Chih-Chung Lo

A novel approach for vector quantization using a neural network, mean shift, and principal component analysis-based seed re-initialization

In this paper, a hybrid approach for vector quantization (VQ) is proposed for obtaining the better codebook. It is modified and improved based on the centroid neural network adaptive resonance theory (CNN-ART) and the enhanced Linde-Buzo-Gray (LBG) approaches to obtain the optimal solution. Three modules, a neural net (NN)-based clustering, a mean shift (MS)-based refinement, and a principal component analysis (PCA)-based seed re-initialization, are repeatedly utilized in this study. Basically, the seed re-initialization module generates a new initial codebook to replace the low-utilized codewords during the iteration. The NN-based clustering module clusters the training vectors using a competitive learning approach. The clustered results are refined using the mean shift operation. Some experiments in image compression applications were conducted to show the effectiveness of the proposed approach.

Using Ant Colony Optimization Algorithm to Solve Airline Crew Scheduling Problems

In this paper, an ant colony optimization (ACO) based ant crew scheduling model (ACSM) is proposed to solve airline crew scheduling problems. In the proposed ACSM, airline crew scheduling problems are first formulated as traveling salesman problems with flight graph representation. Then, the ACO algorithm is applied to search near-optimal solutions for airline crew schedules. The validity of the proposed ACSM is verified by implementing it in real study cases. The results from the implementation and evaluation confirm that the proposed ACSM is suitable for the airline crew scheduling problems with good performance.

Boundary-based corner detection using K-cosine

This study presents a boundary-based corner detection method that achieves robust detection for digital objects containing wide angles and various curves using curvature. The boundary of an object is first represented into curvature measured by K-cosine. Then, by modifying the corner detection error (J.L. Ker, 1989), this study proposes a suitable K value and curvature threshold for robust corner detection. Finally, the K-cosine corner detection (KCD) algorithm is proposed and verified with several commonly employed digital objects. The experimental results reveal that the proposed method is free from translation, rotation and scaling, and is superior to Tsais method (D.M. Tsai et al., 1999) in computation speed in discriminating false targets.

Artificial immune systems for intelligent nurse rostering

Nurse rostering is an essential and important task for hospital administration. In this paper, an intelligent nurse rostering system is proposed by using two types of artificial immune systems, CLONALG and aiNet, as the intelligent mechanisms. The performance of both artificial immune systems is examined, and the results indicate that both artificial immune systems provide good intelligent solutions to solve nurse rostering problems efficiently and effectively. In addition, it is also found that CLONALG is a more efficient intelligent nurse rostering mechanism when time-consuming what-if analysis is frequently needed.

Supplier selection using rough set theory

The purpose of this study was to build a model of supplier selection to improve organizational capability and competitiveness, as well as to apply the model to solve practical problems. The critical criteria for supplier evaluation were chosen and the questionnaire was developed after literature reviewing. The questionnaire differentiates class 1 (excellent firms), class 2 (common firms), and class 3 (disappointed firms) from suppliers to be evaluated by participants. Next step was to use rough set theory (RST) to analyze the rules of supplier selection. After attribute reduct and core were derived, the decision-making rules were created by the supplier selection model. Rough set theory was adopted as main analysis method for enterprises to find the optimum supplier partners quickly and accurately in designing and organizing of supply chain.

A Novel Approach for VQ Using a Neural Network, Mean Shift, and Principal Component Analysis

In this paper, a hybrid approach for vector quantization (VQ) is proposed for obtaining the better codebook. It is modified and improved based on the centroid neural network adaptive resonance theory (CNN-ART) and the enhanced LBG (Linde-Buzo-Gray) approaches. Three modules, a neuronal net (NN) based clustering, a mean shift (MS) based refinement, and a principal component analysis (PCA) based seed assignment, are repeatedly utilized. Basically, the seed assignment module generates a new initial codebook to replace the low utilized codewords during the iteration. The NN-based clustering module clusters the training vectors using a competitive learning approach. The clustered results are refined using the mean shift operation. Some experiments in image compression applications were conducted to show the effectiveness of the proposed approach

Applying Particle Swarm Optimization algorithm to roundness measurement

Roundness measurement has been a crucial issue of quality control in industry. This study proposes a roundness measuring method that applies the particle swarm optimization algorithm (PSO) for computing the maximum inscribed circle (MIC). We first conducted a design of experiment with five PSO models, in which the impact of inertia weight, maximum velocity and the number of particles on the performance of the particle swarm optimizer was analyzed. Then, a simple experiment of verification was conducted, and the experimental results reveal that the PSO-based method performed well in both accuracy and the efficiency.

A two-phased evolutionary approach for intelligent task assignment & scheduling

In this paper, a two-phased evolutionary approach that integrates particle swarm optimization (PSO) and genetic algorithm (GA) is proposed to help assigning and scheduling maintenance crews to maintenance tasks automatically. The validity of the proposed system is verified experimentally. Results from the experiments indicate that the two-phased integration of PSO and GA in the proposed system is a robust and effective intelligent planning mechanism to help make decisions for task assignment and scheduling with higher efficiency.

A SOA-based intelligent system for nurse rostering

Currently the crisis in hospital management is forcing hospital executives to operate their organizations in a more business-like manner. Nurse rostering problem (NRP) is an important on-going staff scheduling problem with multiple decision criteria to be considered in order to provide a high-quality healthcare service, to which todays hospital administrations are paying great attention. In this research, the design of an intelligent decision support system, based on guidelines of Service-Oriented Architectures (SOA), is proposed to help solve nurse rostering problems with high flexibility, efficiency and effectiveness. The design uses three evolutionary computation algorithms (AIS, GA, and PSO) as exchangeable intelligent planning and scheduling mechanisms for rostering nursing staffs.