Search

The public transportation in local areas is mostly supplied by private enterprise buses. But the volume of the bus users has been decreasing for a long time. This trend is especially notable in and around the local small cities where the population has been stagnant and the private cars have found their way into almost all homes.In this paper, the authors have developed a way to predict the bus user numbers in future in an area which is in a southerm part of Fukuoka Pre., and have proposed some ideas so as to maintain the public transportation system.

This paper describes two methods to determine the groups of road segments to be constructed, the priority between them and the construction order of road segments within each group considering the disutility under construction. The dynamic programming and stochastic schemata exploiter method are utilized for optimization procedure. The mathematical modeling and solution technique are emphasized. An example problem is included and illustrated for showing the applicability of the proposed methods. The results indicate that the methods are useful for the multi-stage determination problems such as the road network planning.

So far we have developped several methods to dispose the continual traffic counting points. And it also needs to consider how to distribute the points of general traffic survey on the road network. In this reserch, we analysed the degree that the trip length distribution and the intervals between traffic counting points affect the accuracy of the estimation of the observed frequency and total triplength.This paper shows that the expectation of the observed frequency is in proportion to the trip length, that the estimation error of total trip length is in proportion to the average interval, and that the average interval needs to be less than 5.0km from the analysis of the trip length distribution of the car OD surveyin 1980. On the base of these results, we tried to dispose the traffic counting points nearly equally on the route of the trunk road network of Fukuoka City.

This paper proposes a new method for the optimal allocation of increasing activity demmands and the determination of road sections to be improved in urban area, using goal programming and multiple objective linear programming technique.As the results of case study in SAGA city, it has been found that this method is available for projecting future urban environments and for testing the probable consequences of alternative policies such as improvement of road network and land use controls.

A bunch size distribution models of traffic flow are proposed based on the distinction between leaders and followers in speed distribution. In analysis, a new probabilistic method is applied to decide bunching vehicles instead of usual method of using critical headway.Then the bunch-speed distribution models are induced by the use of bunch size distribution model.Consequently, these models can be obtained only by giving traffic volume. Models are also applied to observed data and those appropriateness are examined.