Frontier-based methods, particularly Data Envelopment Analysis (DEA) and Stochastic Frontier Analysis (SFA), have become the dominant tools for benchmarking efficiency in transportation systems. However, these approaches typically rely on a common-frontier assumption that treats all decision-making units as technologically homogeneous. This assumption is especially restrictive in transportation, where systems differ substantially across modes, countries, institutional contexts, and stages of infrastructure development. This paper systematically reviews the frontier-based efficiency literature in transportation through the lens of heterogeneity. It examines how different forms of heterogeneity are addressed or overlooked in existing studies and assesses the potential of Meta-Frontier Analysis (MFA) as a framework for distinguishing managerial inefficiency from technology gaps. This study conceptualised heterogeneity into three dimensions: stochastic (random shocks), structural (scale, regulation, institutional settings), and technological (production technology differences). The review reveals three key findings. First, while extended DEA and SFA models partially address stochastic and structural heterogeneity through techniques such as noise decomposition, variable returns to scale, bootstrapping, and second-stage regressions, technological heterogeneity remains largely unmodelled or unobserved. Second, MFA offers a framework for decomposing inefficiency into managerial and technology-gap components, yet its application in transportation remains limited and fragmented, with MF-DEA emerging slowly and MF-SFA virtually absent, revealing a persistent "stochastic gap". Third, the failure to account for technological heterogeneity leads to systematic policy misdiagnosis, in which technological constraints are misattributed to managerial failure. Beyond synthesising existing evidence, this paper proposes a conceptual methodological direction, an entropy-based clustering framework for endogenous technology grouping for MFA. This data-driven approach shifts heterogeneity treatment from assumption to inference, enhancing reproducibility and policy credibility. The review concludes by outlining a research agenda that prioritises: (i) development of stochastic meta-frontier models, (ii) integration of sustainability and undesirable outputs, (iii) dynamic MFA frameworks, and (iv) a policy diagnostic toolkit linking efficiency decomposition to intervention design. Through this review, scholars can build an understanding of the impact of heterogeneity in efficiency analysis and policy making.
In developing countries, limited data availability and research funding often hinder the assessment of the potential impact of electric vehicle (EV) penetration on greenhouse gas (GHG) emissions. Despite these challenges, providing policymakers with data-driven insights is essential for fostering governmental support for EV adoption. This paper presents an integrated framework designed to estimate GHG emission reductions resulting from a transition from internal combustion engine vehicles (ICEVs) to EVs in developing countries. The framework utilizes a Well-to-Wheel analysis to quantify the GHG reduction potential of EV adoption. Applying this framework to Tehran, Iran, our findings indicate that a 1
Driver speed behavior constitutes a critical aspect in the assessment of the performance of road infrastructure vis-a-vis safety in mountainous terrain. The present study has been conducted to study driver speed behavior on roads in mountainous terrain by analyzing the variability of minimum speed positions on curved and maximum speed positions on tangent road sections. Significant variability is observed to be associated with the minimum speed position on curves and maximum speed position on tangents on four-lane roads in mountainous terrain. Further, the associated variabilities in maximum and minimum speed positions are also characterized with the best-fitted distributions. Higher variability of minimum and maximum speed positions on curves and tangents signifies the necessity of considering the complete length of curves and tangents in the formulation of the operating speed prediction models and the findings also have crucial implications for the safety evaluation of four-lane highways in mountainous terrain.
Activity-based travel demand models have reached a state of maturity but, several challenges remain in their survey design, format, administration technique, sample size determination and constraints faced during field survey. This research is an attempt to explore these issues and suggest way forward towards developing an efficient survey instrument and survey strategy for deployment in emerging countries like India and particularly during the post-pandemic period. A two-step approach of pre-pilot and pilot survey has been adopted to test our survey format and strategy before the conduct of final survey while considering the low literacy rate, qualitative inputs by enumerators, quantitative performance measures; average survey duration, percentage of error and response rate, acceptance rate and cost of survey and impact of the pandemic. A new compact, open-interval based, hybrid activity-travel diary format is proposed along with a hybrid method for data collection and validation and various recommendations for field survey strategies.
Transportation investments are primarily driven by the objective to enhance the accessibility and mobility of people and goods, thereby contributing to the social and economic development of communities. This study examines transportation equity in Bangalore in the context of two types of disturbances on equity: recurring urban floods and fare hikes. While urban floods have a short-term but profound impact, fare hikes result in prolonged economic impacts, both significantly affecting the Quality of Life (QoL) and well-being of citizens. To assess equity, an exploratory survey focusing on parameters of accessibility and affordability was employed. Statistical methods such as cluster analysis and McNemar’s test were carried out to extract patterns and insights from the collected data. The impact of fare hikes assessed using McNemar’s test revealed a 31.5
