An adaptive agent based model for improving agility indicators in JIT manufacturing systems

Authored by S. M. Seyed-Hosseini, M. R. Gholamian, A. R. Movassagh

Date Published: 2011-12

Sponsors: No sponsors listed

Platforms: No platforms listed

Model Documentation: Other Narrative Flow charts Mathematical description

Model Code URLs: Model code not found

Abstract

One of the major problems in JIT manufacturing systems is decreasing agility indicators, such as throughput or mean of waiting time, when confronting large fluctuations in average demand. To solve this problem, this article proposes an adaptive model to adjust the number of kanban by means of intelligent agent technology. In this way, three kinds of agents are introduced. These agents are Monitoring Agents (MA) that are assigned to each indicator, Fluctuation Detector Agents (FDA) that detect demand fluctuation from the environment, and Kanban Adjuster Agents (KAA), that make decisions about kanban numbers. In each iteration, these agents try to improve Total Performance (TP) of the system by cooperation and negotiation wherever needed. Application of the model is analyzed by simulation. Results show that the improvement in agility indicators derived from the new model is tremendous, especially when it is compared with the present kanban-based system.

Tags

Multi-agent systems agility just in time kanban