A modified particle swarm optimization for production planningproblems in the TFT Array process

Authors:
Yin-Yann Chen;James T. Lin
Affiliations:
Department of Industrial Management, National Formosa University, Yunlin 632, Taiwan, ROC;Department of Industrial Engineering and Engineering Management, National Tsing-Hua University, Hsinchu 300, Taiwan, ROC
Venue:
Expert Systems with Applications: An International Journal
Year:
2009

Citing 5
Cited 2

A combined model of network design and production/distribution planning for a supply network

Computers and Industrial Engineering - Supply chain management
Integrated process planning and scheduling with minimizing total tardiness in multi-plants supply chain

Computers and Industrial Engineering - Supply chain management
Comparison between Genetic Algorithms and Particle Swarm Optimization

EP '98 Proceedings of the 7th International Conference on Evolutionary Programming VII
A Lagrangean-based heuristic for multi-plant, multi-item, multi-period capacitated lot-sizing problems with inter-plant transfers

Computers and Operations Research
An algorithm for production planning in a flexible production system

Computers and Industrial Engineering - Special issue: Selected papers from the 30th international conference on computers; industrial engineering

Monthly streamflow forecasting based on improved support vector machine model

Expert Systems with Applications: An International Journal
Particle swarm optimization algorithm for the berth allocation problem

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	12.05

Visualization

Abstract

In this study, a practical production planning problem in the TFT (thin film transistor) Array process is introduced. Several researchers have referred to the capacitated production lot-sizing allocation problems as NP-Hard. Naturally, it is harder to solve the capacitated production allocation problem considering its practical characteristics and constraints, such as allocation problems among bottleneck machines, photo masks, and products with different re-entrant layers. In response to this, we proposed a novel variation of the particle swarm optimization (PSO) model called the modified PSO (MPSO), which is a binary PSO model with dynamic inertia weight and mutation mechanism. It improves some weaknesses as opposed to the original version of the PSO, including a propensity for obstruction near the optimal solution regions that hardly improve solution quality by fine tuning. In addition, it is converted to be able to solve the model of binary decision variables. In order to illustrate effectiveness, the traditional PSO (TPSO), genetic algorithm (GA), and the proposed MPSO are compared by application of the literature's well-known test problems as well as the practical production planning problem in the TFT Array process. Based on the results of the investigation, it can be concluded that the proposed MPSO is more effective than the other approaches in terms of superiority of solution and required CPU time.