AI can unlock sustainable value chains for Vietnam manufacturing

The promise of AI-powered circularity

Industries such as electronics, textiles, logistics, and packaging in Vietnam have long relied on linear “take-make-dispose” models, i.e. resources are extracted, made into products, and then disposed as waste. However, such models are no longer sustainable. Researchers have pointed to circular economy principles, which keep products and materials in use for as long as possible, as a future-proof alternative.

According to Associate Professor Pham Cong Hiep, Deputy Dean of Research and Innovation at The Business School, RMIT University Vietnam, artificial intelligence (AI) is emerging as a key enabler to make the circular economy practical and scalable.

“AI does far more than improving operations – it transforms how we design, manufacture, and recover value in supply chains,” he says.

AI can be used in smart design automation. With generative AI design and machine learning, engineers can develop modular, lightweight, and repairable products tailored for circularity. They can generate thousands of product configurations using a single software program. This reduces material use while supporting disassembly and reuse in future product cycles. In Vietnam’s textile industry, AI tools are already being explored to design clothing patterns that minimise fabric offcuts, supporting zero-waste manufacturing. 

Real-time process optimisation can also benefit from AI. Predictive analytics and AI systems with Internet of Things (IoT) help manufacturers monitor machine health, energy use, and production efficiency. Siemens MindSphere, for example, connects industrial equipment to the cloud and uses AI-driven analytics to detect anomalies, forecast equipment failures, and recommend proactive maintenance schedules. In a Vietnamese garment factory, such a system could analyse data from sewing machines in real time, predict wear-and-tear, and schedule servicing to prevent unexpected downtime.

Finally, AI-powered vision systems and robotics are transforming how materials are identified, sorted, and recovered in recycling operations, thus maximising value recovery. AMP Robotics, a leading innovator in this space, uses AI-enabled robotic arms to recognise and sort recyclable materials from waste streams with up to 99 per cent accuracy. In electronics assembly plants, a similar system could scan returned devices, classify parts by condition, and automate disassembly for reuse or recycling. This approach accelerates material recovery and prevents valuable resources like rare earth metals from ending up in landfills, helping manufacturers close the loop in their supply chains. 

“These capabilities aren’t theoretical anymore,” Associate Professor Hiep notes. “They’re already reshaping global supply chains and they’re within reach for Vietnamese firms, too.”

Affordable cloud services and ready-made AI frameworks allow companies, including small and medium-sized enterprises, to deploy predictive models, image recognition, and optimisation tools with minimal setup.

AI-driven circularity also delivers measurable returns: fewer machine downtimes, reduced material waste, and new revenue streams from refurbished products.

“This return on investment often outweighs the initial deployment costs,” Associate Professor Hiep says.

How Vietnam’s manufacturing companies can act

Associate Professor Hiep outlines three key steps for manufacturing companies to start adopting AI.

1. Start with predictive and prescriptive applications

Vietnamese manufacturers can begin by applying AI to high-impact areas. For example, AI algorithms combined with sensor data can monitor the health of critical machinery in real time, predicting potential breakdowns before they occur. This not only minimises costly downtime but also extends equipment life and reduces the waste of prematurely discarded parts.

By starting with these practical applications, businesses can achieve quick gains in operational efficiency and build confidence for broader AI integration.

2. Combine AI with data and sensors

Circularity requires visibility across the entire supply chain, so businesses should equip their products and production lines with IoT sensors that generate high-quality data for AI systems. These sensors track usage patterns, component wear, and environmental conditions, enabling AI models to recommend timely repairs, upgrades, or recycling decisions.

For example, a textile factory could use RFID tags and AI-powered image recognition to trace fabric waste across production stages, identifying where offcuts can be reused or resold. In mechanical parts manufacturing, AI can use sensor data to suggest process adjustments that reduce material consumption without compromising quality.

“This is where the magic happens,” says Associate Professor Hiep. “The more data you feed your AI systems, the smarter they become and the more value they unlock in your supply chain.”

3. Redesign for circular business models

Finally, AI adoption for circularity is not just a technological shift; it requires a cultural and leadership transformation within businesses. Leaders must cultivate an innovative mindset across teams, empowering employees to experiment with AI tools and reimagine traditional workflows. Managers need to align sustainability goals with business strategies, fostering collaboration between technical and non-technical staff.

Associate Professor Hiep highlights, “For AI to drive circular success, companies must invest in human capital, develop digital capabilities, and create an organisational culture where innovation and sustainability are seen as shared responsibilities.”

Story: Linh Do

Masthead/thumbnail image: Samon – stock.adobe.com