In order to investigate the current situation of efficiency for rail transport,and provide a theoretical guidance in improving efficiency,a class of network data envelopment analysis model was constructed through re-viewing relative undesirable outputs of existing literatures and comparing their advantages and shortcomings.In this model,rail transport system was divided into passenger service process,freight service process and freight business process.Then it measures the network efficiency ,technical efficiency of each sub -process for China's railway transport from 2003 to 2012.The results show that the efficiency of transport service process is satisfacto-ry,but the efficiency of freight business process is lower,which affects the network efficiency of rail transport to some extent.Comparing with traditional DEA model,the new model is more practical because it determines the source of the inefficiency of railway transport well,which proposes theoretical basis and strategies for improving operational performance.

In order to study the relationship between transportation and economic growth from the macro, in this p per the concepts of impact factor and conversion efficiency of transportation level are introduced. Based on the analysis of New-Classical Growth Theory, an economic growth model with transportation factors IS set up. Through analysis of the model the action mechanism and influence path of transportation on regional economic growth is revealed. And the results indicate that the positive equilibrium economic growth rate exists when impact factor transportation level satisfies certain condition. And economic growth is inversely related to impact factor of transportation level. This model has some practical values for transportation planning and the coordinated development of regional economic and transportation.

In data mining, fuzzy C-means clustering algorithm has demonstrated advantage in dealing with the challenges posed by large collections of vague and uncertain data. This paper reviews concept of fuzzy C-means clustering which is widely used in context of pattern recognition. Based on the study of the fuzzy C-means algorithm, we propose a weighted local fuzzy regression model. The efficiency of the new modified model is demonstrated on real data from 1980 to 2010 collected for road freight of China. And with this method, analysis of correlation between economic development, transport facilities and demand for road transport is given. At last, we apply this method to predict the demand for road freight of China from 2011 to 2013. The results shown that weighed fuzzy model has more practical value than least squares regression on prediction problems with the small sample, non-linear and high-dimensional pattern recognition of transport system.

According to the fact that users are interested in the changes of multiple interest measures of association rules,a new kind of emerging pattern is presented which considers not only the changes of support but also the changes of confidence and other measures.Based on the Pareto sorting,a corresponding mining algorithm for Emerging Pattern is designed.Empirical evaluation shows that the proposed algorithm is very effective to recognize emerging patterns from two-period datasets.

In order to improve the shortcomings of change mining algorithms of association rules mined from two datasets from different time periods,this paper presents a new change mining algorithms of association rules with mixed attributes.The research makes an attempt to account for the distance between the attribute-values according to different attribute types while measuring the similarity between two patterns,and then identifies accurate change patterns.The experiment results show that the proposed algorithm has good performance.

Aiming at the shortcomings of existing change mining methods from multi-period datasets, the concept of abrupt rules based on two existing types of changes, stable rules and those which exhibit trends, is presented. The rules that outliers exist in support or confidence time series are defined as abrupt rules, and the Grubbs test is used to discover the abrupt rules. The abrupt rules exhibit the abrupt changes of customer shopping behavior, which can reflect the dynamic information of customer shopping patterns with time comprehensively and help managers take effective strategies timely to gain competitive advantage. In order to evaluate the degree of change, the average value of support or confidence of rules is used as the interest measure for stable rules; the Kendall slope of the Mann-Kendall test is used as the degree of change for rules which exhibit trends; and the departure degree of outliers is used as the interest measure for abrupt rules. The results of empirical evaluation show that the proposed method can effectively recognize the change of customer shopping patterns from multi-period datasets.

Based on the best splitting point,a change mining methods of customer classification is proposed,which conquers the shortcomings of the existing change mining methods.In terms of the features of classification rule,the new definition of change types is proposed,based on which the metric of change and degree of change measures are redefined.In order not to affect the search for the best splitting point of each node in the decision tree,an extended measure of value match of quantitative attribute is designed to analyze the change measures between patterns.Empirical evaluation shows that the methodology is very effective to recognize the change of customer classification from two different datasets.

Bases on the stochastic models of economic optimization,this paper adopts Prescott's paradigm to analyze the problem of optimal allocation of investment for the co-development of a region and its transportation,and develops a stochastic optimal investment model for the coordinated development.Furthermore,it derives an Euler equation for the optimum allocation,performs a numerical experiment based on the proposed model with the statistical data of Jiangsu province.The numerical results show that Jiangsu's development is good.The results are helpful for Jiangsu to allocate optimally its total investment in transport and other areas.The detailed suggestions for improving development of Jiangsu's transportation are presented as well.

To identify the stocks that have entered the bull wave is of great significance to maximize returns on investment.Discussions have been made in this article to define the start of optimal investment region and set up the evaluation index system and membership functions based on HWSME and fuzzy comprehensive evaluation method.Through the interdisciplinary and diachronic analysis of some stocks in Chinese stock market,results have come to prove the validity of the method,which can provide a powerful analytical tool in maximizing returns on investment.

About the price of railway,traditional solution is to get the sole price,event the dynamic price,the set of prices is also assigned.This paper mainly studies the dynamic pricing of railway under the competition between short-distance road and railway based on the sensitivity analysis and bi-level programming model,gets the dynamic price according to a nonlinear programming model and the data indicates that the income based on the dynamic price is higher than the sole price.The results indicate the model is feasible.

A simple network DEA (data envelopment analysis) model is proposed in terms of directional distance function under the assumption of invariable returns to scale. With this model, air transport is divided into stages of production and service and the efficiency of the air transport system in 16 selected airlines in China for the year 2008 is analyzed. This model not only represents the airline technical efficiency in the whole transport process, but also gives both production and service efficiency in two different stages. With these results, the influence of production and consumption efficiency on airlines transportation system efficiency can be identified, and thus the strategies for improving operational performance can be proposed. This new model is more practical than the traditional data envelopment analysis model.

In view of the undesirable forecasting results from the conventional group method of data handling(GMDH) network modeling that uses the least square method to identify coefficients,this paper introduces a fuzzy reasoning model into the GMDH network to take the place of partial descriptions of the traditional GMDH network,i.e.the integral binary quadratic polynomial forms.Based on the GMDH network,this paper proposes a model of traffic volume forecasting based on a fuzzy GMDH network.Simulation results show that the average relative error of the new model prediction is only 2.31%,and it is less than 3.35% which is the average relative error of prediction of the traditional GMDH network.The new model is proved effective.

A novel algorithm is presented on the base of quantum behaved particle swarm optimization,which is aimed at resolving the problem of slow convergence rate in optimizing higher dimensional sophisticated functions and being trapped into local minima easily.Chaos algorithm is incorporated to traverse the whole solution space;besides a new strategy of self adapting adjustment of inertia weight according to the current particles'fitness is combined also to balance the capability of local search and global search.The experimental results of some typical trial functions show that the proposed algorithm not only has great advantage of fast convergence rate and computational precision in solution,but also can avoid the premature effectively with a better performance than Standard Particle Swarm Optimization(SPSO)and Quantum behaved Particle Swarm Optimization(QPSO).

On the stochastic models of economic optimization of Prof. Chen M F, this paper chooses Prescott's paradigm to analyze China's optimum economic growth, finds a stochastic optimal control model for China's economic growth, acquires a worthy Euler equation of this model, and gives the optimal numerical solution of this model, which would help planners could easily allocate investment of capital and labor on China's economy best.

科学准确地预测公路货运量是制定公路网规划的基础。文中运用灰色Verhulst模型和一次指数平滑模型相结合的组合预测模型对我国的公路货运量进行预测。结果表明,组合预测模型能够提高我国公路货运量的预测精度。

With Chen M F's original researches on the stochastic models of economic optimization, this paper finds Prescott's paradigm is a suitable method to integrate LU Da-dao's regional balanced development theory with ZHANG Jin-shui's dynamic economics control model. As results of the integration process, this paper acquires a stochastic optimal control investment model for the coordinated developing of transportation and non-transportation within a region in China, derives an effective Euler equation for the optimum allocation. This paper does a numerical experiment using the model and the data of Jiangsu province, finds the numerical results show Jiangsu's developing well and the results helpful for Jiangsu to distribute optimally its total investment in transport and transport else, and presents concrete proposals for improving the developing of Jiangsu's transportation.

In view of the situations that urban water demand have trend of growth and periodicity,and urban water management system have definite estimating parameters,a discrete time system dynamic optimization model of urban water supply and demand management was constructed.Based on predicting of urban water demand and controlling of urban water supply investment reasonable,the urban water demand could be satisfied.The system's optimal control equation and optimal trajectory was solved by means of integrating discrete maximum principle with singular optimal control.With implemented dynamic optimal investment decision-making on the system,the total square of biased deviation of urban water resources supply and demand could be minimized in the period of urban water demand growing.The simulation results show that the system has strong sensitive dependence to initial value under certain parameter condition,so it should be determined prudently;in addition,the results indicate the significance of keeping the coordinated increment speed between urban water supply investment upper limit and the water demand.

We derived a nonlocal kinetic traffic model from a generalized car-following model with multiple look-ahead through the BBGKY hierarchy. A nonlocal hydrodynamical model was obtained consistently by the method of kinetic theory. The novel feature of nonlocality in equilibrium velocity and variance as well as in relaxation time was discovered. Simulation results showed that traffic hysteresis was induced by anticipation-dependent equilibrium velocity, relaxation time and equilibrium variance, indicating a 2D-region of synchronized flow. When drivers looked farther ahead, the amplitude of stop-and-go traffic reduced, and the resulting equilibrium velocity and flow increased, thus traffic stabilized as expected. Numerical results also indicated the existence of optimum anticipation vision. Finally, the model could also capture the physical nonlinear dynamic characteristics of metastability and the stop-and-go phenomenon in traffic flow.

On the stochastic models of economic optimization of Prof. Chen M F, this paper chooses Prescott's paradigm to analyze how optimally to allocate the investment of Jiangsu Province to coordinate the comovements of Jiangsu's transportation and non-transportation, acquires a optimal stochastic controllable model for the coordinated developing, finds a worthy Euler equation of this model, does a numerical experiment on the model with the statistics of Jiangsu province. The numerical results describe Jiangsu's development well, and the results are helpful for Jiangsu to distribute optimally its total investment in transport and transport else. At last this paper presents concrete proposals for improving the developing of Jiangsu's transportation.

In order to enhance the level of public transport services, this paper presents a multi-agent model of public transport, including guidance and management of multi-agent. The authors set forth the role of public transport agents, to put forward the components of public transport multi-agents model, as well as the specific contents of guide agent components and management agent components; Finally, the analysis of coordinated between guide agents and management agents of the public transport system is given, it demonstrate the feasibility of the establishment of guide multi-agent system and management multi-agent system of urban public transport for improve public transport services.