Research And Design On Models Of Automobile Loss Synthetic Evaluation Decision Support System

To confirm the vehicles' loss is a both important and complex work of motor vehicle insurance.It not only needs a uniform loss confirmed standard,but it also needs the loss adjusters' experience and professional knowledge.It is vehicle insurance industry's urgent problem currently that how to guide people make a scientific,rational and correct assessment through computer system and artificial intelligence technical.This dissertation's purpose is to tackle the problems with the studies and design on models of automobile loss synthetic evaluation decision support system,which based on the relative theories and technologies of synthetic evaluation and decision support system.The main contents are as follows:(1) It proposes the theory basis for the system modeling of automobile loss synthetic evaluation.The basic theories of fuzzy math,grey theory and neural network have been deeply analyzed and researched.And their features and application in synthetic evaluation are discussed.(2)According to the components of automotive,the two-level indicator system model of automobile loss synthetic evaluation is established.(3)In allusion to request and character of automobile loss synthetic evaluation,the automobile loss neural network evaluation model is built,and the generalized analysis and detailed discuss to its performance is carried on.(4) The system architecture of automobile loss synthetic evaluation decision support system is established.And a prototype system has been designed and implemented,which has completed the predestinate function and performance requirements.At present,the prototype system of automobile loss synthetic evaluation decision support system has been put into practical trail stage.The system can effectively guide loss adjusters to make scientific and reasonable assessments on the vehicles' loss,and it has been widely acclaimed.