A note on the use of correlation coefficients for assessing goodness-of-fit in spatial interaction models

The use of correlation coefficients to evaluate the accuracy of spatial interaction models is inappropriate unless such models have been fitted using least squares techniques. In other cases the correlation involves an implicit intercept value and a regression coefficient that may significantly modify the interaction model's estimates. Researchers have not acknowledged the role of these two parameters when the correlation is used. A generalized root mean square error is proposed as an alternative indicator of accuracy that may be used with any model.     

A practical assessment of stated preferences methods

Stated preferences data in the form of rankings, ratings and choices were collected in Santiago and discrete choice models estimated with them. The models were compared in terms of accuracy v/s the cost of obtaining the information and models. All methods produced reasonable but different models and fairly close subjective values of time. In terms of production costs the ranking method was a clear looser although the experimental design was slightly biased against it. Finally, the use of computerised interviews is highly recommended particularly for dealing with low income people.     

Dynamics of the railway track and the underlying soil: the boundary-element solution, theoretical results and their experimental verification

A combined finite-element boundary-element method is presented in detail to calculate the dynamic interaction of the railway track and the underlying soil. A number of results are shown for ballasted and slab track, demonstrating the influence of the stiffness of the soil and the rail pads on the vertical compliance of the track. The compliance of the track is combined with a simple model of the vehicle giving the transfer function of vehicle-track interaction. An experimental verification of the theoretical results is achieved by harmonic and impulse excitation with and without static (train-) load and by combined measurements of train-track-soil interaction. A clear vehicle-track resonance is found for the slab track with elastic rail pads and for higher frequencies at highspeed traffic, the dynamic axle loads due to sleeper passage are reduced.     

A study on the stability of a motorcycle wheel–swingarm suspension with chain transmission

The present study describes a possible driving mechanism for a self-excited oscillation observed in motorcycle dynamics, often referred to as chatter. This phenomenon, affecting the performance of road racing motorcycles, has been simulated in straight running braking manoeuvres with multibody motorcycle models. It involves rear suspension bounce and driveline oscillation in the frequency range between 17 and 22?Hz. A simplified model of a motorcycle rear suspension with chain transmission is proposed and its stability in equilibrium configurations is studied via eigenvalue analysis. The sensitivity with respect to all its governing parameters is analysed by means of stability maps and the self-excitation mechanism is explained with the aid of energy balance analysis and phase diagrams. It is found that the key role for the instability onset is played by the gradient of the nonlinear characteristic slip function of the tyre.     

Application of Sensitivity Methods to Analysis and Synthesis of Vehicle Dynamic Systems

The development and application of sensitivity methods for determining the effects of parameter changes on the response of vehicle dynamic systems is presented. The procedures shown can be used to enhance the analysis and synthesis processes of virtually any road or rail vehicle system regardless of its complexity. The parametric sensitivity of vehicle models in time domain, steady state models and vehicle models in frequency domain can be investigated using different types of sensitivity functions, both dimensional and dimensionless including first order standard, percentage, logarithmic, second order standard, and logarithmic and percentage sensitivity measures. These sensitivity functions and measures are determined as functions of partial derivatives of system variables taken with respect to system parameters. In the case of sensitivity functions in the frequency domain the variable values are computed as either the magnitude or phase angle of a complex element of the transfer function matrix. The methods presented enable to determine the influence of all system primary (constant) and secondary (non-constant) parameters on system primary and secondary variables. The primary variables are state variables or elements of the transfer function matrix and the secondary variables may be any functions of primary variables and system parameters. Typical secondary system parameters which can be examined include initial conditions, time variant coefficients, natural frequencies, loads, and typical secondary variables are forces, weight transfers, stability factors and energy components. The analysis of sensitivity results obtained for three vehicle handling models in both linear and nonlinear regimes of vehicle performance and utilizing various types of sensitivity functions is also presented.     

Aggregating mobile object trajectories: cumulative time geographic density estimation for GPS data

We present new approaches that expand upon the time geographic density estimation (TGDE) framework previously employed to estimate potential path trees. In the past, TGDE metrics have identified possible locations an individual moving object may have passed between, given known origin and destination points. This paper utilizes a new form of TGDE to investigate taxicab GPS traces over a specified time horizon with position ‘gaps’. To this end, we propose a new extension to the TGDE framework, TGDE-C, which is used to determine the cumulative TGDE values for a group of GPS traces, at a given location. These metrics are applied to multiple taxis and allow for time of day analysis. Additionally, we combine these new extensions with existing TGDE metrics that allow us to determine how accessible individual or groups of vehicles are to urban opportunities.     

Visualisation of uncertainty in probabilistic traffic models for policy and operations

This study investigates different methods to visualise uncertainty in static representations of probabilistic traffic models predictions on road-networks. Although various graphical cues may be used to represent uncertainty it is not a priori clear which of them are most suited for this purpose, since their legibility, intelligibility and the degree to which they interfere with other graphical elements in a representation differ widely. Several graphical uncertainty representations were therefore developed and analysed in expert sessions. A selection of the initial set of uncertainty visualisations was further evaluated in a cognitive alternative task-switching experiment. The results show that graphical representations are able to convey uncertainty information relatively accurately, while some uncertainty visualisations outperform others. It depends on the model and scenario which representation is most suited for a given application. This paper presents an overview of possible graphic uncertainty representations and the considerations involved when applying them to uncertainty in traffic model visualisations.