Abstract: The cities of Johannesburg and Pretoria both located in the Gauteng province are separated by a distance of 58 km. The traffic queues on the Ben Schoeman freeway which connects these two cities can stretch for almost 1.5 km. Vehicle traffic congestion impacts negatively on the business and the commuter’s quality of life. The goal of this paper is to identify variables that influence the flow of traffic and to design a vehicle traffic prediction model, which will predict the traffic flow pattern in advance. The model will unable motorist to be able to make appropriate travel decisions ahead of time. The data used was collected by Mikro’s Traffic Monitoring (MTM). Multi-Layer perceptron (MLP) was used individually to construct the model and the MLP was also combined with Bagging ensemble method to training the data. The cross—validation method was used for evaluating the models. The results obtained from the techniques were compared using predictive and prediction costs. The cost was computed using combination of the loss matrix and the confusion matrix. The predicted models designed shows that the status of the traffic flow on the freeway can be predicted using the following parameters travel time, average speed, traffic volume and day of month. The implications of this work is that commuters will be able to spend less time travelling on the route and spend time with their families. The logistics industry will save more than twice what they are currently spending.
Abstract: The Japanese version of the SF-36 has been employed
to assess individuals’ health-related QOL (HRQOL). This study aimed
to clarify the HRQOL of motorists with a spinal cord injury, in order to
compare these individuals' SF-36 scores and national standard values.
A total of 100 motorists with a spinal cord injury participated in this
study. Participants’ HRQOL was evaluated using the Japanese version
of the SF-36 (second edition). The score for each subscale was
standardized based on data on the Japanese population. The average
scores for NPF, NRP, NBP, NGH, NVT, NSF, NRE, and NMH were
10.9, 41.8, 45.9, 47.1, 46.1, 46.7, 46.0, and 47.4 points, respectively.
Subjects showed significantly lower scores for NPF and NRP
compared with national standard values, which were both ≤ 45.0
points, but relatively normal scores for the other items: NBP, NGH,
NVT, NSF, NRE and NMH (> 45.0 points). The average scores for
PCS, MCS and RCS were 21.9, 56.0, and 50.0 points, respectively.
Subjects showed a significantly lower PCS score (≤ 20.0 points);
however, the MCS score was higher (> 55.0 points) along with a
relatively normal RCS score in these individuals (= 50.0 points).
Abstract: Due to the call of global warming effects, city planners aim at actions for reducing carbon emission. One of the approaches is to promote the usage of public transportation system toward the transit-oriented-development. For example, rapid transit system in Taipei city and Kaohsiung city are opening. However, until November 2008 the average daily patronage counted only 113,774 passengers at Kaohsiung MRT systems, much less than which was expected. Now the crucial questions: how the public transport competes with private transport? And more importantly, what factors would enhance the use of public transport? To give the answers to those questions, our study first applied regression to analyze the factors attracting people to use public transport around cities in the world. It is shown in our study that the number of MRT stations, city population, cost of living, transit fare, density, gasoline price, and scooter being a major mode of transport are the major factors. Subsequently, our study identified successful and unsuccessful cities in regard of the public transport usage based on the diagnosis of regression residuals. Finally, by comparing transportation strategies adopted by those successful cities, our conclusion stated that Kaohsiung City could apply strategies such as increasing parking fees, reducing parking spaces in downtown area, and reducing transfer time by providing more bus services and public bikes to promote the usage of public transport.
Abstract: Road rage is an increasingly prevalent expression of
aggression in our society. Its dangers are apparent and understanding
its causes may shed light on preventative measures. This study
involved a fifteen-minute survey administered to 147 undergraduate
students at a North Eastern suburban university. The survey
consisted of a demographics section, questions regarding financial
investment in respondents- vehicles, experience driving, habits of
driving, experiences witnessing role models driving, and an
evaluation of road rage behavior using the Driving Vengeance
Questionnaire. The study found no significant differences in driving
aggression between respondents who were financially invested in
their vehicle compared to those who were not, or between
respondents who drove in heavy traffic hours compared to those who
did not, suggesting internal factors correlate with aggressive driving
habits. The study also found significant differences in driving
aggression between males versus females, those with more points on
their license versus fewer points, and those who witnessed parents
driving aggressively very often versus rarely or never. Additional
studies can investigate how witnessing parents driving aggressively
is related to future driving behaviors.
Abstract: We proposed a technique to identify road traffic
congestion levels from velocity of mobile sensors with high accuracy
and consistent with motorists- judgments. The data collection utilized
a GPS device, a webcam, and an opinion survey. Human perceptions
were used to rate the traffic congestion levels into three levels: light,
heavy, and jam. Then the ratings and velocity were fed into a
decision tree learning model (J48). We successfully extracted vehicle
movement patterns to feed into the learning model using a sliding
windows technique. The parameters capturing the vehicle moving
patterns and the windows size were heuristically optimized. The
model achieved accuracy as high as 99.68%. By implementing the
model on the existing traffic report systems, the reports will cover
comprehensive areas. The proposed method can be applied to any
parts of the world.