A supply chain as a network of auctions

This paper proposes and evaluates a model of supply chains as networks of auctions. In this model, companies are represented according to the first level of the Supply Chain Council’s SCOR model and the trading strategy of the agents is adapted from a model proposed by Steiglitz and colleagues. Specifically, the highest level of SCOR treats a company as comprising three functions, namely Source, Make, and Deliver. Our companies may also have these three functions, where Source and Deliver are modelled in a way similar to the agents proposed by Steiglitz and his colleagues, that is, they hold products in their inventory and place shouts to buy or sell products so that their inventory remains at its target level. After presenting our model, we study its behaviour with a simulation based on the JASA auction simulator. This study provides insights into supply chain dynamics and associated trading behaviours. In particular, the simulation results show that price dynamics are more complicated than simply balancing consumption, transformation capacities, and the supply of raw materials. In addition, we identify three patterns of price dynamics in our auctions, explain their cause, and propose rules linking initial conditions and the occurrence of such price patterns.