Search

Accurate estimates of manpower are still heavily dependent on well-experienced personnel. The objectives of this study are to prove the feasibility of using rough set theory to classify and weigh the impact attributes, and to develop a model to assess the total quantities of labor needed for a construction project using a rough set enhanced artificial neural network (ANN). Experts suggest 14 attributes that influence the estimation of on-site manpower for construction projects. After trimming and analyzing the basic data, the rough set approach is used to classify and weigh the attributes into three levels of impact based on their frequency. A rough set enhanced ANN is accordingly developed that yields an accuracy rate of 91.903%, higher than that of a regular ANN. A practical and effective prediction model benefits personnel having to estimate on-site manpower needs for construction projects.

Transit-oriented development (TOD) is a crucial method for sustainable urban development and encouraging the use of public transit, especially for large densely populated cities such as Can Tho City, one of the top five cities in Vietnam. This study is aimed at identifying the factors which have a significant effect on the citizens’ engagement of public transport in the TOD environment. Based on the literature review and expert knowledge, 3 hypotheses related to citizens’ perceptions and citizen’s engagement are developed. A questionnaire was accordingly established using structural equation modeling (SEM). As a result, a total of 8 factors with 31 stems were obtained. The survey questionnaires were distributed to citizens of Can Tho City with 749 responses collected for KMO test Chi-square of 0.905. The analysis shows that all three hypotheses stand: (1) the citizens’ perception of local government’s capacity plays an important role in the citizens’ willingness to engage on public transport; the citizens’ perception of the usefulness of (2) TOD and (3) public transport has a positive influence on the engagement on public transport system. The major contribution of this study is that it can help other cities in developing countries to apply TOD.

Debris flow is a natural hazard typically triggered by heavy rainfall. Previous research aimed at forecasting the occurrence of debris flows have led to the development of several rainfall duration-intensity thresholds for different areas using the second-percentile method that allows a missed-alarm probability of up to 2%, while disregarding the occurrence of false alarms. The current study aims to develop rainfall duration-intensity thresholds for debris flow forecasting taking into account both missed alarms and false alarms. Specifically, the new optimization approach seeks to determine the optimal duration-intensity threshold associated with the lowest missed- and false-alarm probabilities combined. In addition to the methodology, a case study is presented to show that the new optimization approach is feasible for determining rainfall duration-intensity thresholds in debris flow forecasting, and is more efficient than the method currently in use, which is associated with a higher probability of missed and false alarms combined.

This paper presents a new application of the Bayesian approach to objectively determine logic-tree weights for ground motion prediction equations (GMPEs) used in seismic hazard analysis. This is a considerable improvement over the current practice based on subjective judgment to determine the logic-tree weights for GMPEs. To demonstrate the feasibility of this Bayesian application, case studies at three sites in Taiwan were conducted. The results demonstrate that the logic-tree weights for three Next Generation Attenuation GMPEs can be determined in an objective and repeatable manner using this Bayesian calculation, which is based on the difference between predicted and observed peak ground accelerations.

The objective of this research was to explore the impact factors of sustainable corporate governance for top consulting engineering companies in Taiwan, to facilitate managers in meeting stakeholders’ needs and adapting to the challenges of the global markets. Nine hypotheses derived from a literature review were proposed and used to develop a survey. Based on the concept of structural equation modeling (SEM) and these hypotheses, a questionnaire containing six aspects and comprising 46 stems was developed using the Likert 5-scale format. The survey took around four months to administer with 324 effective returns, with only five hypotheses confirmed. This was followed by factor analysis to determine the weight sequence for the 28 impact factors and four aspects. The contributions of the findings are as follows: (1) the weighted factors provide practitioners with guidelines for the proper order for the implementation of measures to improve corporate governance, and (2) they answer questions about the degree of influence and the relationship among all aspects and factors for sustainable corporate governance.

Uncertainty studies and reliability assessments have become important topics/areas in geotechnical engineering, and different site-characterization approaches have been developed in an attempt to incorporate the sources of uncertainty into the analysis. Focusing on a target problem with multiple measurement profiles (with depth) from different testing procedures, this paper presents a site-characterization modeling to estimate the probability distribution of soil properties for a given site, considering the uncertainties from site-specific test-data scatter and from inherent soil variability in space (or in horizontal direction). More specifically, the modeling adopts the Bayesian approach to integrate the multiple sources of data, by using one as prior and the other as the so-called likelihood function. In addition to the algorithms, the approach was applied to a site with two sources of undrained shear strength from laboratory triaxial tests and in-situ plate bearing tests, and suggested a series of characteristic values for undrained shear strength for the site, corresponding to a given depth of interest, and to a given level of inherent soil variability in the horizontal direction.

From theory to experience, earthquake probability associated with an active fault should be gradually increasing with time since the last event. In this paper, a new non-stationary earthquake assessment motivated/derived from the Mohr–Coulomb failure criterion is introduced. Different from other non-stationary earthquake analyses, the new model can more clearly define and calculate the stress states between two characteristic earthquakes. In addition to the model development and the algorithms, this paper also presents an example calculation to help explain and validate the new model. On the condition of best-estimate model parameters, the example calculation shows a 7.6 % probability for the Meishan fault in central Taiwan to induce a major earthquake in years 2015–2025, and if the earthquake does not occur by 2025, the earthquake probability will increase to 8 % in 2025–2035, which validates the new model that can calculate non-stationary earthquake probability as it should vary with time.

Debris flows often cause catastrophic damage to communities in the downstream area, by direct impact and deposition. Theoretical predictions of impact pressure and volume of discharge, however, still remain very challenging, mainly due to inadequate understanding of the complex problems and limited field data at the local scale. In this study, the maximum impact pressure (P-max) and total discharge (Q(total)) of 139 debris flow events that occurred during 1961 and 2000 in the "debris museum" of China (i.e., the Jiangjia Ravine) are reported and interpreted with statistical tests and probabilistic analyses. Four common probabilistic models (Normal, Lognormal, Weibull and Gamma distributions) are used to simulate the distributions of P-max and Q(total). The level of fitting of each model is assessed by performing two quantity-based statistic goodness-of-fit tests (Chi-square and Kolmogorov-Smirnov tests). The field data show that during the period from 1961 to 2000, the maximum values of P-max and Q(total) are 744 kPa and 1,751,537 m(3), respectively. It is suggested by the goodness-of-fit tests that the Weibull distribution is the only model (among the four probabilistic models) that is able to capture the distributions of P-max and Q(total) of both surge and continuous flows. Using the verified Weibull distributions and Gaussian copula approach, univariate and bivariate exceedance probability charts considering P-max and Q(total) are developed. Regression models between P-max and Q(total) are also established. (C) 2014 Elsevier B.V. All rights reserved.

Characterizing the standard deviation of soil properties is important to a geotechnical probabilistic analysis, and the task is usually achieved with sufficient samples. However, sometimes soil samples or soil tests in a project could be limited (e.g., only one sample), making classical statistics approaches less applicable to the estimating. In this technical note, we introduce a new Bayesian algorithm to estimate the standard deviation of soil properties, using limited project-specific samples along with relevant prior information from the literature. In addition to the methodology, a few demonstrations are also given in the paper, to re-evaluate the standard deviation of soil properties with the new algorithm. Like many Bayesian algorithms, the new application could be useful for site characterizations when samples are limited.

Engineering risk assessment is to estimate the structure’s failure probability multiplied by its failure consequences. On the other hand, the analysis was sometimes performed on a more general basis, utilizing a few indicators to quantify relative risk levels for decision-making. On the basis of such a framework, this paper introduces a novel risk assessment on 18 active faults in Taiwan, considering the fault’s location, earthquake size and return period from the literature. The result shows that the Chelungpu Fault in central Taiwan has the highest risk score among the 18 active faults, in contrast to the Hengchun Fault in southern Taiwan which has the lowest risk. Besides this, the 18 active faults in Taiwan were classified into four groups based on this risk assessment, and the classification could help the region’s sustainable development against earthquake hazard.

The region around Taiwan is known for active seismicity, and a few studies have reported a high seismic hazard in this region, including a deterministic seismic hazard analysis (DSHA) study. Essentially, DSHA is to estimate earthquake ground motions considering the worst-case earthquake size and location, but without considering the seismic hazards from non-controlling sources. Understandably, when many non-controlling sources are present, the original DSHA framework could be insufficient. Therefore, using the extreme probability theory, this study introduces a new DSHA framework taking non-controlling seismic sources into account during DSHA calculations. The new method was applied to a seismic hazard assessment for Taiwan, showing that near the conjunctions of seismic source zones, the increase in seismic hazard could be substantial after considering a total of 19 non-controlling sources. More importantly, like other seismic hazard assessments for Taiwan, this study conveys the same alarming message that a high level of seismic hazard should be present around the region.

Earthquake size can be described with different magnitudes for different purposes. For example, local magnitude ML is usually adopted to compile an earthquake catalog, and moment magnitude Mw is often prescribed by a ground motion model. Understandably, when inconsistent units are encountered in an earthquake analysis, magnitude conversion needs to be performed beforehand. However, the conversion is not expected at full certainty owing to the model error of empirical relationships. This paper introduces a novel first-order second-moment (FOSM) calculation to estimate the annual rate of earthquake motion (or seismic hazard) on a probabilistic basis, including the consideration of the uncertain magnitude conversion and three other sources of earthquake uncertainties. In addition to the methodology, this novel FOSM application to engineering seismology is demonstrated in this paper with a case study. With a local ground motion model, magnitude conversion relationship and earthquake catalog, the analysis shows that the best-estimate annual rate of peak ground acceleration (PGA) greater than 0.18 g (induced by earthquakes) is 0.002 per year at a site in Taipei, given the uncertainties of magnitude conversion, earthquake size, earthquake location, and motion attenuation.

According to the Central Geological Survey Taiwan (CGST), the Meishan fault in Central Taiwan, which induced a catastrophic earthquake in 1906, is considered capable of triggering equivalent events with a return period of 162 years. Therefore, as the next event is expected around 2070, the Meishan fault poses a high level of earthquake risk in Central Taiwan, especially for those cities and townships very close to the fault. However, the best-estimate return period (162 years) and earthquake magnitude (7.1 M L) reported by the CGST must be subject to some uncertainty, because such an event is very unlikely to recur every 162 years. Therefore, this study carried out a series of seismic hazard assessments for three major cities close to the Meishan fault, with the uncertainty of the best-estimate information being determined using the Rosenblueth algorithm. The results show that for Chaiyi city, which the Meishan fault passes through, the probability of the current design value (i.e., PGA = 0.33 g) being exceeded by the earthquake motion induced by the Meishan fault in the next 50 years would be as high as 55 %. Therefore, further studies should be carried out to obtain a better understanding of the Meishan fault, to develop a robust hazard mitigation plan for nearby cities.

Taiwan is located in an active seismicity region with more than thousands of events occurring every year in average. On the other hand, this small inland is a home of around 23-million people, and nearly 20% of the daily activity is powered by the three existing nuclear power plants (NPPs), with a new one under construction located in Northeastern Taiwan, which is officially referred to as the Longman nuclear power plant, also known as the “fourth NPP” or the “NPP 4” locally. However, when seeing the seismicity in the last 110 years, one can not help but wonder how the decision-making process finalizes such a location for such a critical structure. As a result, the underlying scope of this study is to discuss the site selection with the fundamentals of engineering decision-making given three prior objectives: maximizing the design safety, minimizing the construction cost, and minimizing the risk. Given the NPP 4 following the same earthquake-resistant design in Taiwan, the construction cost is very likely the same as building other NPPs, but owing to the high earthquake potential in Northeastern Taiwan, the risk should be higher than those located in North Taiwan and South Taiwan. Therefore, without further information, possibly confidential, being available at this moment, we must conclude that the NPP 4 site in Taiwan is not an optimal location with a scientific, engineering decision-making analysis given the earthquake activity and seismic hazard. We hope and believe that such a decision was made with a condition that the NPP 4 was constructed with a strict standard to compensate the high seismicity around the site. Under the circumstance, the location would be a logical option with the risk being minimized to an acceptable level, regardless of the construction cost.

This note provides the overview of the calculation in soil liquefaction probability in Central Taiwan carried out by Lee et al. Although that estimation is by all means original and offers an in-depth evaluation on the liquefaction risk for Central Taiwan, we raised a few questions about its earthquake magnitude function, source-to-site distance function, and the fault distributions in Central Taiwan. We personally think that with those concerns being accounted for, that estimation should be more robust.

In this work, two approaches were employed for estimating the spatiotemporal distribution of seismicity density in Taiwan. With the use of the rate-and-state friction model, a model for short-term forecasting according to the fault-interaction-based rate disturbance due to seismicity was considered. Another long-term forecasting model that involves a smoothing Kernel function is proposed. The application of the models to Taiwan led to good agreement between the model forecast and actual observations. Using an integration of the two approaches, the application was found to be capable of providing a seismicity forecast with a higher accuracy and reliability. To check the stability related to the regression the bandwidth function, the forecasted seismicity rates corresponding to the upper and lower bounds of the 95% confidence intervals are compared. The result shows that deviations within the bandwidth functions had an insignificant impact on forecasting reliability. Besides, insignificant differences in the forecasted rate change were obtained when Aσ was assumed to be between 0.1 and 0.4 bars for the application of the rate-and-state friction model. By considering the maximum Coulomb stress change among the seismogenic depth, the model presents a better forecasting ability than that using any single fixed target depth. The proposed methodology, with verified applicability for seismicity forecasts, could be useful for seismic hazard analyses.

This paper reports on the seismic performance of three geosynthetic-reinforced segmental retaining wall systems backfilled with a silty sand mixture, using a shaking table excited by 1995 Kobe earthquake loadings. The preparation of the backfill mixture and its properties, the tested wall configurations, the reinforcement layouts and instrumentations, and the observed wall performance are described. Visual observations and test results indicate that walls having 0.4-m vertical reinforcement spacing, backfilled with soil containing a large percentage of fines, performed better than those having good-quality sandy soil under otherwise identical conditions. Vertical spacing of 0.8 m with removal of interlocking facing blocks in one of the walls did not lead to global collapse under repeated applications of the Kobe earthquake loadings. Only localized shear failure behind the top block layer was observed as the top facing blocks tended to topple. The good performance was attributable to apparent cohes...

Earthquake early warning systems have been developed to help minimize the loss of life and other damages resulting from catastrophic earthquakes. In the past several decades, these systems have evolved to function using a variety of methodologies. This study introduces an onsite system developed from a correlation between the precursor measurement (Pd3) and the peak ground acceleration (PGV) from 780 historical earthquake events recorded in Taiwan, Japan and South California. The study analyzes and integrates the uncertainty of the onsite system, establishing the statistical relationship between the PGV forecast, the Pd3 precursor, and the exceedance probability (PE). A method is presented for forecasting the PGV by scaling the Pd3. Furthermore, this study also suggests a decision-making model to determine whether or not the early warning notification should be issued, based on factors such as the uncertainty of the early warning system, the magnitude of the precursor, and warning recipient characteristics.

The region around Taiwan is well known for its unique geological setting which has resulted in a high seismicity. The Central Geological Survey, Taiwan (CGST), has launched a series of investigation on characterizing the active faults on the island. The CGST revealed that there are 33 active faults and their fault parameters are suggested in the latest publication in 2010. However, the uncertainties of return period and earthquake magnitude are not provided. The main objective of the paper is to study the impact of “unspecified” parameter uncertainties on recurrence earthquake probabilities. In addition, a numerical approach involving the use of Monte Carlo simulation was proposed to estimate the exceedance probabilities in which the parameter uncertainties were taken into account. The Meishan fault in Central Taiwan with the best-estimate return period of 162 years and earthquake magnitude of 7.1 was used as an example. The analyses verified the impact of parameter uncertainties on the exceedance probability in fault-induced earthquake recurrence. When uncertainty is high, the exceedance probability associated with large magnitudes is high. It was also observed that the probabilistic analysis in this study can generate a realistic, continuous distribution of exceedance probabilities that cannot be obtained by a deterministic approach. The best-estimate distribution of exceedance probability was established by enveloping the curves estimated by both deterministic and probabilistic approaches. Accordingly, the best-estimate exceedance probability for a 7.5-magnitude earthquake recurring in 2011–2060 induced by the Meishan fault in Central Taiwan was estimated as 11%.

The annual maximum earthquake magnitudes around Taiwan from 1900 to 2009 are presented in this paper. Using the distribution of the AMEM, a probabilistic framework to estimate the recurrence probability of a large-size earthquake is also proposed and an illustration was made in this paper. The mean value of the 110-AMEM is 6.433, and the coefficient of variation is around 10%. The results of two goodness-of-fit tests show that the Gamma and lognormal distributions are relatively suitable to represent the AMEM around Taiwan among five common probability distributions. Using the proposed approach, the recurrence probability is 4% for an earthquake with magnitude greater than 7.5 in a 1-year period around Taiwan. More site-specifically, the probability is around 5% in Central Taiwan for such an earthquake to occur in a 50-year period.

Cyclic loading has been known to induce fluid flow and thus mechanotransduction in bones. In the past, four-point bending tests have been used exclusively in studying fluid flow in bones. In order to better understand the mechanism of deformation and fluid flow under loading, compression tests were done on trabecular bone specimens under drained and undrained conditions. In the drained tests, the volume change was observed, whereas in the undrained tests, excess pore fluid pressure was measured. Cyclic loading tests were conducted in addition to monotonic loading tests to observe the permanent volume change or excess pore fluid pressure with loading cycles. A fast loading rate gave a sharp rise in the excess fluid pressure compared to a slow loading rate. The strength and stiffness of the specimens appeared to deteriorate with an increased speed of loadings, but there was no appreciable difference between the results obtained from drained and undrained tests. The drained and undrained tests as described allowed a better understanding of bone behavior under loadings for a coupled stress-flow analysis.

Track 2 in KDD Cup 2013 aims at determining duplicated authors in a data set from Microsoft Academic Search. This type of problems appears in many large-scale applications that compile information from different sources. This paper describes our solution developed at National Taiwan University to win the first prize of the competition. We propose an effective name matching framework and realize two implementations. An important strategy in our approach is to consider Chinese and non-Chinese names separately because of their different naming conventions. Post-processing including merging results of two predictions further boosts the performance. Our approach achieves F1-score 0.99202 on the private leader board, while 0.99195 on the public leader board.

The track 1 problem in KDD Cup 2013 is to discriminate between papers confirmed by the given authors from the other deleted papers. This paper describes the winning solution of team National Taiwan University for track 1 of KDD Cup 2013. First, we conduct the feature engineering to transform the various provided text information into 97 features. Second, we train classification and ranking models using these features. Last, we combine our individual models to boost the performance by using results on the internal validation set and the official Valid set. Some effective post-processing techniques have also been proposed. Our solution achieves 0.98259 MAP score and ranks the first place on the private leaderboard of Test set.