datasets. The multi-agent approach improves resource utilization to 90.2%, lowers CO2 emissions by 26%, boosts conversion lift by 8% and ROI by 12%, and maintains competitive profit levels, according to key results. It also exhibits better sustainability scores and 13.6% less energy use. Human-centered assessments emphasize greater usability, clarity, and reliability. These findings elevate the prospect of AI-supported multi-agent models to optimize business financial performance and environmental sustainability and further improve goal-aligned, transparent decision-making in business. The work sets the stage to the wider implementation of sustainable, intelligent decision support in industries.