Abstract
Any model is an idealization as well as elimination of an actual phenomenon. Traffic flow model is no exception. The authors have already advocated ‘Spring-mass system’ model, where car-following behavior is idealized as an analogy of physical dynamics by data oriented approach, instead of eliminating a driver's perception, reaction etc. from discussion. Therefore, each parameter of the model does not correspond one-to-one with driver's behavior. Furthermore, the relation between the microscopic model and a macroscopic phenomenon in traffic flow is nontrivial. The paper tries 1) to observe the traffic flow generated by the model through simulation and 2) to consider that the parameters of the model by a sensitivity analysis. The trial is positioned as a first step to illustrate and mitigate bottleneck phenomena on the interurban expressway.
