Research On Traffic Operation Situation Prediction Method Based On Online Mapdata Mining

The urban transportation system is increasingly perfect,and the options for residents’ travel are also increasingly diversified.Travelers can use one or several modes of transportation to reach their destination according to their own preferences.This series of changes makes travel more convenient.It complicates the choice behavior of residents,which makes it difficult for the management of traffic management departments and the release of ATIS information.The existing literature studies traditional single-mode travel behaviors,and rarely analyzes them in multi-modal transportation systems.In the field of multi-modal research,there are many studies on traffic distribution,and there are few research options from the perspective of individual travellers.Some studies are basically based on the theory of expected utility.This paper attempts to consider the risk attitude of travellers in the framework of prospect theory based on multi-mode networks to study travel behavior and establish a selection model.main tasks as follows:Firstly,the travel modes studied in this paper are four modes: car,bus,track,bus and track transfer.By analyzing the purpose of travel,the factors affecting travel and the bounded rationality of travellers in uncertain environment,this paper The travel time is the main analysis object,and the travel behavior of the traveler in commuting travel is studied.Then,the individual risk attitudes were studied in depth,and they were differentiated from the risk appetite of prospect theory,and the measurement methods of individual risk attitude levels were given.Compare the theory of expected utility with the theory of prospects,and analyze the content of prospect theory with traffic travel.Secondly,it analyzes the characteristics of the operation network of different modes of transportation in urban multi-mode,and constructs the network of multi-modal transportation system by using super-network theory,and explains the construction process with the actual starting point and traffic network.A multi-mode network attribute storage method considering the impedance of road intersections and public transportation stations is proposed.The quantification model of car travel time,bus travel time,track travel time and multi-mode travel time in super-network is given.The survey data is used to verify.The normal distribution is a good reflection of the distribution of travel time.Finally,according to the multi-modal transportation system network and the actual travel rules,the feasible super-path is defined,and the reserved travel time calculation method based on the risk attitude level of the traveler is given.Taking the reserved travel time as the supernetwork impedance,the improved Dijkstra algorithm and the improved depth-first algorithm were adopted to obtain the reasonable path set of people with various risk attitudes.Then,based on the path reserved time,the range of departure time is obtained,the dual reference point foreground theoretical model is set,and the corresponding parameters are calibrated to give the data calculation method needed for the foreground theoretical model calculation.The applicability of the dual reference point foreground theoretical model is verified,and the path-departure time joint selection model based on the foreground theory framework is established.The network calculation of the actual traffic system verified the availability of the model.