Research On Dynamic Multi-stage Scheduling Decisions Model Of Railway Emergency Resources And Its Application

In recent years, with the increase of the railway operation mileage, introductions of the new equipment and technology after the speed-raising, the increasing rate of traffic flow, frequent natural disasters and the affection of man-made factors, the railway emergency event shows the increasing of numbers and kinds and increscent losses. How to deal with the railway emergency event scientifically and effectively, minimize the incidental losses has become a focus issue in today's society and the railway departments..Once the railway emergency event happens, the key to rescue treatment quickly, control development of the events timely and reduce disaster loss is the reasonable railway emergency resource scheduling decisions. At present, the study of emergency resources scheduling decisions adopts mainly the nearest principle or principle of the emergency point priorities to established the scheduling decisions model. The model based on the nearby principle can ensure that emergency resources is dispatched from the nearest rescue points to each emergency point on each stage, but it can't balance the severity between different emergency points. The model based on emergency priority principle divides the priorities according to the severity of the different emergency points, first ensures the high priorities emergency needs to be met, but it weakens the most important time factors in the process of the emergency resource scheduling, and the utility that different emergency resources is different in each emergency point, only considering the priorities of itself is not enough. At the same time, most of scheduling model is built on the basis of static demand and allocates each kind of resource when confirms fixed quantity of demand for each emergency point. It ignores the dynamic change characteristics of emergency resource demand and emergency points competition relation for resources in emergency resource shortage situation, may lead to local surplus or shortage of emergency resource allocation, then makes the system total losses not to the optimal value, because of the emergency scheduling decision-making errors.Therefore, on the basis of the predecessors' research results, combining the advantages of all kinds of emergency resources scheduling model and contraposing the deficiency of in the existing models, the thesis makes the further research on the railway emergency resources dynamic multi-stage scheduling decisions model of multi-emergency points. First of all, through the analysis of the characteristics of the railway emergency resource demand, introduces fuzzy comprehensive evaluation method to establish the evaluation model for the railway emergency resource demand urgent degree, thus calculates the emergency resource demand urgent degrees specific value of the multiple emergency points as the important parameter that can distinguish the different emergency resource demand utility value in the railway emergency resources dynamic multi-stage scheduling model. After that, analyzes the characteristics of the emergency resources scheduling model for the huge and complicated railway system, combines with the advantage of the nearest principle and the priorities principle, adds the dynamic demand function in traditional static model, uses the emergency point time accumulation of resource wantage to describe system loss, and considering emergency resource demand urgent degree and capacity constraints in the process of the emergency resources scheduling, using game theory to describe the multi- points emergency resources scheduling as a multi-stage process of non-cooperative game, establishes the railway emergency resources dynamic multi-stage scheduling non-cooperative game model of multi-emergency points — multi-rescue points — variety resources with the target of minimizing the system total loss. Since then, in order to meet the time urgency requirements of the railway emergency decision and to accelerate the speed of solving the model further, the thesis designs an improved cuckoo algorithm which is suitable for solving the railway emergency resources dynamic multi-stage scheduling model, and uses the algorithm to solve the concrete example of railway emergency resources dynamic multi-stage scheduling in the MATLAB environment. It verifies the feasibility of the model and the superiority of the algorithm. Finally, in the Visual Studio 2010 environment, uses the SQL Server to manage the database and realizes the application of the railway emergency resources dynamic multi-stage scheduling model in the railway emergency resource scheduling system by means of MATLAB and VB mixed programming. The system achieves the accurate calculation of the railway emergency resources dynamic multistage scheduling model under the relevant algorithm.