Impact Of Travel Time Uncertainty And Traffic Network Scale On Travelers' Day-to-day Route Choice Behavior

With the rapid advancement of urbanization,the urban road network has expanded dramatically.At the same time,with considerations of reducing costs and improving efficiency,travelers choose to switch between different routes during the trip.In real life,it is a daily decision for travelers to choose a route under the conditions of different traffic network scales and travel time uncertainty.So,it is of great significance to study the impact of travel time uncertainty and traffic network scale on day-to-day route choice behavior.Aiming at the aforementioned issues,in this thesis,using laboratory experiments and statistical methods to carry out two aspects of work:(1)the impact of travel time uncertainty on travelers' day-to-day route choice behavior.Firstly,through hypothesis testing,it was found that the uncertainty of the route travel time had almost no effect on the number of travelers,but had a significant impact on the route travel time.For travelers,the mastery of travel information is not as high as possible.Incomplete travel information is more conducive to the efficient operation of the entire transportation road network than complete travel information.Secondly,through the analysis of individual path selection behavior,it is found that certain path preference will lead to extreme distribution of travel time of travelers.The travelers who have three types of information preference are considered,a normality test method is used.It is found that as the uncertainty of travel time increases during travel,the degree of information preference of the traveler for the last travel perception time decreases,and the volatility of traveler's information preference gradually weakens.(2)the impact of traffic network scale on travelers' day-to-day route choice behavior.Firstly,the impact of changes in road network scale on road network equilibrium is explored.It is found that as the scale of the road network increases,the number of times the road network reaches user equilibrium is getting fewer and fewer.The relationship between the User Equilibrium(UE)and System Optimum(SO)of each road network and the average of the traffic of each path are compared and analyzed.It is found that the average value of the traffic of each path is very close to their UE value,but it differs greatly from the SO value.Secondly,the travelers' day-to-day route choice switching activities under different traffic network scales are analyzed from two aspects,i.e.day-to-day traffic network flow evolution and individual choice variability.It was found that the number of daily route selection switchers and the number of individual route selection switches were positively linearly related to the scale of the road network.The ordinary least squares regression method is used to fit linear and non-linear models to experimental observation data,it is found that the conversion frequency of individual route selection is significantly related to the average empirical travel time.It can be seen from this that unilaterally expanding the scale of the road network does not always promote the evolution of the road network's traffic flow to the equilibrium state of the user,and there is an interactive relationship between the number of individual path selection transitions and travel time.Finally,The research results obtained above can provide a certain theoretical basis and reference value for the construction and management of the actual traffic road network at the strategic and operational levels.