Recovery of inter- and intra-personal heterogeneity using mixed logit models

Most applications of discrete choice models in transportation now utilise a random coefficient specification, such as mixed logit, to represent taste heterogeneity. However, little is known about the ability of these models to capture the heterogeneity in finite samples (as opposed to asymptotically). Also, due to the computational intensity of the standard estimation procedures, several alternative, less demanding methods have been proposed, and yet the relative accuracy of these methods has not been investigated. This is especially true in the context of work looking at joint inter-respondent and intra-respondent variation. This paper presents an overview of the various different estimators, gives insights into some of the theoretical properties, and analyses their performance in a large scale study on simulated data. In particular, we specify 31 different forms of heterogeneity, with multiple versions of each dataset, and with results from over 16,000 mixed logit estimation runs. The findings suggest that variation in tastes over consumers is captured by all the methods, including the simpler versions, at least when sample size is sufficiently large. When tastes vary over choice situations for each consumer, as well as over consumers, the ability of the methods to capture and differentiate the two sources of heterogeneity becomes more tenuous. Only the most computationally intensive approach is able to capture adequately the two sources of variation, but at the cost of very high run times. Our results highlight the difficulty of retrieving taste heterogeneity with only cross-sectional data, providing further evidence of the benefits of repeated choice data. Our findings also suggest that the data requirements of random coefficients models may be more substantial than is commonly assumed, further reinforcing concerns about small sample issues.     

Cities on the move – Ten years after

The World Bank urban transport strategy review, “Cities on the Move” analyzed urban transport problems in developing and transitional economies and articulated a proposed strategy framework for national and city governments. This paper describes how the urban transport problems of the developing world have changed in the last decade and assesses the extent to which the strategies recommended in 2002 have been successfully implemented. It shows that progress has been widespread in some areas – particularly in mass transit analysis and investment and some environmental policies – and that there have developed some good planning and public transport practices in a smaller number of model cities. But more strategic institutional and policy issues, including the mobilization and regulation of private sector initiative in meeting infrastructure and public transport supply deficiencies, have tended to be poorly developed. Above all, the growth of medium sized cities with weak institutions and finance highlights the need for the international development institutions to put greater emphasis on helping those cities by dissemination of best practice in strategic transport planning and traffic management.     

The allocation of buses in heavily utilized networks with overlapping routes

Many transit systems outside North America are characterized by networks with extensively overlapping routes and buses frequently operating at, or close to, capacity. This paper addresses the problem of allocating a fleet of buses between routes in this type of system; a problem that must be solved recurrently by transit planners. A formulation of the problem is developed which recognizes passenger route choice behavior, and seeks to minimize a function of passenger wait time and bus crowding subject to constraints on the number of buses available and the provision of enough capacity on each route to carry all passengers who would select it. An algorithm is developed based on the decomposition of the problem into base allocation and surplus allocation components. The base allocation identifies a feasible solution using an (approx.) minimum number of buses. The surplus allocation is illustrated for the simple objective of minimizing the maximum crowding level on any route. The bus allocation procedure developed in this paper has been applied to part of the Cairo bus system in a completely manual procedure, and is proposed to be the central element of a short-range bus service planning process for that city.     

Design charts for estimating transportation planning statistics from area population

Utilizing data collected for urban transportation studies in Ontario, regression analysis has been used to establish relationships between the daily number of person trips in an urban area and the area population. In particular the number of trips by auto drivers, auto passengers and mass transit riders have been investigated. Further, auto driver trips have been stratified into the following destination trip purposes: return home, work and related business, shopping, social-recreational and miscellaneous. The results of this analysis have been used to prepare a set of design charts. These charts are presented graphically and in the form of a nomogram. The accuracy of these charts has been investigated and found satisfactory for most planning purposes.     

Evaluation of transit operations: data applications of Tri-Met's automated Bus Dispatching System

In this paper, archived Automatic Vehicle Location and Automatic Passenger Counter data are used to evaluate actual bus running time variation in relation to scheduled service for Tri-Met, the transit provider for the Portland, Oregon metropolitan area. Given observed variation in running times, scheduled recovery times are found to be generally (though not universally) excessive. This results in an under-investment of resources in revenue service relative to non-revenue service. Analysis of trip level data reveals that bus operators are an important source of running time variation after controlling for such factors as route design, time of day and direction of service, and passenger activity.     

Testing electric vehicle demand in ‘hybrid households’ using a reflexive survey

The debate over electric vehicles (EVs) pivots largely on issues of market demand: will consumers purchase a vehicle that provides substantially less driving range, yet can be refueled at home, than an otherwise comparable gasoline vehicle? Also, what role do other unique attributes of EVs play in the purchase decision? Most previous studies find that limited driving range is a serious market barrier; many of those same studies ignore or under-value other novel attributes. To probe these future consumer decision processes deeply and robustly, we first devised and conducted detailed, interactive and experiment-oriented interviews. Then, incorporating what we learned, we designed an innovative mail survey and administered it to 454 multi-car households in California. The four-stage mail survey included a video of EV use and recharging and other informational material, completion of a 3-day trip diary and map of activity locations, and vehicle choice experiments. In addition to propulsion systems, respondents made choices of body styles, driving ranges, and other features. We formalized and tested what we call the hybrid household hypothesis: households who choose EVs will be purposefully diversifying their vehicle holdings to achieve the unique advantages of different propulsion systems. The hypothesis is supported, given the assumptions in our experimental design. In fact, a significantly larger number of EVs are chosen than the minimum number that would support our hypothesis. We find that purchases of battery-powered EVs by hybrid households would account for between 7 and 18% of annual light duty vehicle sales in California. EVs sold to fleets and other households would be in addition to those identified by this study.     

Gendered immobility: influence of social roles and local context on mobility decisions in Pakistan

This paper examines the instances of one-day immobility in Pakistan and reports its socio-demographic determinants using the nationally representative dataset of the 2007 Pakistan Time Use Survey. Of 37,830 time diary respondents, nearly 30% did not report travel during the diary day. Homemakers and those out of the workforce were more likely to be immobile than employed or student respondents. Immobility rates were very high among women (55%) as compared to men (4%). Among women, those between 20 and 34 years of age, married, with children, having better education, dependent on other household members and those living in higher income households were more likely to be immobile. The excessive gender nature of immobility seems to be triggered by a gender-based sociocultural environment, which restricts female mobility due to family honor concerns. Other than this, those living in the provinces of Sindh and Khyber Pakhtunkhwa or in urban areas were more likely to be immobile than those living in Punjab and Sindh provinces or in rural areas. The significant geographic effect at broader spatial scale is caused by the demographic structure as well as due to differences in the social and cultural context of these areas. Finally, questions regarding the measurement of immobility and the potential implications of increased female immobility are discussed.     

An intersection turning movement estimation procedure based on path flow estimator

Estimation of intersection turning movements is one of the key inputs required for a variety of transportation analysis, including intersection geometric design, signal timing design, traffic impact assessment, and transportation planning. Conventional approaches that use manual techniques for estimation of turning movements are insensitive to congestion. The drawbacks of the manual techniques can be amended by integrating a network traffic model with a computation procedure capable of estimating turning movements from a set of link traffic counts and intersection turning movement counts. This study proposes using the path flow estimator, originally used to estimate path flows (hence origin–destination flows), to derive not only complete link flows, but also turning movements for the whole road network given some counts at selected roads and intersections. Two case studies using actual traffic counts are used to demonstrate the proposed intersection turning movement estimation procedure. Copyright © 2010 John Wiley & Sons, Ltd.     

Road user charging design: dealing with multi-objectives and constraints

This paper proposes an innovative approach for designing a road user charging scheme to meet multiple policy objectives. Three practical features are integrated into the design methodology including (i) cordon formation, (ii) a set of design constraints, and (iii) multiple objectives of the scheme. The methods also consider possible responses of road travellers to the charging scheme. Two methods based on genetic algorithms (GA) are developed for optimising a charging cordon scheme with constraints and with multiple objectives. The dynamic self-adaptive penalty GA and Non-dominated Sorting GA II (NSGA-II) are applied to the constrained design and multi-objective design respectively. The objective functions or constraints considered include social welfare improvement, revenue generation, and distributional equity impact. A case study of the City of Edinburgh is presented and common characteristics of charging cordon designs which perform well against the three objectives are discussed.

Understanding the modifiable areal unit problem and identifying appropriate spatial unit in jobs–housing balance and employment self-containment using big data

Transportation - Jobs–housing balance (JHB) and employment self-containment (ESC) have been used to examine the jobs–housing relationship. However, the effect of the modifiable areal...