Physical characteristics of paratransit in developing countries of Asia

Paratransit modes play a significant role in the urban transport sectors of developing countries since in many cities more than half of the total public transport demand are carried by them. Rapid increase in urban population, per capita income, along with inadequate existing transport infrastructures have stimulated their usage as cheap and convenient public transport modes. This paper examines their present role and physical characteristics in a comparative form to provide a basic data for discussion of urban transport issues in developing cities.     

Modeling oil spills in a river—lake system

A general model shell, ROSS3, is developed for simulating oil spills in complex river systems using techniques which have not been previously exploited in oil spill models. ROSS3's new approach has several advantages over the approach to model oil spills in the past: (a) The use of a time-varying boundary-fitted coordinate system that allows accurate accounting for complex river/lake boundary as well as the river boundary changes as its water levels fluctuate; (b) The ability to confine two-dimensional hydrodynamic computations to a limited river reach; (c) The ability to interactively layout the channel networks for setting up the model, define extra cross sections to increase the accuracy if needed, in addition to the traditional data entry and visualization interfaces.ROSS3 is a two-layer two-dimensional oip spill model that can simulate the mechanism of advection, horizontal diffusion, mechanical spreading, shoreline deposition, evaporation, dissolution, vertical mixing, resurfacing and sinking. In ROSS3 spilled oil may be a surface slick or suspended oil droplets, or a combination of both. Both free surface and ice cover conditions can be simulated. The flow of conditions can be varied and the unsteady flow model can be run within ROSS3 to simulate the flow conditions in both the river and the lake. The ice conditions can be added or removed from the model input using easy interactive procedures.     

Residual stresses at a longitudinal stiffener — Web frame intersection and their effects on crack growth

Stress coining at geometrical discontinuities may introduce compressive residual stresses which will lower the propagation of surface cracks and thus increase the fatigue life. In a non-linear finite element analysis, the residual stresses after welding of a transverse web-frame to a ship hull plating is first estimated. Stress coining of one corner of a longitudinal cut-out in a transverse web-frame is then simulated in detail. The calculated residual compressive tangential stresses in the corner of the cut-out is seen to lower the propagation of edge cracks. Experiments performed on web-frames taken from a shipyard after manufacturing show indeed that large tensile residual stresses exist in the web-frame after welding. Stress coining is therefore believed to have a beneficial effect on the fatigue life.     

Wheel/Rail Non-linear Interactions With Coupling Between Vertical and Lateral Directions

Summary A theoretical model is developed to explore the high frequency wheel/rail interaction with coupling between the vertical and lateral directions. This coupling is introduced through the track dynamics due to the offset of the wheel/rail contact point from the rail centre line. Equivalent models of the railway track in the time domain are developed according to the rail vibration receptances in the frequency domain. The wheel is represented by a mass in each direction with no vertical-lateral coupling. The vertical wheel/rail interaction is generated through a non-linear Hertzian contact stiffness, allowing for the possibility of loss of contact between the wheel and rail. The lateral interaction is represented by a contact spring and a creep force damper in series and their values depend on the vertical contact force. The vibration source is the roughness on the wheel and rail contact surfaces which forms a relative displacement excitation in the vertical direction. Using the combined interaction model with this relative displacement excitation, the wheel/rail interactions with coupling between the vertical and lateral vibrations are simulated. It is found that the lateral interaction force caused by the offset is usually less than thirty percent of the vertical dynamic force. The lateral vibration of the rail is significantly reduced due to the presence of the lateral coupling, whereas the vertical interaction is almost unaffected by the lateral force.     

Vertical Vibration Analysis of Vehicle/Imperfect Track Systems

Summary This paper studies the vertical vibration of a vehicle traveling on an imperfect track system. The car body and sleepers are modeled as Timoshenko beams with finite length, and the rail is assumed as an infinite Timoshenko beam with discrete supports. Imperfection of the track system comes from a sleeper lost partial support by the ballast. Since deflection of the rail is limited within a certain interval where the vehicle is passing over, the infinite domain problem can be transformed into a finite domain problem with moving boundary. In this work, the equations of motion of the car body, rail and sleepers are discretized first by the finite element method. The discretized equations of motion for the vehicle and track systems are then assembled, respectively. Finally, the Newmark method is applied to obtain the response of the vehicle and track systems at each time step. The effect of the vehicle speed on the response of the vehicle and track systems is investigated.     

Currency adjustment factors: some alternative strategies

The purpose of this paper is to consider some of the issues arising from currency fluctuations or currency disparities and in particular their effects upon freight earnings through the application of a currency adjustment factor (CAF).

The operation of CAFs has been a source of great concern to shippers and shipper's councils since their introduction some seven years ago. The basis of this concern has been the widening differential of applicable CAFs as a surcharge on basic freight rates. The CAF system has been regarded by many as iniquitous and punitive. Conferences have been accused of insulating themselves from currency losses due to exchange-rate variability consequent upon the introduction of flexible exchange rates in 1971. However, this mounting pressure has brought the operation of the CAF formula under close scrutiny especially from four sources.

(a) Some member Councils of the European Shippers' Council (ESC).

(b) The Japanese Shippers' Council (JSC).

(c) The question of the legality of differential CAFs under the Treaty of Rome

(d) Complaints from North Continental Maltsters.     

Currency adjustment factors

The manner in which CAFs have been applied to cargo carried by lines appears to be both arbitrary and secretive. This note intends to give some clarification as to how the CAF is calculated and the manner in which it is applied.

The operation of the two methods, i.e. the monthly review and the radical review, is merely a device which enables conferences to adjust effective charges so as to leave lines without a loss in real revenue, although minor losses and gains will be inevitable as the operation of the CAFs take time to be implemented.

Finally it is hoped to point out the volatilc nature of CAF changes impinges upon the uncertainty of lines' costs.     

A statistical approach to the traveling salesman problem

A statistical approach is shown to be adaptable to the N-city traveling salesman problem by considering route distances to be random variables which are continuous and normally distributed. A solution to the shortest route distance and path can be approximated by utilizing a Monte Carlo simulation to obtain a representative sample of possible journeys. The approach involves recursive statistical inference which is used to select next-city visits leading to the most probable minimum route path. A statistical selection of the minimum route path is computationally efficient and computer run time increases in proportion to the square of the number of cities as opposed to an (N - 1)! increase for a deterministic approach. The accuracy of the statistical approach is directly proportional to the number of Monte Carlo simulations.