A robust adaptive scheduler for an intelligent workstation controller

Abstract

A shop floor control system (SFCS) consisting of three hierarchical control levels (shop, workstation, and equipment) is described. Each controller plans, schedules, and executes the activities necessary to process an order. An intelligent workstation controller (IWC), which is a part of the SFCS, is described in detail. The 1WC receives information such as part type and quantity, part routeing specifications, and process plans from the shop level controller and coordinates production activities. The IWC performs three main functions—planning, scheduling, and execution in real-time in order to ensure completion of jobs assigned by the shop controller. The focus of this paper is to develop a robust adaptive scheduler to support the IWC which fits within the functional SFCS architecture. The objectives of this paper are: (1) to develop a neural network model that generates several part dispatching strategies based on workstation status; (2) to develop a mutti-pass simulator that evaluates the generated strategies and selects the best strategy to maximize system efficiency; and (3) to compare the efficiency of the scheduling function with other single-pass strategies with respect to several performance criteria.