Yu-Chiun Chiou

Genetic clustering algorithms

Abstract This study employs genetic algorithms to solve clustering problems. Three models, SICM, STCM, CSPM, are developed according to different coding/decoding techniques. The effectiveness and efficiency of these models under varying problem sizes are analyzed in comparison to a conventional statistics clustering method (the agglomerative hierarchical clustering method). The results for small scale problems (10–50 objects) indicate that CSPM is the most effective but least efficient method, STCM is second most effective and efficient, SICM is least effective because of its long chromosome. The results for medium-to-large scale problems (50–200 objects) indicate that CSPM is still the most effective method. Furthermore, we have applied CSPM to solve an exemplified p -Median problem. The good results demonstrate that CSPM is usefully applicable.

Driver responses to green and red vehicular signal countdown displays: Safety and efficiency aspects

This study investigates the effects of green signal countdown display (GSCD) and red signal countdown display (RSCD) on driver behaviours, and thus on intersection safety and efficiency. Three driver responses to GSCD, including late-stopping ratio, dilemma zone and decision to cross, and three driver responses to RSCD, including early start ratio, start-up delay, and discharge headway are observed and analyzed. Results show that although GSCD can reduce late-stopping ratio, the dilemma zone is increased by about 28 m and the decision to cross will be more inconsistent among the approaching vehicles, creating a potential risk of rear-end crashes. Additionally, following the provision of a green countdown the number of vehicles ejecting to cross the intersection reduces. On the other hand, comparisons among four observation periods examining the effects of RSCD-before-RSCD, 1.5 months after-RSCD, 3.0 months after-RSCD and 4.5 months after-RSCD, show that although RSCD significantly reduces the early start ratios of the leading vehicles in various waiting areas, the ratios soon return to their before-RSCD levels, suggesting that RSCD does not significantly improve intersection safety over the longer term. However, RSCD effectively reduces start-up delay, saturated headway, and cumulative start-up delay at 4.5 months after-RSCD installation. Thus, RSCD enhances intersection efficiency. RSCD is clearly less controversial and more beneficial than GSCD.

A joint measurement of efficiency and effectiveness for non-storable commodities: Integrated data envelopment analysis approaches

Efficiency and effectiveness for non-storable commodities represent two distinct dimensions and a joint measurement of both is necessary to fully capture the overall performance. This paper proposes two novel integrated data envelopment analysis (IDEA) approaches, the integrated Charnes, Cooper and Rhodes (ICCR) and integrated Banker, Charnes and Cooper (IBCC) models, to jointly analyze the overall performance of non-storable commodities under constant and variable returns to scale technologies. The core logic of the proposed models is simultaneously determining the virtual multipliers associated with inputs, outputs, and consumption by additive specifications for technical efficiency and service effectiveness terms with equal weights. We show that both ICCR and IBCC models possess the essential properties of rationality, uniqueness, and benchmarking power. A case analysis also demonstrates that the proposed novel IDEA approaches have higher benchmarking power than the conventional separate DEA approaches. More generalized specifications of IDEA models with unequal weights are also elaborated.

Genetic fuzzy logic controller: an iterative evolution algorithm with new encoding method

Logic rules and membership functions are two key components of a fuzzy logic controller (FLC). If only one component is learned, the other one is often set subjectively thus can reduce the applicability of FLC. If both components are learned simultaneously, a very long chromosome is often needed thus may deteriorate the learning performance. To avoid these shortcomings, this paper employs genetic algorithms to learn both logic rules and membership functions sequentially. We propose a bi-level iterative evolution algorithm in selecting the logic rules and tuning the membership functions for a genetic fuzzy logic controller (GFLC). The upper level is to solve the composition of logic rules using the membership functions tuned by the lower level. The lower level is to determine the shape of membership functions using the logic rules learned from the upper level. We also propose a new encoding method for tuning the membership functions to overcome the problem of too many constraints. Our proposed GFLC model is compared with other similar GFLC, artificial neural network and fuzzy neural network models, which are trained and validated by the same examples with theoretical and field-observed car-following behaviors. The results reveal that our proposed GFLC has outperformed.

Modeling crash frequency and severity using multinomial-generalized Poisson model with error components

Since the factors contributing to crash frequency and severity usually differ, an integrated model under the multinomial generalized Poisson (MGP) architecture is proposed to analyze simultaneously crash frequency and severity--making estimation results increasingly efficient and useful. Considering the substitution pattern among severity levels and the shared error structure, four models are proposed and compared--the MGP model with or without error components (EMGP and MGP models, respectively) and two nested generalized Poisson models (NGP model). A case study based on accident data for Taiwans No. 1 Freeway is conducted. The results show that the EMGP model has the best goodness-of-fit and prediction accuracy indices. Additionally, estimation results show that factors contributing to crash frequency and severity differ markedly. Safety improvement strategies are proposed accordingly.

Modeling two-vehicle crash severity by a bivariate generalized ordered probit approach

This study simultaneously models crash severity of both parties in two-vehicle accidents at signalized intersections in Taipei City, Taiwan, using a novel bivariate generalized ordered probit (BGOP) model. Estimation results show that the BGOP model performs better than the conventional bivariate ordered probit (BOP) model in terms of goodness-of-fit indices and prediction accuracy and provides a better approach to identify the factors contributing to different severity levels. According to estimated parameters in latent propensity functions and elasticity effects, several key risk factors are identified-driver type (age>65), vehicle type (motorcycle), violation type (alcohol use), intersection type (three-leg and multiple-leg), collision type (rear ended), and lighting conditions (night and night without illumination). Corresponding countermeasures for these risk factors are proposed.

Service quality effects on air passenger intentions: a service chain perspective

This study divides entire airline services into eight service stages and uses a second-order confirmatory factor analysis (CFA) to constitute service quality and to examine the causal relationships between two consecutive service stages from a service chain perspective. Two conceptual frameworks – overall framework and service chain framework are proposed. The former incorporates the constructs of service quality, sacrifice, servicescape, service value, satisfaction, switching barriers, and behavioural intentions combined with seven hypothetical causal relationships. The latter depicts seven hypothetical causal relationships between two consecutive service stages. The proposed models are validated by a case study of Spring Airlines, a low-cost carrier (LCC) based in China. The empirical results support all hypotheses except hypothesis 7 that service quality positively impacts behavioural intentions. Notably, test results for the interrelationships between two consecutive service stages suggest that a lack of satisfaction at a specific service stage will affect customer perception of the consequent service stage. Therefore, to improve the service quality for a service stage, the service quality of all upstream service stages must be improved first. This study also found that service quality has a large effect although not direct on behavioural intentions while sacrifice has the smallest effect. A low-fare strategy may not be effective when an airline fails to deliver high-quality service.

Direct and indirect factors affecting emissions of cars and motorcycles in Taiwan

This article proposes Direct Correlation (DC) models and Integrated Correlation (IC) models for cars and motorcycles emissions, respectively. The DC models regress emissions on vehicle-related variables (i.e. direct factors), while the IC models further account for such variables (i.e. indirect factors) as driver/rider demographics, vehicle mileages travelled and regional types by using structural equation modelling. Results show that vehicle characteristics are the most influencing factors affecting hydrocarbon and carbon monoxide emissions. The old vehicles with small engine, manual transmission, high cumulative mileages travelled, using unleaded gasoline #92 and 2-stroke engine (for motorcycle) can be identified as ‘high-emitting’ vehicles. The second most influencing construct is the driver/rider demographics. The aged, male, low-educated car drivers and motorcyclists with high income and long driving/riding experience tend to use high-emitting vehicles.

A dynamic analysis of motorcycle ownership and usage: A panel data modeling approach

This study aims to develop motorcycle ownership and usage models with consideration of the state dependence and heterogeneity effects based on a large-scale questionnaire panel survey on vehicle owners. To account for the independence among alternatives and heterogeneity among individuals, the modeling structure of motorcycle ownership adopts disaggregate choice models considering the multinomial, nested, and mixed logit formulations. Three types of panel data regression models--ordinary, fixed, and random effects--are developed and compared for motorcycle usage. The estimation results show that motorcycle ownership in the previous year does exercise a significantly positive effect on the number of motorcycles owned by households in the current year, suggesting that the state dependence effect does exist in motorcycle ownership decisions. In addition, the fixed effects model is the preferred specification for modeling motorcycle usage, indicating strong evidence for existence of heterogeneity. Among various management strategies evaluated under different scenarios, increasing gas prices and parking fees will lead to larger reductions in total kilometers traveled.

A Novel Method to Predict Traffic Features Based on Rolling Self-Structured Traffic Patterns

In this study, a novel method is proposed to predict the traffic features in a long freeway corridor with a number of time steps ahead. The proposed method, on the basis of rolling self-structured traffic patterns, utilizes the growing hierarchical self-organizing map model to partition the unlabeled traffic patterns into an appropriate number of clusters and then develops the genetic programming model for each cluster to predict its corresponding traffic features. For demonstration, the proposed method is tested against a 110-km freeway stretch, on which 48 time steps of 5-min traffic flows are predicted (i.e., a 4-h prediction). The prediction accuracy of the proposed method is compared with other models (ARIMA, SARIMA, and naive models) and the results support the superiority of the proposed method. Further analyses indicate that applications of the proposed method to larger scale freeway networks require sufficient lengths of observation to acquire enough traffic patterns for training and validation in order to achieve higher prediction accuracy.