Multi-Agent Based Beam Search for Real-Time Production Scheduling and Control

Multi-Agent Based Beam Search for Real-Time Production Scheduling and Control PDF

Author: Shu Gang Kang

Publisher: Springer Science & Business Media

Published: 2012-10-11

Total Pages: 136

ISBN-13: 1447145755

DOWNLOAD EBOOK →

The Multi-Agent Based Beam Search (MABBS) method systematically integrates four major requirements of manufacturing production - representation capability, solution quality, computation efficiency, and implementation difficulty - within a unified framework to deal with the many challenges of complex real-world production planning and scheduling problems. Multi-agent Based Beam Search for Real-time Production Scheduling and Control introduces this method, together with its software implementation and industrial applications. This book connects academic research with industrial practice, and develops a practical solution to production planning and scheduling problems. To simplify implementation, a reusable software platform is developed to build the MABBS method into a generic computation engine. This engine is integrated with a script language, called the Embedded Extensible Application Script Language (EXASL), to provide a flexible and straightforward approach to representing complex real-world problems. Adopting an in-depth yet engaging and clear approach, and avoiding confusing or complicated mathematics and formulas, this book presents simple heuristics and a user-friendly software platform for system modelling. The supporting industrial case studies provide key information for students, lecturers, and industry practitioners alike. Multi-agent Based Beam Search for Real-time Production Scheduling and Control offers insights into the complex nature of and a practical total solution to production planning and scheduling, and inspires further research and practice in this promising research area.

Multi-Agent Based Beam Search for Intelligent Production Planning and Scheduling

Multi-Agent Based Beam Search for Intelligent Production Planning and Scheduling PDF

Author: Shugang Kang

Publisher: Open Dissertation Press

Published: 2017-01-27

Total Pages:

ISBN-13: 9781361476208

DOWNLOAD EBOOK →

This dissertation, "Multi-agent Based Beam Search for Intelligent Production Planning and Scheduling" by Shugang, Kang, 康書剛, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled Multi-agent based beam search for intelligent production planning and scheduling submitted by KANG SHUGANG for the degree of Doctor of Philosophy at the University of Hong Kong in September 2007 To survive fierce global competition, manufacturing enterprises need to improve their responsiveness to market changes and the productivity of their manufacturing systems. Production planning and scheduling requirements have become more demanding than ever before. Due to the complexities of scheduling problems, the impact of scheduling research on real-world applications is limited, and much research work is needed to develop effective methods and tools that can be directly applied to the planning and scheduling functions of modern manufacturing systems. This study attempts to address this issue by incorporating the philosophies and techniques of multiple approaches into a unified framework for development of a powerful hybrid method, an open software platform, and subsequently an integrated system for production planning, scheduling, and execution. Hence, an effective total solution is developed for real-world applications. In the theoretical aspect, a Multi-Agent Based Beam Search (MABBS) method is developed by integrating AI (Multi-Agent System, Expert System) and OR (Beam Search) methods. Although the multi-agent system (MAS) is a promising technique for modeling complex dynamic systems, the solution quality of MAS-based method is usually questionable. The MABBS method can take advantage of the representation power of MAS, and enhances MAS with beam search method for improving solution quality. As such, it provides a powerful hybrid method to solve complex real-world problems. To simplify the use of the MABBS method, computer languages and an open software platform are developed. The Knowledge Markup Language (KML) is an XML based markup language to model intelligent agents; the Information Markup Language (IML) is an XML based markup language for rapid construction of graphical user interfaces (GUI) for user interaction; and the Embedded eXtensible Application Script Language (EXASL) is embedded into KML and IML to provide a simple and versatile programming language for both knowledge representation and user interaction. It also provides a flexible tool for configuration of complex systems. The software platform includes an inference engine, a smart client and a toolkit. The inference engine is responsible for the MABBS-based computation. It can exploit the parallel computation power of network-connected computers to reduce computation time. The smart client renders graphical user interfaces (GUI) with IML, which provides rich functionalities and enables remote access over the Internet. The toolkit provides an integrated development environment for both back-end knowledge representation and front-end GUI development. The languages and the software platform provide a reusable and flexible approach for the rapid construction of customized production planning and scheduling systems. To apply the MABBS method, the languages and the software platform to practical production planning and scheduling problems, an Integrated Production Planning, Scheduling, and Executing System (IPPSES) is developed. In this system, a multi-agent system is designed and the MABBS method is used for complex production planning and scheduling. A workflow-based execution system is developed

Generic Multi-Agent Reinforcement Learning Approach for Flexible Job-Shop Scheduling

Generic Multi-Agent Reinforcement Learning Approach for Flexible Job-Shop Scheduling PDF

Author: Schirin Bär

Publisher: Springer Nature

Published: 2022-10-01

Total Pages: 163

ISBN-13: 3658391790

DOWNLOAD EBOOK →

The production control of flexible manufacturing systems is a relevant component that must go along with the requirements of being flexible in terms of new product variants, new machine skills and reaction to unforeseen events during runtime. This work focuses on developing a reactive job-shop scheduling system for flexible and re-configurable manufacturing systems. Reinforcement Learning approaches are therefore investigated for the concept of multiple agents that control products including transportation and resource allocation.

Dynamic Scheduling in Large-scale Manufacturing Processing Systems Using Multi-agent Reinforcement Learning

Dynamic Scheduling in Large-scale Manufacturing Processing Systems Using Multi-agent Reinforcement Learning PDF

Author: Shuhui Qu

Publisher:

Published: 2019

Total Pages:

ISBN-13:

DOWNLOAD EBOOK →

Scheduling in manufacturing plays an essential role in building smart manufacturing from multiple points of view, including social, economic, and environmental. Optimal scheduling, or the allocation of jobs with different requirements for a manufacturing processing system to meet various objectives, has been discussed for several decades. However, advanced scheduling methods in modern processing systems have not significantly improved, nor have they been widely adopted by staff working on manufacturing production lines despite extensive research conducted into scheduling. Most traditional scheduling methods require statistical assumptions, which cannot support operations for a dynamic and stochastic modern processing system. In addition, most proposed scheduling methods are not sufficiently scalable for managing real-world, large-scale processing systems. To address these limitations, we focus on the dynamic scheduling approach, which involves scheduling real-time events in large-scale modern manufacturing systems, from a data-driven perspective. We implement reinforcement learning (RL) to learn adaptive, scalable, and optimal dynamic scheduling policies, since RL can learn the underlying processing system's patterns and adaptively make allocation decisions based on real-time job and server measurements. The direct application of existing RL methods on the scheduling problem in such large-scale processing systems is impractical and undesired due to the extremely high computational complexity of learning a good scheduling policy. This thesis presents a practical and systematic computational framework that integrates RL with existing expert knowledge at three levels: (1) System-level planning. The planning procedure characterizes the processing system by the nominal feasible region of the scheduling problem. (2) Algorithm-level design. The design of the algorithm in RL is carefully selected as the index-policy-based, multi-agent RL, significantly reducing control policy search complexity. (3) Learning-level demonstration. During the learning process of RL, the existing expert knowledge is used as a demonstration to increase search efficiency and stabilize the RL learning process. We conduct various experiments in both real factory scenarios and simulated environments to evaluate the performance of the framework on processing system scheduling problems. The effectiveness of the proposed index-policy-based, multi-agent reinforcement learning (MARL) method is evidenced by its performance over traditional dynamic scheduling methods, with a linear computational time complexity in regard to the number of machines and job classes.

Industrial Applications of Holonic and Multi-Agent Systems

Industrial Applications of Holonic and Multi-Agent Systems PDF

Author: Vladimír Mařík

Publisher: Springer

Published: 2015-08-10

Total Pages: 247

ISBN-13: 3319228676

DOWNLOAD EBOOK →

This book constitutes the refereed proceedings of the 7th International Conference on Industrial Applications of Holonic and Multi-Agent Systems, HoloMAS 2015, held in Valencia, Spain, in September 2015. The 19 revised full papers presented together with one invited talk were carefully reviewed and selected from 27 submissions. The papers are organized in the following topical sections: surveys, conceptual design and validation, digital factories and manufacturing control systems, ARUM: adaptive production management, and smart grids, complex networks and big data.

Lanterna Rally - Theme by Grace Themes