1. Introduction

Amid global economic integration and digital transformation, supply chain management is confronted with challenges like personalized consumer demands, uneven market resource distribution, and prominent environmental issues [1]. From 2020 to 2024, global supply chain disruptions surged by 67%, causing over $4.2 trillion in economic losses, while logistics industry carbon emissions account for 11.5% of total emissions and continue to rise. AI's data processing and decision-optimization capabilities are transforming supply chain management: enterprises using AI can reduce operational costs by 15–20%, boost resource utilization by 25%, and cut carbon emissions by 30% on average. Exploring the link between AI-driven supply chain optimization and sustainability is crucial for industry innovation and green transformation [2].

Research has established a comprehensive theoretical framework and accumulated extensive practical experience, leveraging technologies like edge computing, high-performance computing, and federated learning to reinvent supply chains across multiple dimensions. There is a growing emphasis on AI interpretability, privacy protection, and environmental impact. Despite the rapid development in theoretical, technical, and industrial practice, several common gaps persist, such as insufficient interdisciplinary integration and a scarcity of empirical investigations, which collectively form the research background preceding the research questions [3].

This study employs multiple research methods to analyze the relationship between AI-driven supply chain optimization and sustainable development. It systematically reviews global theoretical frameworks via literature analysis, uses case studies such as JD.com’s drone delivery to dissect AI applications, and integrates quantitative (2019–2023 data on efficiency, costs, and emissions) and qualitative methods (expert interviews/questionnaires). Finally, a theoretical model is constructed to explore the intrinsic mechanisms between AI and supply chain sustainability, enhancing comprehensiveness and research depth. This study provides a new perspective for understanding the role of artificial intelligence in supply chain management and sustainable development, and provides reference for relevant policy formulation and enterprise practice.

2. The application of artificial intelligence in supply chain optimization

2.1. Supply chain optimization model driven by artificial intelligence

This model is the core of modern supply chain management. It integrates technologies such as big data and machine learning to build an intelligent decision-making system, which includes models such as predictive demand planning, intelligent inventory optimization, and multi-objective network optimization. In 2023, the proportion of Chinese enterprises using artificial intelligence to optimize the supply chain reached 38.6%, a significant increase compared to 22.3% in 2019. These enterprises have, on average, reduced operating costs by 15.7%, shortened the logistics cycle by 23.4%, and increased the prediction accuracy rate by 31.2% [4]. Among them, deep learning is used for demand forecasting, reinforcement learning optimizes inventory, and genetic algorithms, etc., achieve the optimal allocation of resources for multiple objectives, providing efficient and sustainable solutions.

2.2. Applications of artificial intelligence in various supply chain links

Artificial intelligence is deeply integrated into various supply chain links, driving the optimization of the value network. In the procurement process, intelligent systems analyze data through machine learning to achieve intelligent supplier evaluation. In 2022, 78% of large global enterprises applied AI in procurement, reducing costs by 15%-22%. In production, predictive maintenance systems reduce equipment failures. In 2023, the equipment failure rate in smart factories decreased by about 35%. In inventory management, AI algorithms optimize inventory. From 2021 to 2023, the inventory turnover rate of retail enterprises increased by 24%, and inventory costs decreased by 19%. In logistics and distribution, with the help of technologies such as intelligent route planning, the distribution mileage in 2023 decreased by 15%-30%, and carbon emissions were reduced [5]. In after-sales service, intelligent customer service shortens the time to solve customer problems by over 40%, jointly building an intelligent supply chain ecosystem.

2.3. Advantages and challenges of supply chain optimization by artificial intelligence

Artificial intelligence has significant advantages in supply chain optimization, enabling accurate prediction and real-time decision-making Enterprises that adopt green packaging have an average reduction of 12% in their carbon footprint, an 8% drop in packaging costs, a 30% increase in resource utilization through the circular supply chain model, and a 40% reduction in waste emissions

However, challenges coexist. Data quality and privacy issues are prominent, and 64.3% of Chinese enterprises are troubled by data quality [6]. Technical barriers and talent shortages are obvious, with the supply-demand ratio of AI talents reaching 1:10. Ethical governance is a concern, and the “black box”（unclear route optimization logic in traditional supply chains）of AI and algorithmic bias have drawn attention. There is insufficient synergy between technology and business models, and only 35% of AI projects can successfully achieve large-scale application.

3. Analysis of supply chain sustainability driven by artificial intelligence

3.1. Mechanisms of artificial intelligence promoting supply chain sustainability

3.1.1. Economic sustainability

AI applications in supply chain management significantly enhance economic sustainability.

Through predictive analysis and optimization algorithms, enterprises can more accurately anticipate market demand, reducing inventory overstocking and shortages. A McKinsey’s 2023 report states AI - enabled supply chains can cut inventory costs by 15%-25% and improve service levels by 5%-10%. In terms of operational efficiency, AI - driven automated decision - making and process optimization increase supply chain response speed by about 30%, reduce human intervention, and save labor costs. Financially, AI optimizes pricing strategies and supplier selection, generating direct economic benefits [7]. A 2022 Deloitte study shows that AI applications in the global supply chain bring enterprises over $40 billion in direct economic value annually, expected to reach $100 billion by 2025. AI - assisted risk management systems can identify supply chain disruption risks in advance. During the 2020 - 2022 global supply chain crisis, enterprises using such systems recovered 40% faster than traditional ones, ensuring business continuity and long - term profitability.

3.1.2. Environmental sustainability

AI has great potential in promoting environmental sustainability in the supply chain. AI driven route optimization algorithms can significantly reduce carbon emissions in logistics transportation. A 2023 World Economic Forum report shows that intelligent route planning systems can cut fuel consumption in logistics by 15%-20%, directly reducing carbon emissions. AI-supported resource optimization also greatly improves energy efficiency. Data from the IBM Environmental Research Institute indicates that applying AI to supply chain energy management can save 10%-15% of energy consumption. In waste management, AI accurately predicts product demand, reducing overproduction and inventory [8].

In 2022, the China Circular Economy Association reported that retail enterprises using AI prediction technology reduced product waste by about 18%. AI also plays a crucial role in material recycling. AI-enabled intelligent identification and classification systems enhance waste recycling efficiency. A 2021 Greenpeace survey shows that AI - assisted recycling systems can increase material recovery rates by 25%, reducing resource consumption and environmental pollution.

3.1.3. Social sustainability

AI driven supply chain optimization deeply impacts social sustainability and transforms work patterns. AI reduces workers' repetitive labor and improves work safety. A 2022 International Labour Organization report shows that after AI is applied in the supply chain, the accident rate in hazardous work environments decreases by 22% [9]. It also creates more high - tech jobs. Although some traditional jobs are automated, new positions such as AI system maintenance and data analysis have increased by about 15%. Regarding social equity, AI enables small and medium-sized enterprises to compete with large enterprises. According to the 2023 data from the China E-commerce Research Center, SMEs’ market competitiveness increases by about 20% after adopting AI - based supply chain management systems, narrowing the gap with large enterprises. AI applications also enhance supply chain transparency, especially in product origin tracking and labor condition monitoring. A 2021 printed wiring card survey shows that transparent supply chain systems combining block chain and AI reduce enterprises’ non-compliance risks by about 30%, promoting human rights protection and fair trade in the supply chain.

4. Case analysis of supply chain optimization and sustainability driven by artificial intelligence

4.1. Case analysis of the manufacturing industry

4.1.1. Introduction to the enterprise background

The object of this case analysis is Haier Group, a leading home appliance manufacturing enterprise in China. Established in 1984, after nearly four decades of development, it has become a globally leading provider of home appliances and smart home solutions. In 2022, the operating revenue of Haier Smart Home reached 228.529 billion yuan, with a year-on-year growth of 4.25%. Despite the complex and changeable global economic environment, it has maintained steady growth and actively promoted digital transformation. According to the data of the China Household Electrical Appliances Association, when the global supply chain faced challenges from 2022 to 2023, the COSMOPlat industrial Internet platform independently developed by Haier has been the key for the enterprise to deal with challenges of complexity and sustainability [10]. Haier’s supply chain network covers more than 160 countries and regions around the world, controls over 10,000 suppliers, and processes millions of logistics and production data every day. In 2021, under the carbon neutrality strategy, Haier put forward the “Green Intelligent Manufacturing” plan to build a sustainable intelligent manufacturing ecosystem.

4.1.2. Application of artificial intelligence in supply chain optimization

Haier Group comprehensively applies artificial intelligence technology in supply chain optimization, forming a unique “Smart Supply Chain Ecosystem”. Firstly, it constructs an intelligent demand forecasting system based on deep learning. This system can integrate different types of data from various sources, including historical sales data, social media sentiment analysis, search engine trends, and weather information. Its prediction accuracy is 32% higher than that of traditional methods. During the period from 2020 to 2023, this system helps Haier increase its inventory turnover rate by 18.7% and reduce inventory costs by 15.3% [11]. Then, it develops an AI-driven green supplier evaluation system. With the help of machine learning algorithms, it can monitor and evaluate suppliers’ environmental performance, resource utilization efficiency, and carbon emissions in real-time. By using blockchain technology, it ensures data traceability and transparency, thus establishing a complete green supply chain system. Moreover, Haier uses its self-developed COSMOPlat intelligent logistics system. After integrating robotic warehousing, intelligent path planning, and driverless technologies, the logistics distribution efficiency isincreased by 25% and energy consumption is reduced by 20%. The Smart Logistics Cloud Brain project implemented in 2022 is particularly remarkable. Relying on AI algorithms to optimize the warehousing layout and distribution routes, it can reduce carbon emissions by approximately 28,000 tons per year.

4.1.3. Assessment of sustainability effects

Haier Group has achieved remarkable results in sustainability thanks to the optimization of its supply chain driven by artificial intelligence. In terms of the environment, after being optimized by AI, the intelligent energy management system has increased the efficiency of electricity consumption in Haier’s production by 21.3%. From 2019 to 2023, it has cumulatively reduced carbon dioxide emissions by approximately 960,000 tons. Moreover, the intelligent supply chain planning system has optimized transportation routes, leading to a 28.5% reduction in carbon emissions in the logistics process. In terms of the economy, Haier’s AI supply chain management system has reduced the overall operating cost by 17.2%, improved the supply chain response speed by 34%, and shortened the product delivery cycle by 42%. It is particularly important that Haier's COSMOPlat platform has utilized AI technology to empower more than 2,300 small and medium-sized suppliers, helping them enhance their production efficiency and quality control capabilities. In terms of society, Haier’s AI supply chain transparency system has ensured the protection of labor rights and fair trade practices. With the intelligent monitoring system, the workplace safety accident rate has decreased by 31%. Haier’s 2023 sustainable development report shows that its AI supply chain optimization project has obtained multiple international certifications, such as the ISO14001 environmental management system certification and the ISO50001 energy management system certification, thus becoming a model case of sustainable development in the manufacturing industry [12].

5. Challenges and policy recommendations

Data security and privacy risks are significant concerns for Chinese enterprises, with 64.3% facing issues related to data quality. The "black box" feature of AI may lead to algorithmic bias and information leakage. Additionally, technical barriers and talent shortages also pose challenges. The supply-demand ratio of AI talents reaches 1:10. Small and medium-sized enterprises find it difficult to afford technological investment, and only 35% of AI projects have been applied on a large scale. Finally, there is a lack of ethical governance in AI applications. The opacity of algorithms raises questions about fairness, such as implicit discrimination in supply chain decision-making.

To address these challenges, it is crucial to improve the data governance framework. This involves establishing data security standards specific to the supply chain and promoting the application of blockchain technology to enhance transparency. For instance, Haier's green supply chain traceability model exemplifies how such measures can be implemented effectively.

Strengthening technological inclusiveness and talent cultivation is also essential. The government takes the lead in building an AI technology sharing platform and sets up special subsidies for small and medium-sized enterprises. Additionally, universities have added interdisciplinary courses on "AI + Supply Chain" to better prepare the workforce.

Establishing an ethical review mechanism is another important step. By referencing the EU's "Artificial Intelligence Act", enterprises could be required to disclose algorithmic logic, and a supply chain AI ethical assessment committee could be formed to oversee compliance and fairness.

Finally, promoting interdisciplinary research and international cooperation will be vital for advancing these initiatives Funding for multi-field joint research projects can facilitate collaboration, and drawing on mature overseas theoretical systems, such as experiences in edge computing application, can expand the dimensions of sustainability research.

6. Conclusion

This study systematically explores various aspects of artificial intelligence (AI) in promoting supply chain optimization and sustainable development. In the context of growing complexity in global supply chains, evolving consumer demands, and climatic transformations. the research finds that AI technology has already become a key innovative force in supply chain management. Through technical approaches such as multi-data fusion prediction, selection of green suppliers, and drone delivery, AI significantly improves supply chain efficiency, greatly reduces operating costs, and minimizes environmental impact. Moreover, AI-driven supply chain optimization achieves excellent environmental benefits in optimizing energy optimization and carbon monitoring, and it can also empower small and medium-sized enterprises, generating positive social-economic impacts. However, challenges in data security, ethics, and technical barriers require improved governance, ethical review, and technology adoption. In the future, with the development of technology and policy support, AI will more deeply promote the sustainability of the supply chain, achieving a harmonious unity of economic benefits, environmental responsibilities, and social responsibilities. The research results can provide important decision-making references for corporate practitioners and policymakers, and also offer new perspectives for the academic community to understand the intersection of artificial intelligence, supply chain management, and sustainable development.

However, this study mainly sorts out the current research status, but lacks in-depth criticism of the conflicts and consensus among different theoretical frameworks. For instance, the applicability differences between edge computing and federated learning in supply chain applications were not compared, nor were the limitations of the existing sustainability assessment models pointed out. It only remained at the descriptive summary level.

Looking ahead, future research should strengthen basic theoretical innovation, deepen interdisciplinary integration, improve empirical methods, and pay attention to issues such as ethics, data security, and sustainable development. Furthemore, it is necessary to construct interdisciplinary theories, strengthen empirical research, and expand the dimensions of sustainability research.