How Fujitsu Optimises Supply Chains with Multi-AI Agents

Fujitsu Limited has developed a new collaborative multi-AI agent technology in order to optimise supply chain efficiency.

Through this technology, it is securely connecting agents across multiple companies, helping businesses react quickly to changing environments.

Joint trials will soon be taking place, putting the security and agility of these agents to the test, with the aims of expanding into new industries.

Supply chain agility

Fujitsu is a Japanese multinational information and communications technology company. It has established itself the digital transformation partner of choice for many customers around the world, by offering a range of services and solutions.

It has a focus on five key technologies: AI, computing, networks, data & security and converging technologies. 

Fujitsu has now announced the development of a new technology which encourages secure collaboration within the supply chain. Its new technology consists of a multi-AI agent collaboration tool which allows companies to respond quickly to changing circumstances. 

It also considers changes in other AI agents from other companies within a supply chain.

This helps businesses stay agile and protects their supply chain from disruption. If these agents can track changes early on the supply chain, business leaders can make the appropriate changes in order to mitigate risk.

Fujitsu will begin field trials of this technology in January, aiming to help increase supply chain optimisation of Rohto Pharmaceutical Co, in collaboration with the Institute of Science Tokyo (Science Tokyo).

The multi-agent technology will allow for the streamlining of regular daily supply chain operations, but its strongest asset is its ability to facilitate fast recovery during demand shifts or external disasters which impact the business.

Risk mitigation

After trials with Rohto Pharmaceutical, Fujitsu will continue advancing its technological enhancement, with the aim of expanding into industries like manufacturing. It will also work to implement technology for more complex supply chains, meaning that every business will be able to mitigate risk through its multi-AI agents. 

Through this, the company will help with global supply chain transformation. Its technology can enhance supply chain resilience, sustainable business operations and could transform corporate supply chain strategies. With its Uvance business model, Fujitsu's technology can implement secure data collaboration, with AI agent cooperation across industries and borders. 

As a result, businesses can see more resilience, greater reliability and governance in multi-vendor environments and sustainable industrial growth.

“Science Tokyo is actively promoting Cyber-Physical Systems (CPS) research and working to improve efficiency across the entire industrial value chain," explains Katsuki Fujisawa, Professor, Digital Twin Research Unit at Institute of Science Tokyo.

"Moving forward, by collaborating with Fujitsu's agentic AI technology to optimise the entire supply chain, we aim to contribute to the advancement of industry and the resolution of societal challenges.”

Fujitsu technology

The new technology features two key components: global optimal control for AI agents and a secure inter-agent gateway.

Optimal control
This technology approximates the preferred conditions for a partner company using limited information and can quickly identify the overall optimal state for the entire supply chain. It enables effective collaboration between various agents from different companies, without the need of confidential or sensitive data.

Fujitsu secure inter-agent gateway
This gateway allows for seamless and secure collaboration among external AI agents, protecting confidential information while providing distributed AI training. Throughout setup, the AI agents can learn supply chain characteristics by leveraging knowledge distillation - data is drawn from multiple 'teacher models' into a 'student model'. 

While being used, this model is able to uncover malicious attacks and prevents the spread of confidential information. It repeatedly simulates AI agent behaviour and updates information in a safe format.