Mathematical Decomposition and Simulation in Real-Time Production Scheduling

Published

January 1, 1987

Author(s)

W Davis, Albert T. Jones

Abstract

This paper discusses an on-line, real-time production scheduling algorithm for automated manufacturing systems. Decomposition theory is used to transform a multi-criteria, production scheduling problem from a block angular structure into a two-level hierarchical structure. The top level, called the supremal, considers a list of jobs, due dates, precedence constraints, and objectives. It generates a set of potential scheduling rules and evaluates those rules using an on-line, distributed simulation package. The supremal outputs a list of tasks with proposed start and finish times to each of the lower level systems under its control. Each lower level system, called on infimal, then uses a similar simulation approach to sequence those tasks and generate actual start and finish times. These times, together with status on all other tasks, provide the feedback needed by the supremal to close the control loop.

Citation

NIST Interagency/Internal Report (NISTIR) - 3639

Report Number

3639

NIST Pub Series

NIST Interagency/Internal Report (NISTIR)

Pub Type

NIST Pubs

Keywords

mathemeatical decomposition, simulation, real-time production, scheduling, automated manufacturing

Citation

Davis, W. and Jones, A. (1987), Mathematical Decomposition and Simulation in Real-Time Production Scheduling, NIST Interagency/Internal Report (NISTIR), National Institute of Standards and Technology, Gaithersburg, MD (Accessed October 11, 2024)