2021 Selected Papers in English
This paper proposes a risk assessment method of school commuting accident and identify the crucial influencing factors, by integrating traffic accident data, road operation data and school district zoning data, the study developed the accident assessment method and analyzed the influencing factors using the random forest model. With the assumption that pupils' school commuting is within the school district, a road exposure model was developed through traffic accident data and road length data to evaluate the pupils' school commuting accident risk. A verification analysis was conducted by taking the central city of Shenzhen as an example. The results show that: the high- risk school districts in the central city of Shenzhen are mainly in the north of Nanshan and Luohu, while the low-risk school districts are mainly distributed in the south-central part of Nanshan and the central part of Futian. The risk assessment model based on the random forest produced 85.93% prediction accuracy. Which shows the performance of the proposed model for the pupils' school commuting accident risk assessment. Primary school density and school district zoning area are two major influencing factors of pupils' school commuting accident risk, which can explain respectively 37.15% and 22.86% of school commuting accident risks.
In order to explore the revolution of the metro network's resilience, this paper constructs a series of indexes for assessing the stability of metro flow-weighted network from the perspectives of network connectivity, selfadaptivity, and availability, and proposes a method of analyzing cascading failures based on coupled map lattice model with the consideration of the impact of OD flow distribution on the network. Taking two metro networks of Beijing in 2014 and 2018 as examples, this paper constructs the dynamic stability analysis model based on the P-space weighted complex network and carries out the simulation attacks with different interferences on three nodes with different importance in these two networks to compare the evolution of network's stability. The numerical results show that with the topological structure of the Beijing metro network evolved from a small-world to scale-free network, the stability and invulnerability of the network in 2018 were significantly improved compared with 2014. The greater the attack on the stop with the larger flow, the more the cascading failures, the higher the probability of network paralysis, and the faster the paralysis speed. The research results can provide scientific theoretical basis for optimizing the network structure and site location and strengthening the safe operation and management of metros.
Due to the complex system composition, various maintenance tasks, and limited maintenance resources of urban tail transit infrastructure systems, a long- term maintenance task arrangement method is proposed considering attributions of tasks and constraints from resources. An optimization model based on mixed-integer programming is established to minimize the total cost by arranging the start time and execution period of maintenance tasks with various types. The occupation penalty is imported to reduce the task conflicts and improve the maintenance arrangement. A parallel hybrid algorithm combining the commercial solver and a heuristic method is designed for solving the large-scale problem with complicated constraints. The case study verifies the importance of task occupation on improving the executability of the maintenance arrangement. The proposed method can save the total cost by coordinating various types of maintenance tasks and provide decision support for arranging long- term maintenance tasks on urban rail transit infrastructure systems.
To choose an efficient and safe transportation route and manage the transport risk for China Railway Express (CRE), a Weibull distribution was implemented to measure the probability distribution of delay time, according to the delay time features of different types of borders. And a novel risk value measurement was constructed based on the expected loss theory. Considering the minimum comprehensive transportation cost of risk, time, and cost, a new route utility model was established. A depth- first traversal algorithm was used to solve the route selection of goods with different value of CRE from Chengdu to Berlin. The results show that high-value goods are more sensitive to border delay, and the rate of cargo loss has less impact on the delay cost; the delay risk of the optimal route is high with a 0.46 average delay rate; when the unit value of goods is 3 million yuan, the border delay cost of the optimal route is 1.2 times of the transportation cost. Attentions need to be paid on monitoring delay risk and dealing with an emergency of borders by related stakeholders.
This paper predicts the 2020 to 2035 external passenger transport demand of a typical sea island city-Haikou in Hainan Province. A system dynamics model was developed to incorporate multi- dimensional qualitative and quantitative influencing factors such as economy, policy, and infrastructure. The forecast results show that the demand for foreign passenger transport in Haikou will experience a steady increase in the next 15 years. and the total number of foreign passenger transport is expected to exceed 80 million in 2035. After a sensitivity analysis of the Gross Domestic Product (GDP), the result shows that when the growth rate is greater than 9%, the current facility construction plan needs to be further strengthened to proactively improve the supply capacity of transportation facilities. Based on the sensitivity analysis of the introduced population, it is concluded that the maximum economic benefit can be achieved when the scale of talent introduction is controlled at about 1 million people, the benefit is 6.9% higher than 0.6 million talent introduction and 5.1% higher than and the 1.4 million talent introduction. Based on the sensitivity analysis of the construction time sequence of airport and high- speed rail, the result shows that comparing with high- speed rail, the opening of airport can boost the growth of demand by 6.3% , which also indicates that aviation is a major way of external travel in sea island cities.
In view of the market competition between the domestic node cities in the network of the China Railway Express, this paper establishes a competition cooperation game model to analyze the competition cooperation relationship between governments and platform companies by considering the competition and cooperation decisionmaking of both local government and platform company. In the model, the local government, as one of the game participants, formulates the subsidy strategy for platform company according to the maximum goal of social welfare, and the platform company makes demand according to the maximum goal of profit under the government subsidy. The paper compares the profit and government social welfare of platform companies under three different scenarios: (i) fierce competition, (ii) platform company cooperation and (iii) government led the cooperation. The optimal strategy of competition and cooperation of node cities is analyzed with numerical examples. The results show that the government led cooperation between platform companies of two places is beneficial to the increase of local social welfare. However, such cooperation mode is not optimal for platform companies. When platform companies and competitors take the initiative to seek cooperation, platform companies can obtain higher demand and profits. Further numerical analysis shows when the competition is more intense, the government and platform companies are more motivated to cooperate. The study provides important reference for the reconstruction of China Railway Express ecology.
This paper investigates the horizontal alignment and station location of the underground urban rail transit considering the fixed starting and ending points and the study area population distributions. A mathematical model was established to optimize the number and location of stations and the horizontal alignment of the rail transit. The model considers the constraints of geographical requirements, station spacing, transfer reservations, and horizontal track alignment design. The objectives are maximizing the attracted passenger flow while minimizing the overall costs which include construction investment, operating expenditure and environmental impact. A hybrid heuristic algorithm based on non- dominated sorting genetic algorithm (NSGA- II) and A* algorithm was designed to solve the model. A case study on Chinese Metro shows that the pareto frontier of the model can be solved by the proposed algorithm, which provides the solution with the minimal overall costs under the different attracted passenger number. In addition, compared with the practical alignment, the horizontal alignment optimized by the proposed model decreases the overall costs of line by about 4.4% without a reduction of attracted passenger flow.
It appears that the current research on the relationship between built environment and walking activities of the elderly hasn't considered the relationship between individual subjective perception and the spatial hierarchy. This paper develops a multi-level linear model of subjective perception, built environment and walking distance efficiency of the elderly to reveal the nature and degree of interaction among variables at all levels. The results show that: safety and mobility are two subjective factors affecting the walking efficiency of the elderly. The impact of built environment factors on elderly walking distance efficiency shows either consistency or difference. The intersection density, NDVI vegetation coefficient and public service facilities density are the consistency factors, while the plot ratio and road network density have different impacts on the travel distance of 300 meters or longer. The mediating effect test shows that the floor area ratio is the key factor relevant to the traffic safety within 300-meter walking space; the community open form affects the elderly's long-distance travel through the security of surrounding areas; NDVI vegetation coefficient is also an important factor related to the elderly's walking environment security.
To analyze drivers' visual characteristics at the entrance and exit of undersea tunnel, this study carried out the real vehicle test with 26 drivers operating in the off peak period under similar traffic conditions. The data of illuminance, eyelid closure, gaze duration, vehicle speed and position were collected using the vision instrument and illuminance meter under the real traffic condition. The change rules of eyelid closure, gaze duration and driving speed were analyzed under the coupling effect of different illuminances and longitudinal slopes. The mathematical models were proposed to describe the co- relations between eyelid closure, gaze duration, driving speed and illuminance and slope. The results show that: when driving through the entrance and exit section of the undersea tunnel, (1) driver's eyelid closure is significantly reduced, and the illuminance has significant impact on eyelid closure; (2) the fixation time is increased, which is significantly affected by illuminance; (3) the fixation points are mostly distributed on the front vehicles; (4) the vehicle speed shows a trend of decline- rise, then keep steady, and then decline-rise. The illuminance and longitudinal slope also have significant impact on vehicle speed.
This study investigates the relationship between pedestrian crossing speed and the elderly pedestrian proportion under different pedestrian flow conditions. The micro simulation model was used to obtain the crossing speed and elderly pedestrian proportion data considering different pedestrian level of services (PLOS). The pedestrian free-flow speed and accelerations for the old people, the young and middle-aged people were collected at three signalized intersections in Chongqing, China to calibrate the micro-simulation model. The relationship between the elderly pedestrian proportion and the pedestrian crossing design speed was established by the principal component analysis and the regression model. The results indicate that the design speed of pedestrian crossing is 0.94 m ⋅ s-1 when the elderly pedestrian accounts for 21% to 41% of the pedestrian flow. The design speed of pedestrian crossing is recommended to be 0.86 m ⋅ s-1 when there are more than 41% of pedestrians are elderly people, which meets the needs of the elderly crossing at signalized intersections.
