|Law, Ko Wah Edward
|Modelling a production workflow using interaction matrix, ant colony and information entropy optimization
|Yung, Winco (ISE)
|Hong Kong Polytechnic University -- Dissertations
Production management -- Mathematical models
Workflow -- Management
|Faculty of Engineering
|iii, iii, xii, 205 pages : color illustrations
|Over the last two decades, there has been a paradigm shift in product manufacturing to one that can serve diversified and volatile customer needs. It becomes a trend to introduce increasingly complex production line configurations to meet the individual demand of customers. Both planning and production people do not like complexities, as it needs additional information to understand the exact assembly situation. The primary purpose of this project is to develop a mathematical model and methodology to minimize the complexity (information entropy) of manufacturing systems in various configurations generated by production workflow. The manufacturing system to be examined in this research is a mixed-model assembly where the same production stations can be used to produce various product models at the same time. The methodology is derived from the traditional method of line balancing applied to mixed-model production, but the complexity aspects are quantified and optimized. Line balancing methods as reviewed in the literature, only consider tangible factors such as cycle time or number of work terminals. It induces complexities in the configuration of the terminals and also its execution. There are two types of complexities: (i) structural (planning), and (ii) dynamic (execution). Structural complexity is inherent in planning and occurs even if actual activities are exactly followed. In this project, an interaction matrix is adopted to consider the effect of interaction among tasks. Dynamic complexities occur real-time when the actual activity cannot follow that which is planned. Therefore, the greater the differences it creates, the more complex it becomes. Of course, the decision may also depend on historical records where some processes are likely to be delayed by external factors. Optimization for such combinational problems is time consuming and therefore a meta-heuristic methodology-ant colony optimization (ACO) is utilized to speed up the process. ACO is an optimization algorithm that simulates ant behavior when finding the shortest path in carrying food back to their nest. It has been proven to be successful when tackling various types of Non-Deterministic Polynomial-Time (NP) hard problems by reducing the computational time but still finding acceptable solutions. Furthermore, the mathematical model is programmed by the computer language Matlab; this is a high-performance language for technical computing. To illustrate the application of the model and its optimization in real life circumstances, processes and data from an electronic manufacturing company are analyzed for demonstration purposes. The scope of application is not only to demonstrate the effectiveness of the mathematical model in a factory, but also to illustrate that input data in the mathematical model can be captured from a typical manufacturing circumstance without special arrangement for its collection. To conclude, it is shown in the results that it is feasible to reduce the complexities of the production workflow through the framework developed. The findings of the case study also reveal that the data can be captured in normal operations without special arrangement. Even though the mathematical model and methodology are applied to line balancing in a manufacturing environment, it can also be extended to other types of process with hierarchy relationships among tasks (or activities).
|All rights reserved
Files in This Item:
|For All Users (off-campus access for PolyU Staff & Students only)
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: