Archive

25 June 2016, Volume 16 Issue 3 Previous Issue    Next Issue

---

For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Select

Incentive Mechanism of Public-private Partnership Rail Transit Project from the Perspective of Multitask Principal-agent Model YI Xin 2016, 16(3): 1-7.  Abstract ( )   PDF (1208KB) ( )   To design an effective PPP rail transit project supervision incentive mechanism, and to improve current government regulatory system defects, a new PPP rail transit project supervision organization structure is proposed in this paper. According to the responsibility of supervision organization for economic and quality supervising tasks in the PPP rail transit project, the incentive mechanism with multi- task principal-agent model is constructed, which includes government, project supervision organization and SPV. Solution of the model is got in accordance with the incentive compatibility and participation constraint. The results show that the most effective incentive mechanism is that government should make rewards and punishments to the supervision organization with reasonable presuppose supervising goals, which enforce the supervision organization intensify supervising the SPV. Based on feedback information from the supervision organization, government can take the corresponding strategies and measures to the SPV. References | Related Articles | Metrics

Select

Public Transit Improvement Measures of Low-income Commuters Based on Market Segmentation CHENG Long, CHEN Xue-wu, YANG Shuo, YUAN Ming-yi 2016, 16(3): 8-14.  Abstract ( )   PDF (1289KB) ( )   To identify low- income commuters with different public transit preferences, market segmentation using subjective attitudinal factors is conducted based on travel diary survey data of Fushun. First, factor analysis is utilized to determine latent attitudes of low- income commuters. Then, structural equation modeling is used to explore the causal relationship between attitudinal factors and public transit travel. Last, K- means cluster method is employed to segment low- income commuters. The entire public transit market is segmented into five sub- markets by three attitudinal factors including preferences for comfort, preferences for reliability and environment awareness. Commuters in the same sub-market possess similar travel preferences while those in different sub-markets have distinct preferences. Travel and attitudinal characteristics of different sub- markets are examined to analyze effects of latent attitudes on public transit choices. Polices that serve differentiated public transit travel needs are proposed References | Related Articles | Metrics

Select

A Combined Trip Distribution and Traffic Assignment Model under Regret Theory View LI Meng，HUANG Hai-jun 2016, 16(3): 15-20.  Abstract ( )   PDF (1479KB) ( )   It is important to develop newly theoretical and practical approaches of studying various travel behaviors. The majority of route choice models assume that travelers are completely rational and always choose paths with the maximum expected utility or the minimum expected cost, which comes into conflict with bounded rational behavior in reality. Random regret minimization model has attracted scholars’ attention for overcoming the drawbacks inherent in random utility maximization model. This paper proposes a combined trip distribution and traffic assignment model under the principle of random regret minimization. Numerical results from an example are presented to compare and explain the similarities and differences between the two models governed by random regret minimization and random utility maximization, respectively. It is shown that the proposed model under regret theory view could more realistically depict travelers’route choice behavior. References | Related Articles | Metrics

Select

Modeling and Simulation of Risk Communication in Multimodal Transportation Networks LEI Kai, ZHU Xiao-ning, HOU Jian-fei 2016, 16(3): 21-27.  Abstract ( )   PDF (1670KB) ( )   The dynamic theory of complex networks is introduced into the study of the risk communication of multimodal transportation network. This paper aims to reveal the complex internal rule of the risk communication in the multimodal transportation network. Firstly, the applicability of complex network dynamics model is described. Secondly, the risk communication process is deeply analyzed. Then, the risk communication dynamics model is built based on the disaster spreading dynamic model. And then the simulation experiment of the risk communication is designed in multimodal transportation network. Finally, the risk communication range is simulated from three aspects, such as node type, initial risk value and node risk threshold. The research shows that the risk communication dynamics model in this paper is in line with the process and can reflect the process of the risk communication of the multimodal transportation network. At the same time, the simulation experiment reveals the risk communication rule based on the differentiation of network nodes, and can provide certain theoretical basis for the risk control of the multimodal transportation network. References | Related Articles | Metrics

Select

Dynamic Volatility of China Containerized Freight Index Based on MCMC Algorithm of AR-GARCH Model WANG Si-yuan, YU Si-qin, PAN Jing-jing 2016, 16(3): 28-34.  Abstract ( )   PDF (1351KB) ( )   China has a large number of importing and exporting containerized cargo. Shippers and shipping companies face enormous risks from liner freight rates volatility. An AR- GARCH model is proposed to capture dynamic volatility of CCFI with Griddy- Gibbs sampling applied to simulate in WinBUGS. CCFI weekly is from April 1998 to December 2013. The empirical results of MCMC algorithm to a Bayesian inference show that the AR(3)-GARCH(1,1) model well fit the data. The strong persistence of volatility is reflected by the estimations, but no risk- premium or leverage effects. Results show that AR- GARCH- T model has better fitting effect. The AR-GARCH-T model estimated by ML within the sample is more fitting, while the counter party inferred by Bayesian beyond the sample is more predictable. References | Related Articles | Metrics

Select

Review on the Application of EEG in Traffic Driving Behavior Study GUANWei, YANG Liu, JIANG Shi-xiong, ZHANGWen-yi 2016, 16(3): 35-44.  Abstract ( )   PDF (1210KB) ( )   Drivers are“self- driven particle”factors of a traffic system, and its perception characteristics have close relationship with traffic driving behavior. It is an effective way to detect the drivers’perception characteristics by using electroencephalography (EEG) to analyze their brain signals quantitatively. This paper presents the key scientific problems of EEG researches, experimental environment, EEG signal processing methods and data analysis methods from four aspects which are fatigued driving, distracted driving, sleep-deprived driving and driving under some other specific conditions. It is founded that the research essence is to study the qualitative and quantitative relationship between various driving states and EEG; the common study approaches including using simulation driving experiments to collect various data, such as EEG data; and then some signal processing methods, such as power spectrum analysis, are adopted to process EEG signals; after that, statistical methods, such as variance analysis, are used to analyze the data. In the end, the potential future directions of EEG research in traffic research fields are also proposed References | Related Articles | Metrics

Select

Dynamic Programming Approach for Optimization of Arterial Offset with Multi-bandwidth HE Zhong-he, ZHANG Xing-yuan, LI Ying-hong 2016, 16(3): 45-52.  Abstract ( )   PDF (1615KB) ( )   The control objective of traditional arterial offset optimization usually takes green-wave bandwidth, or delay as a control objective. It does not build an analytic relation between delay and greenwave bandwidth. Consequently, according to graphic method and queue analysis based on vehicle arriving rate and discharging rate, a linear relation between delay and green-wave bandwidth is built; furthermore, making number of vehicle passing intersections over bandwidth maximum is regarded as a control objective, and geometry function is used to distribute forward and backward green-wave bandwidth; and it is solved based on the dynamic programming. In a simulation of Yi-Le Middle Road in Beijing, it shows that, although, the green-wave bandwidth acquired by this new method is not definitely maximum, the bandwidth and the distribution of the forward and backward green-green wave bandwidth can reflect the change of the traffic flow more accurately and control queuing delay and green-wave bandwidth at the same time. References | Related Articles | Metrics

Select

A Transit Priority Signal Timing Optimization Strategy Based on Dynamic Weight LIU Yu-gang, LI Jia-li, ZHANG Qing,WANG Hai-ning, DONG Dao-jian 2016, 16(3): 53-59.  Abstract ( )   PDF (1438KB) ( )   The signal timing of signal controlling intersections should consider increasing traffic capacity with the person unit so as to reduce traffic delay per capita. Vehicle parking rate and traffic delay are main relevant factors of cycle length, and decreasing them is the direct purpose of improving the timing design of signal intersections. To aim at minimizing the vehicle parking rate and traffic delay, a timing model of multiobjective planning with dynamic bus priority weights is built. The traditional signal timing plan is improved by signal phase design which is based on transit priority, and the effectiveness of transit priority strategy is ensured. Finally, a real intersection is taken as a simulation prototype in order to verify feasibility of the best cycle timing model and the effectiveness of the improved signal timing plan. The results show that transit priority signal timing optimization strategy based on dynamic weight makes an anticipated good effect on decreasing passengers and vehicles’traveling delay and parking rate. References | Related Articles | Metrics

Select

Prediction Model of Bus Arrival Time Based on Particle Swarm Optimization and Wavelet Neural Network JI Yan-jie, LU Jia-wei, CHEN Xiao-shi, HU Bo 2016, 16(3): 60-66.  Abstract ( )   PDF (2036KB) ( )   Real-time bus arrival time prediction is the key technology of bus travel information release, bus trip guidance, and bus dynamic scheduling. Based on the characteristic analysis of bus operation, the bus arrival time is divided into section running time and platform docking time. With the consideration of differences between running characteristics of the weekday and weekend, a forecasting model is proposed based on iterative thinking, particle swarm optimization and wavelet neural network to forecast bus arrival time. Example analysis shows that the particle swarm optimization can effectively reduce the training error of wavelet neural network model. Combined with the iterative method, the use of bus running time as forecast input can effectively improve the prediction accuracy. The bus arrival time prediction model is built in this paper can reach high precision on weekday and weekend, and mean absolute error is 10.82% and 9.85%. References | Related Articles | Metrics

Select

Step-by-step Guidance Algorithm for the Frequently Changing Congestion Degree of Stochastic Road Networks GONG Yana，ZHANG Jiea，LI Su-jiana，LAN Jin-huib 2016, 16(3): 67-72.  Abstract ( )   PDF (1221KB) ( )   When the congestion degree of stochastic road networks changes frequently, shifts in the congestion degree of roads or intersections may lead directly to vehicle congestion. This occurrence can cause guidance algorithms to fail. To address this problem, the frequent changes in the congestion degree of stochastic road networks are first analyzed with the use of a novel variable cell transmission model for stochastic road networks with a time-varying congestion degree. Moreover, a frequency index is established for changes in congestion degree. Second, a step-by-step guidance algorithm is established for the frequently changing congestion degree of stochastic road networks based on the least-time path algorithm. Part of a road network in Beijing is used in the simulation, which validates the frequent changes in the congestion degree of stochastic road networks and the step- by- step guidance algorithm. The simulation results indicate that the change frequency and time of the congestion degree of stochastic road networks are random. Furthermore, the step-by-step guidance algorithm is suitable for investigating the frequently changing congestion degree of stochastic road networks, the long distances between origins and destinations, and long road sections. References | Related Articles | Metrics

Select

Research on En-route Behavior under Short-term Incident CHEN Ling-juan，WANG Dian-hai 2016, 16(3): 73-80.  Abstract ( )   PDF (2304KB) ( )   In order to study the non-equilibrium state under short-term incident, an en-route model for two types of travelers facing delay is established. Travelers without information choose en- route diversion behavior when arriving at incident section and computing delays exceeding tolerance; Travelers with information change route at pre-trip and adjust departure time through real-time information, also choose enroute diversion at any nodes existing effective transfer routes. Transfer coefficient is adopted to determine flow transition probability. Link transmission model is adopted to calculate travel time and traffic density, and results are released to travelers as real- time information. A numerical example has compared network performance under three mechanisms. Results show that under condition of higher degree information, network congestion is lower but network stability is poorer, the probability of en- route is larger; whereas information degree is lower, network congestion is higher and network stability is better. References | Related Articles | Metrics

Select

Random Forest Based Operational Missing Data Imputation for Highway Tunnel QIAN Chao, CHEN Jian-xun, LUO Yan-bin, DAI Liang 2016, 16(3): 81-87.  Abstract ( )   PDF (1745KB) ( )   Real- time & completely accessing and deeply mining of tunnel operational data such as environment state and traffic status is a foundation work to improve emergency response capacity and realize safety early warning. An imputation method is proposed based on Random Forest algorithm. Missing data set is separated according to missing features. Random Forest regression model is built to iteratively impute after the determination of stopping criterion. The optimal combination of decision tree numbers and variables numbers randomly sampled at each split in Random Forest are identified by taking the minimum normalized root mean square error as objective function. Imputation results on highway tunnel operational missing data indicate that the method provides significantly higher precision and better robustness than KNN, SVD, MICE, PPCA, reducing normalized root mean square error by at least 25% . Moreover, the imputation efficiency is improved significantly by using parallel computation. It covers the shortage of slow imputation speed and provides a warranty of effectiveness and timeliness in missing data imputation. References | Related Articles | Metrics

Select

3D Visualization Model of Vessel Trajectory in Bridge Area Based on AIS LEI Jin-yu，CHU Xiu-min，XUWu-xiong 2016, 16(3): 88-94.  Abstract ( )   PDF (2491KB) ( )   With the rapid development of AIS (Automatic Identify System) applied in maritime transportation, large amounts of vessels’trajectory data are collected and employed to analyze vessel traffic. The AIS data usually contain multidimensional attributes including time attribute and speed attribute. However, the traditional visualization approaches only show the spatial properties of the AIS data, and rarely related to time attributes and other information. Therefore, a 3D visualization model, which can present time, latitude, longitude and speed simultaneously, is proposed for describing the multidimensional nature of AIS data in this paper. Additionally, a comparison experiment is carried out to illustrate that the 3D visualization model can prompt the understanding of the AIS data for analyzer, especially under the help of the time information. Finally, the visualization model is applied to explore the AIS data of bridge area and results indicate that the model can reflect the navigation situation in the bridge area properly. References | Related Articles | Metrics

Select

Feature Extraction Model of Urban Traffic Network Data Based on Locality Sensitive Discriminant Analysis Algorithm XU Li-xiang，WANG Yun-peng，YU Hai-yang 2016, 16(3): 95-100.  Abstract ( )   PDF (1435KB) ( )   To simplify the way of expressing road network state and maximize the value of network information, this paper constructs a model of extracting feature parameter from massive historical traffic data to express road network running state. In this model, the flow, speed and density data of road network in urban areas are selected, considering the nonlinearity and correlation of traffic data, the feature of urban road network data is extracted based on adaptive neighborhood selection of local sensitive discriminant analysis algorithm (ANS-LSDA). Examples demonstrate the effectiveness of the model, results show that feature parameter obtained in this paper can effectively describe the road network 24 h periodicity, directly reflect the phenomenon of morning and evening peak as well as the difference between weekday and weekend. Compared to kernel principal component analysis (KPCA), the feature parameter of ANS-LSDA model has better divisibility, which can express macro road network running state and provide basis for traffic managers in decision making References | Related Articles | Metrics

Select

A Bus Departure Time Interval Transition Model Considering Traffic Congestion DONG Hong-zhao，KONG Juan-juan，LIU Qing-hui 2016, 16(3): 101-106.  Abstract ( )   PDF (1281KB) ( )   The existing planning models of bus departure time interval usually ignored the important impact factor of real- time traffic congestion. To solve the issue, a departure time interval transition method is presented considering passenger travel demand and traffic congestion. The bus operating hours is divided into several temporary periods separately according to the passenger travel demand and the congestion conditions by the clustering method of samples with sequence. Then these temporary periods are combined to form a period set, in which each period possess the features including passenger travel flow and traffic congestion. Consequently, the departure interval transition model is established to adjust the transient time separately between two neighbor periods to match the passenger travel demand and departure plan. Finally a practical experiment for one bus route in Hangzhou is implemented to verify the proposed model by PARAMICS simulation. The result shows that the model can keep the load ratio in the range of 40% to 70% with the precondition of satisfying the passenger travel demand, and help to improve the service level and the efficiency of the bus transit system. References | Related Articles | Metrics

Select

A Multi-objective Bus Rapid Transit Dispatching Optimization Considering Multiple Types of Buses YANG Xin-feng, LIU Lan-fen,LI Yin-zhen, HE Rui-chun 2016, 16(3): 107-112.  Abstract ( )   PDF (1178KB) ( )   The reasonable departure frequency, bus type, loading rate and queuing time of buses affect the service quality of Bus Rapid Transit (BRT). The dispatching of BRT for multiple types of buses is studied. Firstly, the dwell time and arrival time of bus, passenger numbers on bus platform and failed to get on the bus are analyzed. Moreover, considering multiple types of buses, capacity of bus and BRT and platform capacity, a multi-objective departure frequency model of BRT is established to optimize the average loading rate, operating costs and queuing time of buses. Furthermore, the problem is divided into two sub-problems: bus queue and without queue, and an analytical method is put forward to find this multi-objective model’s Pareto solution set. At last, Lanzhou BRT line is chosen for the dispatching optimization model and the results illustrate the effectiveness of the optimization model and algorithm References | Related Articles | Metrics

Select

Optimizing Fares and Transfer Pricing Discounts for a Subway and Bus Corridor Considering Difference in Stop Distance between Two Services LIU Bing-zheng, GE Ying-en, CAO Kai 2016, 16(3): 113-119.  Abstract ( )   PDF (1219KB) ( )   Considering the stop spacing difference between subway and bus services, passengers served by both services are classified into three types according to their chosen trip paths. The fares and transfer pricing discounts of the two services are optimized in an integrated model that captures the interaction of different passenger type trip and the influence of it on optimal fares and transfer pricing discounts. For this purpose, a bi-level program is formulated, of which the upper level aims to maximize the social welfare and the lower level is a variable-demand stochastic user equilibrium assignment model. The genetic algorithm is applied to solve the bi-level program. Numerical experiments are carried out to illustrate the performance of the model. The results show that: ① it is of great significance to embed the factor of stop spacing difference between subway and bus services into the integrated optimization of fares and transfer pricing discounts of the two services, otherwise overcrowding may emerge in a set of routes in each service line; ② the optimal fares of each service can increase and both the differences between fares of the two services and transfer pricing discounts are reduced considering the stop spacing difference References | Related Articles | Metrics

Select

Optimization of Classification Track Assignment Based on the Railcar Connection Plan in Hump Yard ZHANG Ying-qun, SONG Rui, LI Hao-dong 2016, 16(3): 120-125.  Abstract ( )   PDF (1651KB) ( )   Classification track assignment decides how to build each outbound train on different classification tracks in hump yard based on the railcar connection plan and plays an important role in rail yard. In this paper, an optimization model for classification track assignment is proposed, which aims to minimize the number of mixed tracks and pullback operations. The constraints of this model mainly include track capacity constraint, and track number constraint for each outbound train. Tabu Search algorithm is proposed to solve the model and the result is compared with IBM ILOG CEPLEX. As last, a numerical example is presented to valid the proposed model and algorithm. The result shows that the classification track assignment is influenced by the number and the length of classification tracks, the railcar connection plan and the layout of rail yard. References | Related Articles | Metrics

Select

Analysis of the Steady State Performance of Traffic System Based on Max-plus Theory HAN Yun-xiang, HUANG Xiao-qiong 2016, 16(3): 126-132.  Abstract ( )   PDF (1232KB) ( )   For the running process of the train in the rail transit line, the urban railway traffic system is regarded as a discrete event dynamic system. Based on the rail transit system dispatching rule and railway network space distribution structure, the single train is viewed as a batch in order to obtain the optimization scheduling scheme of rail transit system. Under the maximal algebraic framework, the open loop and closed loop control models of the single link and multi section traffic flow are established. Besides, the constraint relationship between the input variables, state variables and output variables is obtained. Starting from the state evolution of the rail transit system, the influence of different train scheduling forms on the system performance is studied based on the perturbation analysis method. The performance of several kinds of sorting methods and optimal system scheduling scheme are presented regarding rail transit system evolution process shortest period as optimization objective. Furthermore, the system robustness in the case of single element and multi element perturbation is put forward in order to maintain the steady state system evolution parameter. The research results can provide reference for the traffic system control. References | Related Articles | Metrics

Select

Platoon Dispersion Model Based on Mixed Phase Distribution of Speed YAO Zhi-hong, JIANG Yang-sheng, WU Yun-xia, LIU Yu-qiao 2016, 16(3): 133-140.  Abstract ( )   PDF (1576KB) ( )   The platoon dispersion model is used to depict the relationship of the downstream flow profiles to upstream flow profiles, which could provide theoretical support for signal control and other applications. To evaluate the effect of the models, it is critical for them to accurately describe the characteristic of mixed traffic flow and possess an analytical solution. Here the truncated mixed phase distribution (TMPH) is proposed, which has the merits of universality and analyticity. Then we can obtain the mixed platoon flow dispersion model based on this distribution. By analyzing the field data, the TMPH can fully reflect the characteristics of mixed traffic flow. Then on this basis, we can derive the mixed platoon dispersion model. Furthermore, through the comparative analysis of the proposed model, Robertson’s model，mixed Gaussian model and actual arriving traffic flow distribution, we found that the proposed model not only has good analyticity, but also could better reflect the platoon dispersion laws of mixed traffic flow. References | Related Articles | Metrics

Select

A Capacity Simulation Model of Stairway in Subway Station Considering Pedestrian Heterogeneity WU Hao-ling，YUAN Zhen-zhou，LI Hui-xuan，CAO Zhi-chao，ZHANG Si-lin 2016, 16(3): 141-147.  Abstract ( )   PDF (1671KB) ( )   Pedestrians are regarded as homogeneous ones without their own individual characteristics in the current study on the capacity of subway stairways. But actually, pedestrians are different individuals with heterogeneous characteristics. Due to the lack of the research on this issue, this paper classifies persons into pedestrians in pairs, fast ones and common ones by introducing the concept of heterogeneous pedestrian and analyzing the observed data of the subway station stairway. Then the improved floor field model, a capacity simulation model of stairway is presented. We validate the precisions of the proposed model under the conditions of homogeneous and heterogeneous pedestrians. At last, three situations are simulated and analyzed based on different space occupied, different speed, different space occupied and speed respectively. The results show that the capacity of stairway is not a constant value and it changes with different component proportions of heterogeneous pedestrians. The heterogeneity of pedestrian has an important influence on the capacity of the stairway. References | Related Articles | Metrics

Select

Platform Width Calculation at One-platform-transfer Station of Urban Rail Transit LI Ying, MAO Bao-hua, DU Peng，CHEN Zhi-jie 2016, 16(3): 148-153.  Abstract ( )   PDF (1538KB) ( )   This paper mainly studies the width calculation of one-platform-transfer station. The passengers’ dynamic distribution law is studied and a method of calculating the maximum assembling on one-platformtransfer station is given. The control point of calculating the width is analyzed. In the presented method, the two waiting areas are calculated on a consolidated basis, which is different from the current code. Further, in a case of an one-platform-transfer station, the result is 11.8 m which is bigger than 9.8 m in China’s current code for design of metro, smaller than 12.4 m in American’s B level. In the end, with the help of the software, the pedestrian density is tested under the three ways. In the current code, the pedestrian density in the waiting area reaches 2.5 person/m2, and 1.8 person/m2 in American’s B level, while using the presented way, the result is 2.0 person/m2. In the distribution area, the pedestrian density in China’s current code reaches 0.8 person/m2, and 0.5 person/m2 in American’s B level, while using the presented way, the result is 0.6 person/m2. References | Related Articles | Metrics

Select

Container Ship Speed Optimization under Influence of Wind and Wave WEI Zhao-kun，XIE Xin-lian，WEI Ming 2016, 16(3): 154-160.  Abstract ( )   PDF (1437KB) ( )   With enhancing of environmental awareness, shipping market is suffering a downturn that makes shipping companies reduce the fuel consumption for controlling ocean pollution and operating costs when transportation tasks can be completed. Fuel consumption is relation to the speed which is influenced by climate such as wave, wind and so on. To deal with speed optimization problem, the mix integer non-linear mathematical model is proposed based on the influence of wave and wind that is climate factor. Due to the complex of mix integer non- linear problem, it is difficult to be solved. Considering the actual situation of ships, the sailing speed is dispersed. At the same time, we transform the MINP model into MILP. The CPLEX is used to solve MILP model efficiently. By the algorithm of dispersing speed, the optimal speed and the number of ships that is in different voyages with various wave and wind conditions can be obtained. Experiments indicate that the method can apply reasonable ship schedules for shipping companies. References | Related Articles | Metrics

Select

Container Terminal Loading and Unloading Task Scheduling Based on Computational Thinking LI Bin 2016, 16(3): 161-167.  Abstract ( )   PDF (2155KB) ( )   Task scheduling in container terminal logistics systems (CTLS) has been one of the difficulties and focus in the field of logistics engineering. The existing programming, simulation and optimization are all limited and lack of universality while they are applied to CTLS. Computational thinking provides a new solution to control and decision on CTLS. This paper centers on the concept of process and thread in operating system (OS), and consults the task scheduling framework, mechanism, pattern and algorithm in OS, and proposes container terminal logistics scheduling and decision computing architecture (CTLSDCA). Furthermore, for a certain large-scale container terminal operation scene, a comprehensive scheduling algorithm with computational thinking is devised by the integration of multiple scheduling mechanisms and policies in OS. Finally, a lot of simulation and analysis for the case scene is executed and given to demonstrate the superiority of CTLSDCA. References | Related Articles | Metrics

Select

Hama-based Parallel Ant Colony Optimization Algorithm Mode and TSP Applied Research MAJi-hui, XU Ming-jie, CHEN Xin-jie, SONG Cui-ying, YANG Yang 2016, 16(3): 168-173.  Abstract ( )   PDF (1416KB) ( )   Hama is a large-scale parallel computing tools, which works on Hadoop platform and is based on the open parallel computing platform of BSP computing technology. Its main function is to support the scientific calculation of parallel and big data. To enhance the efficiency, many algorithms are transplanted to Hama platform, which becomes an advanced research hotspot. Ant colony algorithm is a highly adaptive heuristic algorithm, which is widely used. Because of the randomness of the individual, there is a contradiction between the convergence speed and the diversity and stability of the solution. But this contradiction can be alleviated by the paralleled ant colony algorithm, and the performance of the algorithm can be improved correspondingly. This paper tries to realize a parallel ant colony algorithm, which based within the framework of Bulk Synchronous Parallel. The experimental results show that the parallel ant colony algorithm is feasible and effectively improves the performance. References | Related Articles | Metrics

Select

Equilibrium Problem of Last-mile Delivery Service Choice Based on Different Interest Participants CHEN Yi-you, ZHANG Jin, ZENG Qian 2016, 16(3): 174-180.  Abstract ( )   PDF (1358KB) ( )   To model the tradeoff between the customer and the enterprise in the choice of pick-up service and home delivery service, the equilibrium problem of last-mile delivery service choice based on different interest participants is studied. The pick-up point is modeled as an M/M/1/1 queue and home delivery as an M/D/1 queue. From an expectation utility- maximizing customer’s perspective and an expectation profitmaximizing enterprise’s perspective, delivery service choice equilibrium models of boundedly rational customers and full rational enterprises are respectively constructed. Based on game theory, the existence and uniqueness of Logit equilibrium and Wardrop equilibrium in last- mile delivery service system are proved. Numerical experiments verify the correctness of the model, and reveal that there is a big gap between the customer and the enterprise, when they select delivery service. The charge and level of customer bounded rationality affect the equilibrium, customer’s utility and enterprise’s profit significantly. References | Related Articles | Metrics

Select

Dynamic Shortest Paths of Emergency Vehicles in Expressway Network Based on Shuffled Frog Leaping Algorithm DUAN Xiao-hong，ZHAO Jian-dong，SONG Shou-xin 2016, 16(3): 181-186.  Abstract ( )   PDF (1480KB) ( )   Considering the discrete dynamic characteristic of the expressway network, a novel shuffled frog leaping algorithm is proposed to solve the shortest paths of emergency vehicles. In this paper, a random coding scheme is designed firstly, and then the concept of node travel time degree is introduced to improve the coding scheme. Then a backward marking update rule is put forward. The entering node moment and the link travel time of the worst individual are compared with those of the advantaged group, and the update rule makes the vehicle enter a shorter road section at the earlier time. Finally, the dynamic link travel time distributed in a normal is used as the weights, and the algorithm is verified by the example of expressway network of Dong Cheng District and Chao Yang District in Beijing. The results show that the proposed shuffled frog leaping algorithm can search the shortest path in one second. The calculation speed of the algorithm based on node travel time degree is raised by 100% comparing to the random coding scheme, and the average accuracy rate of the former is raised by 4.3% comparing to the latter. References | Related Articles | Metrics

Select

Robustness Analysis of Hierarchical Airline Network Based on Penalty Factor PENG Ting, ZHANG Ya-ping, CHENG Shao-wu 2016, 16(3): 187-193.  Abstract ( )   PDF (2191KB) ( )   The robust performance of Chinese domestic airline network (CAN) is systematically analyzed in this paper. In order to consider isolated points (or failure points) which produced by the deleting of edges (or nodes), the concept of penalty factor is introduced. And then, the connectivity efficiency of airline network is characterized by the average shortest distance. At last, the robust performance of airline network is evaluated according to the change rules of the average shortest distance. In addition, CAN is divided into trunk network and branch network, and the differences of network robust performance are systematically compared when the edges (or nodes) of different hierarchies are deleted. Results indicate that CAN is of better robustness to the edge deletion of trunk network under the condition of edge failure. However, under the condition of node failure, it is of better robustness to the node deletion of branch network. In addition, compare to the condition of edge failure, the Chinese domestic airline network is more vulnerable under the condition of node failure. References | Related Articles | Metrics

Select

Optimization of Contra-flow Links in Traffic Evacuation Based on the Critical Edge for Improving Maximum Flow of a Network GAO Ming-xia, HONGWa-yi, SI Qian-qian 2016, 16(3): 194-199.  Abstract ( )   PDF (1300KB) ( )   Contra- flow got effective application in traffic evacuation as a method of traffic organization. The problem of choosing contra-flow links in traffic evacuation is studied with network optimization technique. A road network is abstracted as a directed network, and the critical edge for increasing the maximum flow value of a network is defined, and alternative links for contra-flow can be got by searching critical and sub critical edges etc. for increasing the maximum flow value. A modified algorithm is presented to find such critical edges on the basis of the maximal capacity path algorithm for the maximum flow problem. A numerical example is given and the effects are tested through traffic simulation. The results show that evacuation time can be effectively decreased through increasing the capacity of critical links by contraflow in large-scale evacuation. References | Related Articles | Metrics

Select

A Method of Vehicle Fuel Consumption Estimation Based on Decision Tree ZHU Guang-yu, ZHAO Lei, HUANG Da, ZHANG Peng, BIAN Li-qin 2016, 16(3): 200-206.  Abstract ( )   PDF (1493KB) ( )   Vehicle fuel consumption is one of the important indicators of road construction post-evaluation. At the same time, it is the decision basis for pavement design, the location of gas stations, route choice and so on. Traditional vehicle fuel consumption evaluation mainly uses the method of constructing the regression model, and a vehicle fuel consumption estimation model based on decision tree and data mining is studied in this paper. Namely: firstly, the key factors affecting the vehicle fuel consumption are obtained using principal component analysis method; secondly, the vehicle fuel consumption estimation model is built based on the improved C4.5 decision tree; finally, validate the model presented in this paper combined with a representative sample data of vehicle fuel consumption at freeway scene. By analyzing error of vehicle fuel consumption estimated value and the real value, it shows the effectiveness and practicality of the proposed model. References | Related Articles | Metrics

Select

Bi-level Programming and Algorithm on the Location of Arterial Highway Management Station CHENG Xue-qing, JIA Jiang-tao, LI Yue, LIU Xing-wen 2016, 16(3): 207-213.  Abstract ( )   PDF (1421KB) ( )   Reasonable location of arterial highway management station is important to improve disaster response speed and transport efficiency of relief supplies. Taking into full account the function and nature of the arterial highway management station, with the rational model assumes, the paper established a bi- level programming model including total weighted minimum distance of road management station and the locations of material demand as the upper target while materials transportation fleet shortest running time as the lower target. Improved genetic algorithm based on linear stretching method is designed to solve the bilevel programming model. In an example of arterial highway in the region, according to historical data and emergency conservation, the paper selects 14 material needs points and 8 management station candidate points. Finally, it obtains the management stations sitting plan and verifies the feasibility of the model and algorithm by MATLAB programming. References | Related Articles | Metrics

Select

Apron Taxiway Operation Mode and Simulation Evaluation in Beijing New Airport WU Hao-ning, GUO Yan-chi, MU Tong, LI Xiong 2016, 16(3): 214-220.  Abstract ( )   PDF (2412KB) ( )   Apron taxiways are the important active areas for the flight power-in or push-out gate. The smooth operation of the apron is directly related to the ground operation efficiency of the airport. First, the main operating modes of the apron taxiways in large scale airport are systematically analyzed. Then, using Beijing New Airport (BNA) as the research object, three apron taxiway operational patterns are carried out, respectively dual-channel and unidirectional operation pattern (DCU), setting push-out waiting position operation pattern (SPWP), triple-channel with one-in and two-out operation pattern (TC). Finally, Simmod simulation software is operated to construct the computer simulation models for different utilization patterns of apron taxiways. The simulation result shows that the ground delay time of arrival and departure flights in TC operation mode is the least, compared with the other two kinds of operation patterns, the average reduction of the departure ground delay in different flights is about 10%~15%. Departure flights take off waiting delay accounted for more than 50%, the terminal angle area delay accounted for about 20%; arrival flights runway crossing leading to delays accounted for about 50%. References | Related Articles | Metrics

Select

Empirical Analysis of Speed-density Relationships for Urban Expressway under Rainstorm Condition SUN Hong-yun, YANG Jin-shun, WU Bing 2016, 16(3): 221-227.  Abstract ( )   PDF (1794KB) ( )   This paper investigates the impacts of rainstorm condition on traffic operation of multi-lane urban expressways. Firstly, reductions in free- flow speed and capacity are calculated. Secondly, six oneregime speed-density models are calibrated by least square method. Finally, best models for key parameter estimation and for operating speed prediction are assessed. Results show that reductions in free-flow speed and capacity, and the abilities to estimate key parameter and to predict operating speed vary with different number of lanes under rainstorm condition. Therefore, it necessitates distinguishing rainstorm' s impacts on urban expressway segments with different number of lanes. References | Related Articles | Metrics

Select

Experimental Study of Passenger Evacuation Efficiency in a Metro Station LI Tao，CHEN Juan, MAJian，LO Siu-ming 2016, 16(3): 228-234.  Abstract ( )   PDF (1658KB) ( )   The rapid development of metro system has attracted a large amount of passengers, which makes the ground road network much less congested. However, how to safely evacuate these passengers in the metro system in case of emergency becomes more and more urgent to be solved. The evacuation process of metro station is special because the evacuation is towards upward direction, which is different in normal buildings where people move downward to complete their evacuation. There still lacks studies on this kind of evacuation. Thus this paper in the present study designs and performs a real passenger participated metro station evacuation. Based on the experiment, the emergency evacuation efficiency of escalator and stair in the subway station, as well as the performance of turnstiles and gate- door passengers control devices in the station are extracted and analyzed, data including the average evacuation number of passengers and evacuation time, the flow rates are discussed and compared. Results indicate that escalator performs better than stairs when used in evacuation; the capacity of turnstiles can be larger than regulation values and evacuation gate-door performance is much stable. The research of this paper not only provides basic experimental data for the study of large scale metro passenger evacuation, but also provides a useful reference for the development of the underground emergency evacuation plan. References | Related Articles | Metrics

Select

Theories and Confirmed Model of Household’s Activity-travel Behavior Based on Social Network HE Bao-hong, LIU Yang, HE Yan, LI Zhen-hui 2016, 16(3): 235-240.  Abstract ( )   PDF (1355KB) ( )   The joint decision-making behaviors such as companion, substitution, complementary exist generally in family activities. However, ignoring the irrational characteristics of traveler as a society and the self-organization characteristics of interactions between family members, most studies on travel behavior take individual as decision-making unit and model based on the utility theory of microeconomics. On the basis of time and space constraints, this paper emphasizes the influence of the social relationship to individual behavior, integrates time geography and social network theory of study to individual. Firstly, a role-measured indicator is put forward to reflect the position of individual in family network structure, and we obtain the typical household interaction modes and their characteristics. After that, a SEM of individual activity-travel choice is constructed based on obviously interaction network of family relationships. In conclusion, the external constraints of individual for participating activities not have a directly effect on individual’s own, but balanced through internal network relationship of family, the low-restrict member role who has a higher- priority of household resources would like to help the high- restrict member role. Especially, considering family structure is not only benefit a refined model, but also help a good understand for model, which ultimately realizes the full appreciation for individual’s activity-travel behavior. References | Related Articles | Metrics