A Genetic Fuzzy System for Defining a Reactive Dispatching Rule for AGVs

In the last decades automated guided vehicles (AGVs) have been widely used as automated material handling system in flexible manufacturing system (FMS) aiming to improve the use of factory's resources, increase the productivity, decrease the lead time and maintain the delivery dates. However the definition of which AGV will attend which request is not an easy task. In this work, the authors propose a fuzzy system to decide which job the AGV will attend first based on some specific criteria. The fuzzy system is built automatically from numerical data by a genetic algorithm. The proposed approach was tested and validated by computer simulations in a FMS.

1 Introduction

Flexible Manufacturing System (FMS) have received increasing attention in recent decades. The increasing demand for low cost, low-to-medium volume production of different type of goods creates the need for production flexible systems having low setup times and mainly characterized by many degrees of freedom in the decision making process. A FMS is a production system consisting of multiple controlled heterogeneous resources such as workstations, automated guided vehicles (AGVs), robots, load and unload stations, warehouse, and a control system [1] [2].