Research On Optimization Of Wagon Wheel Maintenance Equipment Investment Based On Stochastic Programming

With the transformation of the railway wagon maintenance method from maintenance by plan to maintenance by status,railway wagon maintenance companies need to update and upgrade existing maintenance equipment.However,there are many uncertainties in the transformation of the maintenance strategy.One-time large-scale fixed asset investment will bring huge risks to the enterprise,so the research on the decision-making of investment in maintenance equipment under uncertain environment is of great significance for the railway wagon maintenance enterprises to steadily promote the maintenance strategy reform.In this thesis,the investment problem of wheel maintenance equipment is taken as the example,and the research on the optimization problem of wheel maintenance equipment investment under uncertain environment is carried out.First,the wheel maintenance amounts and equipment market price are selected as the random parameters of this study,and based on the reliability function,the transformation of wheel failure law into maintenance demand prediction is realized,then the moment matching method is used to complete the generation of the initial set of scenarios in each decision period and the initial scenario tree is also been generated according to the scene construction method.In order to avoid the difficulty in solving the mathmatical model due to the large scale of initial scenario set,the reduction of the initial scenario set was completed based on the backward reduction method and probability measure evaluation of the reduction algorithm is also be conducted.Finally,the optimal set of retention scenarios is determined.Based on the mixed integer linear programming model,this thesis builds a deterministic model and a stochastic programming model for the wheel maintenance equipment investment optimization problem based on the stochastic programming framework.In the General Algebraic Modeling System,the assignment of the two models is solved,and the decisions of purchasing equipment with the corresponding investment cost are also given.In order to realize the dynamic update of the model and highlight the characteristics of the delayed decision of the stochastic programming method,this thesis uses the rolling horizon procedure to dynamically solve the model.At the same time,the results of different solutions in different data scenarios are also discussed.The results show that,in the random environment,the stochastic programming method based on rolling horizon procedure update can bring better decisions than the deterministic planning and one-time-solving stochastic programming methods.In order to verify the correctness of the stochastic programming method and the work done in the previous chapters,this thesis carries out the test of the model and method from two dimensions.First for the model feasibility vertification,this study introduces two evaluation indicators to demonstrate the value of information and the superiority of the method,The calculation results show that taking more information into the decisionmaking process will bring about a more significant cost reduction,and the use of stochastic programming in the random environment is relatively better than deterministic planning.For the test of model stability,this study uses two methods: in-sample test and out-of-sample test.The test results show that the size of the retention scenario set selected in this thesis is reasonable and the model proposed in this thesis is in sample and out of sample stable.Through the study of this thesis,the optimal investment decision of railway wagon maintenance equipment under uncertain conditions is determined,which reduces the investment cost of enterprises and provides certain theoretical support for railway wagon maintenance enterprises to steadily promote the reform of maintenance strategy based on the state.