The Effect of the Application of the Silver-Meal Method in the Aggregated Supply Chain, Analysis Using a Multi-agent System
Nominated for The Planning for Sustainable Land Use and Transport Award
Mateo Pachón Rincón, Universidad Nacional De Colombia, Jorge Eduardo Ortiz Trivino, Universidad Nacional De Colombia
This paper proposes agent-based modeling as a tool to analyze the behavior of the supply chain. The multiagent system begins with a study of the supply chain, and an analysis of the different logistic flows.
This paper proposes agent-based modeling as a tool to analyze the behavior of the supply chain of the bakery industry, regarding the adoption of inventory management policies, particularly the use of Silver-Meal heuristic method. A comparison of the characterized logistics system, taking into account the inventory policies of different agents interacting in the logistics network. On the other hand the impact of the adoption of individual policies in the aggregated supply chain is analyzed. In particular the behavior among raw material suppliers and small producers, who demand to elaborate products.
The multiagent system begins with a study of the supply chain, and an analysis of the different logistic flows, mainly information and physical flow. With this study the base line of the computational model is proposed. Afterwards the model is implemented via software, which has the characteristic of being configured in a way that allows comparison of policies and scenarios through simulation. In this way the ability to observe and analyze emergent behavior of the system arises. It also allows comparing the behavior of the supply chain by using the study data contrasted with the use of the Silver-Meal heuristic method that seeks to minimize the costs of ordering and holding.
The construction of agent-based model and analysis of the data produced by it are used to show the impact of adopting the method of Silver-Meal in the aggregated supply chain and small producers individually. That method is configured in the multi-agent system and compared with the actually observed in the supply chain. Moreover the different model configurations based agents provide information about the degree of coordination. Analysis was also performed near the raised configurations and its impact on the management of the supply chain in the aggregate level.
The agent-based modeling applied to logistics systems and specifically to supply chain and provides a dynamic view that is appropriate when: the control of the same cannot be performed centrally, when the environment is dynamic and complex. Furthermore it is concluded that the multiagents systems work well to compare inventory policies, and also provides some insights of the implementation in the real supply chain.
Association for European Transport