The changing role of information technology in food and beverage logistics management: beverage network optimisation using intelligent agent technology

The last decades, advances in information technologies and increased competition have changed the business environment in the food and beverages industry, particularly in the European Union, which is characterised by the proliferation of small and medium enterprises. Many food companies are now aggressively focusing on logistics management as the last frontier to gain and sustain a competitive advantage. This study describes a model of intelligent food supply chain that improves efficiency within the supply chain. The aim of the paper is to demonstrate that agent technology can optimise food supply chains by (a) reviewing intelligent agents applications for supply chain optimisation and (b) illustrating how a multi-agent system can optimise performance of a beverage logistics network. Firstly, we review and synthesise existing applications in comparison to traditional and Internet-based technologies and critically evaluate agent technology applicability for supply chain management. We model the beer supply network to demonstrate that products can acquire intelligence to direct themselves throughout the distribution network. Optimisation agents can help solve specific problems of beverage supply: reduce inventories and lessen bullwhip effect, improve communication, and enable chain coordination without adverse risk sharing. Further, they gain a capability to be purchased and sold while in transit. Overviews of the supporting technologies that make such a supply chain a reality are fully discussed. In particular, optimisation agents have the characteristics of autonomous action, being proactive, reactive, and able to communicate. We demonstrate that agents enhance the flexibility, information visibility, and efficiency of the supply chain management. Suggestions and recommendations for further research are provided. Simulations of the agent-enabled supply optimisation can be used to benchmark for future research and development associated with building an optimisation agent.