UM E-Theses Collection (澳門大學電子學位論文庫)
Design structure matrix based intelligent process sequencing and clustering methods for networked manufacturing
English Abstract
Design Structure Matrix Based Intelligent Process Sequencing and Clustering Methods for Networked Manufacturing by Ma Yuan M-B2-5549-1 Thesis Supervisor: Dr. Yang Zhixin Department of Electromechanical Engineering, Faculty of Science and Technology Abstract Manufacturing execution system (MES) aims to managing and monitoring work-in-process on shop-floor. In modern manufacturing industries, the manufacturing information and material flow involves a large scale of process tasks that are embedded with rich dynamic data. In order to well plan and keep tracking of process executions, an intelligent informatics approach, called as the hierarchical design structure matrix (H-DSM), is proposed to address such NP-hard process sequencing issue. The hierarchical structure of the proposed H-DSM consists of three sub-modules for transforming the original DSM. Firstly, as a multi-criteria decision analysis, an Analytic Hierarchy Process (AHP) method is adopted to transform the original binary DSM into numerical matrix for accurate measurement of task relationships. Secondly, a partitioning analysis approach is proposed to generate a suitable decomposition scheme for large-scale project with coupled tasks set, which will be partitioned into several small-scale clusters. Finally, according to the obligate brood parasitic behavior of certain VI cuckoo species in combination with the Lévy flight behavior of birds, a cuckoo search (CS) based DSM optimization algorithm is proposed for optimizing the task sequencing in coupled blocks so as to negotiate and cooperate the interdependent tasks. The experiments in MES environment for real industry’s shop floor have verified that the proposed H-DSM approach could significantly improve the execution efficiency via reordering and clustering the tasks set. With witnessing rapid development of the industrial automation, the modern industrial model has entered Industry 4.0 era. In addition, the manufacturing process in Industry 4.0 era trends into in integration and intellectualization. How to quickly respond to market and diverse customer requirements has become a large challenge and a hot research topic. As a system of representing the integrated computational and physical capabilities as well as actuation to physical world, the cyber-physical system plays an essential role in the future manufacturing. In this paper, we firstly introduce a network based Bat-DSM kernelled cyber-physical system for technology managers, designers, engineers working in industry. The proposed cutting-edge cyber-physical system not only optimizes the information and task flow obtained via various smart sensors e.g. RFID, Wi-Fi, Bluetooth, etc., but also delivers the functional tasks to the corresponding subsidiaries in real-time via the proposed novel networked platform including Product Data Management (PDM), Enterprise Resource Planning (ERP), Manufacturing Execution System (MES), and Distributed Control system (DCS). A real-world experiment verifies that the proposed cutting-edge cyber-physical system, in particular, can fulfill the business demands towards agility significantly.
Issue Date
Ma, Yuan
Faculty of Science and Technology
Department of Electromechanical Engineering
Computer integrated manufacturing systems
Manufacturing processes -- Data processing
Electromechanical Engineering -- Department of Electromechanical Engineering

Yang Zhi Xin
Library URL
Files In This Item:
Full-text (Internet)
1/F Zone C