Variances and covariances for origin-destination flows when estimated by log-linear models

An expression is derived for the variances and covariances of the logarithms of origin- destination flows when estimated by one of a family of log-linear models. Among this family is the Furness growth factor model, the gravity model, the Willumsen model and the Van Zuylen-Bell model. Variances and covariances for the logarithms of the parameters incorporating prior information and for the values of the constraints with which the fitted values must conform are transformed linearly into approximate variances and covariances for the logarithms of the fitted values. Additional error due to misspecification of either the model or the constraints is not taken into account. Depending on the model, the prior information parameters may correspond to values of a deterrence function or to a base year trip matrix, while the constraints may represent design year trip end totals or traffic counts. As an illustration, the expression is applied to an example involving Willumsen's model.     

Modeling drivers' diversion from normal routes under ATIS using generalized estimating equations and binomial probit link function

Advanced Traveler Information Systems (ATIS) provide travelers with real time traffic information to optimize their travel choices. The objective of this paper is to model drivers' diversion from their normal routes in the provision of ATIS. Five different scenarios of traffic information are used. Generalized Estimating Equations (GEE) framework with repeated observations and binomial probit link function is introduced and implemented. GEE with four different correlation structures including the independent case are developed and compared with each other and with regular Maximum Likelihood Estimation (MLE). A travel simulator was used. Sixty-five subjects have traveled 10 simulated trial days each on a 40-link realistic network with real historical congestion levels. The results showed that providing traffic information increases the probability of drivers' diversion from their normal routes. Adding advice to the pre-trip and/or en-route information encourages drivers to divert. Providing en-route in addition to the pre-trip information with or without advice increases the diversion probability. High travel time on the normal route and less travel time on the diverted route increase the probability of diversion. High-educated drivers are less likely to divert. Expressway users are more likely to divert from their normal routes under ATIS. Drivers' familiarity with the device that provides the information and high number of traffic signals on the normal route increase the diversion probability.     

The Horsepower Reserve Formulation of Driveability for a Vehicle Fitted With a Continuously Variable Transmission

Summary A systematic methodology is developed for choosing the optimum ratio trajectory of a continuously variable transmission in a passenger vehicle. The optimum CVT ratio schedule is formulated as a constrained optimization problem with maximum fuel economy as the objective function and driveability concerns and physical limitations included as the constraints. The key notion to achieving good driveability is the introduction and definition of a horsepower reserve function that creates a consistent and desirable vehicle response under different driving conditions. Simulation results compare the optimized schedule's performance with several other possible ratio schedules, including the minimum brake specific fuel consumption map. Results from the optimized schedule indicate only a mild tradeoff between driveability and fuel economy relative to the other ratio schedules. The ratio optimization problem formulation and solution provide a novel and unique approach for systematically addressing driveability and fuel economy considerations associated with a continuously variable transmission.     

究竟哪个中国公司会最先进驻美国市场?

上月在拉斯维加斯召开美国汽车经销商联席会议时,我在J．D．Power国际圆桌会议被美国经销商狂轰乱炸,一次又一次被问及这样两个问题:谁来了?什么时候进入美国市场?     

Modeling Travel Time Under ATIS Using Mixed Linear Models

The objective of this paper is to model travel time when drivers are equipped with pre-trip and/or en-route real-time traffic information/advice. A travel simulator with a realistic network and real historical congestion levels was used as a data collection tool. The network included 40 links and 25 nodes. This paper presents models of the origin-to-destination travel time and en-route short-term route (link) travel time under five different types and levels of advanced traveler information systems (ATIS). Mixed linear models with the repeated observation's technique were used in both models. Different covariance structures (including the independent case) were developed and compared. The effect of correlation was found significant in both models. The trip travel time analysis showed that as the level of information increases (adding en-route to the pre-trip and advice to the advice-free information), the average travel time decreases. The model estimates show that providing pre-trip and en-route traffic information with advice could result in significant savings in the overall travel time. The en-route short-term (link) travel time analysis showed that the en-route short-term (link) information has a good chance of being used and followed. The short-term qualitative information is more likely to be used than quantitative information. Learning and being familiar with the system that provides the information decreases en-route short-term delay.     

Pragmatic Tyre Model for Short Wavelength Side Slip Variations

The first-order relaxation length concept is often used to model transient tyre behaviour. It gives a rather good representation of the lateral force response, also at shorter wavelengths. The self aligning moment appears to behave rather differently and its responses cannot be represented by a relaxation length system only. Based on the analytical frequency response functions of the pneumatic trail of the brush type tyre model, a new pragmatic approach is developed. The model consists of a phase leading system in series with the first-order model for the lateral force. The results are compared to the responses of a discrete brush type simulation model. It appears that the aligning moment can be represented by this new pragmatic model very well.     

Braking Force Distribution Control for Improved Vehicle Dynamics and Brake Performance

This paper describes the feasibility of improving the braking performance of a commercial vehicle by using an electronic braking system. An electronic braking system enables the braking force at each wheel to be independently controlled. Braking force distribution control makes the braking force at each wheel proportional to each wheel's load. Results of computer simulation and vehicle test showed that the proposed control laws can eliminate the effects of a laden condition on the braking distance and can increase the degree of deceleration at which wheel lock occurs, resulting in improved vehicle attitude stability during a critical maneuver.     

On the equivalence of common rearrangements for sum constrained travel estimation problems

This paper examines the problem of estimating the parameters of nonlinear sum constrained models by use of rearrangements to permit linear estimation methods. Over the years a number of different, and sometimes ingenious, techniques have been proposed but a remaining difficulty arises from the fact that different rearrangements lead to different parameter estimates. In this paper it is shown that while ordinary least squares certainly results in different estimates, generalised least squares leads to identical parameter estimates provided convergence to the true maximum likelihood estimates is achieved. In particular it is seen that estimates are invariant with choice of base or deletion of one equation and identical for the single base, multinominal logit, geometric mean and link constant versions. The findings of this paper apply to any sum-constrained model and in transportation these occur very frequently as mode choice problems, singly-constrained gravity models, and as route choice models in path assignment. While we develop all results in this paper in the context of an intercity passenger modal market share model, the transferability of these results to other problems should be recognised.