Previous Article in Journal
Filtration of Mineral and Biological Aerosols by Natural Plant Panels
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Atmosphere
Volume 16
Issue 6
10.3390/atmos16060695
atmosphere-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Integrating Traffic Dynamics and Emissions Modeling: From Classical Approaches to Data-Driven Futures

by
Xin Wang
1,*,
Xianfei Yue
1,
Jianchang Huang
2 and
Shubin Li
1
1
Shandong Engineering Research Center of Intelligent Traffic Control and Guidance Technology for Public Security, Shandong Police College, Jinan 250014, China
2
School of Transportation and Logistics Engineering, Shandong Jiaotong University, Jinan 250357, China
*
Author to whom correspondence should be addressed.
Atmosphere 2025, 16(6), 695; https://doi.org/10.3390/atmos16060695 (registering DOI)
Submission received: 3 April 2025 / Revised: 6 May 2025 / Accepted: 4 June 2025 / Published: 9 June 2025
(This article belongs to the Section Atmospheric Techniques, Instruments, and Modeling)
Versions Notes

Abstract

:
A persistent disconnect between traffic modeling and environmental emissions modeling, stemming from their independent disciplinary evolution, continues to impede the accurate integration of traffic dynamics into emissions prediction. This misalignment frequently results in inconsistencies in simulation outputs and limits the reliability of traffic-based environmental assessments. From a traffic engineering perspective, it is essential that emissions models more precisely reflect real-world vehicle behavior and the complexities of dynamic traffic conditions. In addressing this gap, the present study offers a comprehensive and critical review of the integration between traffic dynamics and emissions modeling across macro-, meso-, and micro-scales. Emissions models are systematically classified into four categories—driving cycle-based, speed–acceleration matrix-based, engine power-based, and vehicle-specific power-based—and assessed in terms of their responsiveness to dynamic traffic inputs. Furthermore, the review highlights the emerging challenges associated with connected and autonomous vehicles and AI-driven modeling techniques, underscoring the urgent need for modular, real-time adaptable modeling frameworks. Through a detailed examination of parameter requirements, data integration issues, and validation challenges, this study provides structured insights to guide the development of scientifically robust and operationally relevant emissions models tailored to the demands of increasingly complex and intelligent transportation systems.

1. Introduction

Road traffic emissions are a significant source of air pollution and greenhouse gas emissions, particularly in countries experiencing rapid urbanization and motorization, such as China [1]. Traffic emissions modeling serves as an essential instrument for accurately quantifying pollutant outputs, thereby forming the basis for the formulation of targeted emissions control strategies and informed policy interventions [2]. By enabling the prediction of pollution levels, emissions models facilitate the evaluation of various traffic management measures and offer critical support for the design of long-term, sustainable urban development plans [3].
Although traffic (or motor vehicle) emissions models have been studied for several decades, early research was primarily focused on environmental science and automotive engineering [4]. This disciplinary divide presents significant challenges for the integration of traffic dynamics with emissions prediction models [5]. Parameter mismatches between emissions and traffic models have been a persistent issue, often leading to inconsistent data calibration and a lack of sensitivity to traffic control measures and policy interventions [6]. Consequently, many models fail to accurately capture real-time variations in vehicle behavior that significantly affect emissions, such as congestion patterns, stop-and-go traffic, and signal coordination [7,8].
Given the growing importance of real-time and detailed emissions data for urban planning, it is essential to analyze the generation, development, and evolution of traffic emissions models from a traffic engineering perspective [9]. Focusing on traffic engineering principles can bridge the gap between traffic and emissions models, improve the accuracy of emissions quantification, and ensure that models are responsive to traffic policies. This approach also supports the development of advanced emissions models that account for the complexities of modern traffic systems, such as connected and autonomous vehicles (CAVs) and big-data technologies. Understanding trends and advancements in emissions modeling is crucial for selecting or developing suitable models to evaluate the environmental impact of traffic policies and control measures, especially in regions with high traffic density and air quality concerns, such as China.

2. Database Search and Search Criteria

To conduct a comprehensive and systematic review of the integration between traffic dynamics and emissions modeling, we performed a targeted literature search across three major academic databases: Google Scholar, Scopus, and Web of Science. The search strategy employed a set of focused keywords, including “traffic emissions modeling,” “vehicle-specific power,” “speed–acceleration matrix,” “engine power modeling,” and “traffic–emission integration.” Priority was given to peer-reviewed journal articles, authoritative technical reports, and prominent conference proceedings published between 2010 and 2025.
Studies were included if they involved the development or application of traffic emissions models under dynamic traffic conditions, with particular attention to those incorporating macro-, meso-, or micro-scale traffic parameters such as vehicle miles traveled (VMT), traffic speed profiles, and second-by-second vehicle movement data. Exclusion criteria encompassed purely theoretical research lacking real-world validation, studies not centered on transportation-related emissions, and non-English publications.
In total, over 90 relevant works were shortlisted for detailed analysis. Each selected study was systematically categorized according to its underlying modeling approach: driving cycle-based, speed–acceleration matrix-based, engine power-based, or vehicle-specific power (VSP)-based. From each, we extracted comprehensive information regarding model structure, required input parameters, calibration procedures, validation methodologies, and application scenarios. A thematic analysis was subsequently undertaken to synthesize technological innovations, identify persistent integration challenges, and highlight emerging trends. This approach was designed to offer a clear multiscale perspective on the evolving interplay between traffic dynamics modeling and emissions prediction.

3. The Origins of Traffic Emissions Modeling

More than a decade after the 1943 photochemical smog event in Los Angeles, motor vehicle emissions were identified as the primary cause of local air pollution [10]. This event initiated a series of motor vehicle emissions-related legislation and modeling studies. The California EMFAC model is regarded as the first motor vehicle emissions factor model, and subsequent legislation and standards from the U.S. Congress and the Environmental Protection Agency (EPA) promoted the development of emissions models, including EMFAC and MOBILE [11,12]. The U.S. Clean Air Act (CAA) was amended in 1970 to specifically target the regulation of emissions from traffic sources [13]. In addition to the CAA, the U.S. has enacted laws such as the Intermodal Surface Transportation Efficiency Act and the Transportation Equity Act for the 21st Century to regulate traffic pollution [14]. Of these laws and regulations, the CAA is the cornerstone of pollution control in the U.S. Under its authority, the EPA has established mandatory National Ambient Air Quality Standards that require state governments to create State Implementation Plans (SIPs) and submit them to the EPA for approval [15]. The EPA has also established Transportation Conformity Regulations to assess whether new traffic projects conform to SIP requirements, withholding federal funding from those that do not [16]. This measure ensures that traffic planning and construction projects do not degrade air quality or cause new air quality problems.
Creating an emissions inventory is essential for developing an SIP, and emissions modeling is a critical tool in this process. Since 1978, the EPA has published the MOBILE emissions model to calculate emissions inventories for on-road traffic sources [17]. MOBILE uses a technical approach similar to the EMFAC model, calculating emissions by multiplying the emissions factor per mile by the total vehicle miles traveled [12]. The emissions factor is derived from emissions testing and considers vehicle age and miles traveled (deterioration), while the vehicle miles traveled are derived from traffic surveys or models provided by traffic management authorities [18]. This model established a methodological framework for measuring traffic emissions inventories, which has been adopted by later models, including the European COPERT, HBEF, and DMRB, and U.S. MOVES models for macro-emissions inventories [19].

4. Development of Emissions Models

While the MOBILE emissions model aids in creating emissions inventories and evaluating policies at the planning level, the growth of intelligent traffic technology has increased the need for emissions evaluations in managing and controlling traffic. However, such emissions models struggle to capture the dynamic characteristics of traffic and its impact on emissions at more detailed spatial and temporal resolutions [20]. This has created a need for research on microscopic emissions models that can capture dynamic traffic characteristics, as traditional models used for emissions inventories are now classified as macroscopic models [21]. Emissions models are categorized into four types based on how they characterize vehicle driving states: driving cycle, speed–acceleration, engine power, and specific power distribution, as outlined in Table 1. Table 2 offers a comparative summary of major traffic emissions models, outlining their input requirements, complexity, resolution, application scales, strengths, and limitations.

4.1. Emissions Models Based on Driving Cycle

A driving cycle is a time-varying speed curve that describes the changes in vehicle operating patterns during a typical trip, including various phases such as cold start to hot stabilization, low-grade to high-grade roads, and urban to suburban transitions. It is used to evaluate fuel economy and emissions levels of motor vehicles [22]. Examples include the NEDC (New European Driving Cycle) [23] and WLTC (World-Harmonized Light-Duty Vehicle Test Cycle) [24] in Europe; FTP-75 (Federal Test Procedure) [25] in the U.S.; and JC08 and 10–15 Mode [26] in Japan. These driving cycles are trip-based, reflecting daily travel characteristics and the average level of vehicle activity.
Emissions models based on the driving cycle include MOBILE [27], COPERT [28], EMFAC [29], HBEFA [30], and CVEM [31], all of which share similar emissions calculation methods. Vehicle emissions are calculated by coupling modified emissions factors with driving mileage. The emissions factor is affected not only by the vehicle’s physical properties (e.g., weight, fuel type, and emissions control measures) but also by its activity properties (e.g., speed, acceleration, deceleration, and road type) and environmental factors (e.g., temperature and humidity) [32]. To determine emissions factors, the driving cycle for each vehicle type, across various road types and traffic conditions representing the vehicle’s average driving state, must be established. Then, emissions during the driving cycle are measured through experiments and defined as the basic emissions factor. This factor is further revised by accounting for factors such as component degradation, air conditioning, testing and maintenance, speed, load, and environmental conditions, leading to the final emissions factors. Vehicle mileage and distribution are key parameters reflecting traffic activity [33]. Multiplying these with emissions factors yields total emissions and an emissions inventory, which are crucial for integrating traffic models and emissions models [34]. Mileage is calculated using “non-traffic-volume methods” and “traffic-volume-based methods” [35].
Traffic operation status reflects the activity intensity of motor vehicles in the road network, and its improvement can lead to a reduction in vehicle emissions. Traffic infrastructure planning and construction typically meet (or generate) higher traffic demands. If the emissions reduction effects of traffic operation improvements are not considered, road planning and construction may lead to higher emissions estimates, potentially affecting government approval and funding for projects. Thus, in driving cycle-based emissions modeling systems, it is crucial to accurately represent the impact of traffic operation status on emissions. Currently, speed correction is used to characterize changes in road network traffic operation status. This method uses numerous driving trajectories to establish different speed driving cycles, followed by emissions tests for each cycle [36]. The speed correction factor for the basic emissions factor is derived from the difference between the emissions factors of each driving cycle and the standard cycle [37]. However, since early driving cycles were constructed based on daily travel characteristics, they struggled to capture specific traffic operation states, leading to significant errors and even contradictory results.
Researchers have moved away from speed correction methods based on daily travel characteristics in favor of emissions correction based on link-specific driving cycles, which more accurately reflect speed corrections and traffic conditions [38]. However, this method has several problems:
  • Establishing driving cycles and conducting emissions tests are time-consuming and labor-intensive, and the combination of various road types, speeds, and vehicle categories results in an excessive amount of data and experimental testing.
  • Due to the challenges associated with the first problem, this method lacks fine granularity in differentiating traffic states. For instance, MOBILE6 established only six speed conditions, ranging from free flow to congestion, for highways and only three for urban roads.
  • Fixed driving cycles are inadequate for capturing complex and dynamic traffic conditions, leading to significant deviations between estimated and actual emissions.
  • This method struggles to capitalize on the growing availability of traffic operation data.
Although some models, such as HBEFA using PHEMs to simulate emissions under established driving conditions rather than conducting real vehicle tests, aim to overcome these challenges and reduce the testing workload, issues two, three, and four remain.

4.2. Emissions Models Based on Speed–Acceleration

The speed–acceleration matrix provides an intuitive description of a motor vehicle’s driving state, making it the basis for the most straightforward emissions models that use speed–acceleration matrices and corresponding emissions rates [39]. Typical examples of this emissions model include VT-Micro [40] and MODEM [41]. Using extensive experimental emissions data, this model calculates the corresponding emissions rates within the speed–acceleration matrix. In this matrix, the columns represent continuous speed intervals, the rows represent acceleration intervals, and the matrix values correspond to emissions levels for each speed and acceleration combination. When the driving characteristics of vehicles on a specific road section are entered, the model searches for the corresponding emissions values for each transient state and then simulates and calculates the emissions level of the road section [42].
Using VT-Micro as an example, the model construction method is explained in detail. VT-Micro employs a classification regression tree method to categorize test vehicles into groups and then averages the emissions data for each group based on speed and acceleration to generate representative vehicle emissions data. The best fit for each vehicle type is determined based on the product combination of various powers of speed and acceleration for each emission [43]. The basic form of VT-Micro is structured as follows:
ln ( M O E e ) = ∑ i = 0 3 ∑ j = 0 3 ( L i , j e × v i × a j )   f o r   a ≥ 0 ∑ i = 0 3 ∑ j = 0 3 ( M i , j e × v i × a j )   f o r   a < 0
where MOEe is the instantaneous fuel consumption rate or emissions rate (L/s or mg/s); L i , j e and M i , j e are model regression coefficients for MOEe at speed power i and acceleration j; v is the instantaneous speed of the vehicle, km/h; and a is the instantaneous acceleration of the vehicle, km/s2.
The use of speed and acceleration to characterize the driving state of motor vehicles on actual roads and calculate emissions is an ideal method because it resolves the discrepancies between driving cycles and the actual driving state of vehicles on roads. For instance, Panis et al. [44] demonstrated that emissions are highly sensitive to second-by-second variations in vehicle kinematics, emphasizing the need to move beyond aggregated average speed metrics. Similarly, Rakha et al. [45] developed the VT-Micro model, a microscopic emissions estimation framework utilizing instantaneous speed and acceleration profiles, validated against laboratory data with reasonable accuracy. These studies established the critical importance of high-resolution traffic dynamics in improving emissions estimation. This type of emissions model faces significant challenges in developing the emissions rate module. The first challenge lies in the resolution of the speed–acceleration classification. The emissions rate is highly sensitive to speed and acceleration and requires a highly detailed (high-resolution) division of the speed–acceleration matrix to achieve accurate second-by-second emissions evaluations [46]. However, the development of emissions rates relies on experimental data, which are often derived from driving cycles, making it difficult to ensure the accuracy of emissions calculations. The second challenge is that while this statistical model, which is based solely on speed and acceleration, may be a good fit for specific vehicle data, its generalizability is limited because it fails to account for the underlying vehicle emissions principles.

4.3. Emissions Models Based on Operating Power

To address the limitations of speed–acceleration-based statistical emissions models in explaining the principles of vehicle emissions, researchers have developed emissions models based on vehicle power demand. Examples of this model include CMEM [47], EMIT [48], PHEM [49], and VeTESS [50]. The basic principle shared by these emissions models is the calculation of second-by-second emissions based on vehicle driving modes and engine operating condition parameters.
CMEM is the most widely used of these models. It is designed to model emissions from both light- and heavy-duty vehicles using second-by-second data. However, it goes beyond speed and acceleration by incorporating detailed vehicle parameters such as engine power, fuel consumption, and emissions control technologies [47]. CMEM [48] calculates emissions rates for different combinations of speed and operating mode based on experimental emissions data. Driving mode is defined by vehicle speed, acceleration, power demand, and other factors. After inputting second-by-second data, the driving mode of the vehicle is determined based on the driving mode definition. The corresponding emissions rate for the speed and driving mode combination is then retrieved from the emissions rate module. Finally, the sum of the emissions rates at each moment gives the total vehicle emissions during a trip. The CMEM structure includes two input variable types, three core parameters, and six main modules. The two types of input variables are vehicle-specific variables (weight, maximum torque, etc.) and activity variables (speed and acceleration). The three core parameters are fuel consumption rate (FR), engine emission index (EEI), and catalyst pass fraction (CPF). The six main modules include the engine power module, engine speed module, air/fuel ratio module, fuel consumption module, engine emissions module, and catalytic pass rate module [51].
The calculation is as follows:
Tailpipe   Emission = FR × E E I × C P F
The power parameter-based emissions model poses several challenges to practical application:
  • The model’s detailed characterization of single-vehicle emissions, complex input requirements, and large computational workload make it difficult to support large-scale and rapid evaluation of road networks.
  • Obtaining large volumes of second-by-second speed data from vehicles remains challenging, as the installation of numerous second-by-second data collection devices is costly, and social issues such as user privacy require appropriate solutions. Therefore, simultaneously collecting second-by-second data for all vehicles on an actual road section to calculate emissions is nearly impossible.
  • The volume of wireless data transmission required for second-by-second speed data is extremely large, posing a significant challenge to transmission speed and cost.
  • Handling large volumes of second-by-second data also presents challenges in terms of data storage and processing, making it difficult to manage such extensive datasets.

4.4. Emissions Models Based on Vehicle-Specific Power

The concept of VSP was first introduced by Jiménez Palacios in his doctoral research at the Massachusetts Institute of Technology [52]. VSP is defined as the engine’s power output per ton of mass (including the vehicle’s weight), measured in kW/t (or W/kg). Palacios originally proposed VSP for use in remote sensing exhaust detection. Due to the limitation of remote sensing device test data being collected at specific locations, each vehicle provides only one instantaneous data point and cannot record parameters such as vehicle weight. Therefore, independent of vehicle weight, the VSP variable links the vehicle’s instantaneous motion state to emissions, enabling comparisons and statistical analysis of emissions data from different vehicles and testing methods. In recent years, VSP-based modeling methods have been extensively studied and applied [53,54].
In physics theory, VSP comprehensively accounts for the various tasks performed by motor vehicle engines, including changes in kinetic and potential energy and overcoming rolling friction and air resistance. The derivation process is as follows:
V S P = d d t ( K E + P E ) + F r o l l i n g v + 1 2 ρ a C D A ( v + v w ) 2 v m
where KE is the kinetic energy of the motor vehicle, PE is the potential energy of the motor vehicle, Frolling is the rolling resistance of the motor vehicle, m is the mass of the motor vehicle (kg), v is the speed of the motor vehicle (m/s), vw is the wind speed facing the motor vehicle (m/s), CD (dimensionless) is the wind resistance coefficient, A is the vehicle cross-sectional area (m2), and ρa is the ambient air density, which is 1.207 kg/m3 at 20 °C and at sea level (0 m altitude).
VSP-based modeling methods have been extensively applied to road traffic emissions models, such as MOVES [55] and IVE [56]. The basic approach to the calculation of emissions in these models is similar. The MOVES model’s approach is as follows [57]:
  • Motor vehicles are categorized into 13 types based on technical standards and usage (e.g., motorcycles, light-duty vehicles, buses, trucks, etc.).
  • For each vehicle type, a specific power interval clustering method is used to classify vehicles into 23 driving modes based on speed, acceleration, and specific power. The emissions rate for each mode is then determined through emissions experiments.
  • The proportion of driving time for each emissions source in each mode is calculated and defined as the specific power distribution. This distribution is calculated separately for expressways and non-expressways as well as for different average speeds.
  • The emissions factor for each vehicle type is calculated by multiplying the corresponding emissions rate by the specific power distribution. Regional emissions (national, state/provincial, or city) are then determined by weighting the emissions of each vehicle type according to the proportion of miles it traveled, with adjustments for traffic conditions, fuel, and environmental factors.
MOVES can be applied at various scales, from national inventories to small project-level analyses. It offers flexibility for emissions modeling across various geographic and temporal scales, allowing users to customize the analysis to their specific requirements.
VSP models offer a more dynamic and precise representation of real-world driving conditions by considering variations in speed, acceleration, road gradient, and vehicle load, making them superior in scenarios with fluctuating vehicle behavior [58]. This allows more accurate emissions predictions during real-time driving compared to fixed driving cycle models, which rely on standardized (often idealized) driving patterns that may not reflect the variability in real-world traffic conditions. However, VSP models are typically more data-intensive and require second-by-second driving data, making them harder to implement compared to driving cycle models, which are simpler and rely on average speeds [53]. Compared to speed–acceleration matrix models, VSP models include engine load and road slope, thus providing a more comprehensive view of driving forces, but at the cost of increased computational complexity [59]. Compared to engine power-based models, VSP models offer more detail by calculating emissions based on the interaction between power demand and vehicle mass, whereas engine power models mainly focus on engine output and often overlook critical factors such as road gradient. Overall, VSP-based models are well-suited to high-resolution traffic simulations and policy planning, although they are more complex and data-intensive than other emissions models.

5. Integration of Traffic Models with Emissions Models

In the road traffic system, vehicle driving characteristics and traffic operation status are key factors that determine road traffic activity, thereby influencing emissions levels and pollutant distribution. Therefore, it is crucial to select the appropriate docking parameters, as they represent both vehicle driving characteristics and traffic operating status, while also corresponding to vehicle emissions. Vehicles’ real-time operating characteristics should be linked to pollutant emissions to enable road traffic emissions calculations based on actual traffic conditions.

5.1. Macro-Level Matching Parameters

At the macro level, traffic models supply large-scale, aggregated data for emissions modeling, enabling the estimation of emissions across broad regions, such as national or regional scales. The docking parameters linking traffic models to emissions models at this level typically include the following:
Vehicle Miles Traveled (VMT) or Vehicle Kilometers Traveled (VKT): This parameter represents the total distance driven by vehicles within a specific area. It is one of the most crucial inputs for macro-level emissions estimation, as it reflects overall vehicle activity [60].
Average Speed: The average speed of vehicles on different road types (e.g., urban roads and highways) affects fuel consumption and emissions. For instance, lower speeds caused by congestion can lead to higher emissions, particularly of pollutants such as NOx [38].
Traffic Volume: This represents the total number of vehicles passing a specific point or traveling on a particular road type. Traffic volume data help emissions models estimate emissions based on the intensity of vehicle usage in different areas [61].
Fleet Composition: This refers to vehicle categories (e.g., light-duty vehicles, heavy-duty trucks, electric vehicles) in the traffic stream. Since different vehicle categories have vastly different emissions profiles, an understanding of the vehicle mix is critical for accurate emissions estimation [62].
Fuel Type: Fuel usage patterns across the fleet (e.g., gasoline, diesel, alternative fuels) are crucial for calculating emissions. Emissions, especially of pollutants such as carbon dioxide and particulate matter (PM), vary significantly by fuel type [63].
Road Type and Length: This refers to the types of roads (urban, rural, highway) and the length of each road type within the modeled region. Road types influence average vehicle speed and, consequently, emissions. For example, longer highways with higher speeds tend to have different emissions profiles than congested urban roads [64].
Models requiring macroscopic traffic parameters for emissions estimation include MOBILE, EMFAC, COPERT, HBEFA, CVEM, MOVES, and IVE. These models are designed to estimate emissions over large geographic areas, such as regions, states, or entire countries. To calculate emissions, they rely on cumulative traffic data, including VMT or VKT, average speeds, traffic volumes, road types, and fleet compositions. MOBILE and MOVES are U.S.-based models, with MOVES replacing MOBILE for national and regional emissions inventories. Both use aggregated data, such as VMT and fleet composition. EMFAC is primarily used in California to estimate emissions at the state and county levels using similar parameters. COPERT and HBEFA are widely used across Europe for regional and national emissions inventories, focusing on various vehicle categories and driving conditions. CVEM is designed for China and relies on large-scale data to estimate vehicle emissions across the country’s road networks. IVE is particularly useful for developing countries and offers flexibility in estimating emissions using broader regional data when detailed local data are unavailable.

5.2. Meso-Level Matching Parameters

At the meso level, traffic models focus on smaller geographic areas, such as cities, districts, and corridors, to provide more detailed traffic data compared to macro-level models. The key docking parameters for meso-level models typically include the following:
Traffic Flow: The number of vehicles passing through a specific point or road segment over a given period (e.g., vehicles per hour). This parameter is essential for understanding congestion levels and traffic intensity [21].
Average Speed by Road Segment: The typical speed of vehicles on various road segments or across different road types. Data for this parameter are usually collected via traffic monitoring systems and are crucial for estimating emissions based on traffic patterns and vehicle behaviors [65].
Traffic Density: The number of vehicles per kilometer or mile of road. This parameter is crucial for modeling traffic conditions, particularly in congested urban areas [66].
Fleet Composition: Vehicle categories (e.g., light-duty vehicles, heavy-duty trucks, buses) in specific corridors or urban areas. This parameter is essential because emissions factors may be different between vehicle categories [67].
Vehicle Queuing and Delays: This refers to information about vehicle queuing at intersections or delays caused by traffic signals or congestion. Stop-and-go traffic has a significant impact on emissions, particularly of pollutants such as NOx and PM [68].
Road Types and Characteristics: These include information about road types (e.g., arterial, collector, urban) and their characteristics (e.g., number of lanes and gradients) [69].
Signal Timing and Control Measures: The synchronization of traffic lights and other control measures affects vehicle idling and stop-and-go conditions, both of which affect emissions [70].
Several emissions models require meso-level traffic parameters to estimate emissions at the city or corridor scale. These models include EMFAC, HBEFA, MOVES, COPERT, and PHEM. At this level, the models rely on parameters such as traffic flow, average speed by road segment, and traffic density. These parameters allow the models to more effectively assess emissions for specific road networks or urban environments, where traffic patterns fluctuate due to congestion or signal timing. For instance, EMFAC requires meso-level inputs, such as traffic flow and vehicle speed, to model emissions in California cities, while HBEFA uses similar data to estimate emissions in European urban settings. Although capable of operating at multiple levels, MOVES uses meso-level parameters, such as traffic density, vehicle speed, and queue length, for city-scale or corridor-specific emissions analysis. COPERT, typically used for broader national inventories, can also be applied to meso-level analyses when focusing on specific city networks, relying on fleet composition and traffic patterns. PHEM uses traffic flow and vehicle speed data to model emissions from passenger cars and heavy-duty vehicles, making it suitable for localized scenarios. These models are essential for urban planning, traffic management, and emissions reduction strategies, as they provide detailed insights into how emissions vary across different parts of a city or road network, particularly in areas with variable traffic conditions.

5.3. Micro-Level Matching Parameters

At the micro level, emissions models require highly detailed traffic and vehicle behavior data, often on a second-by-second basis. The key docking parameters at this level include the following:
Second-by-Second Speed and Acceleration: These data provide real-time information on vehicle speed and acceleration at fine time intervals. This detail is essential for calculating emissions during specific driving events like idling, accelerating, braking, and cruising [71].
VSP: VSP, which is the power output of a vehicle per unit mass, captures the impact of different driving behaviors (e.g., climbing, decelerating) on emissions. It helps emissions models more accurately predict emissions under varying load conditions [72].
Engine Load or Vehicle Load: Data on engine load or vehicle load are crucial for estimating emissions related to vehicle performance under specific driving conditions [73].
Road Gradient: Road slope influences engine load, fuel consumption, and emissions, particularly during climbing or descending. Micro-level models must consider the road gradient to provide accurate estimates [74].
Driving Behavior: Parameters such as the frequency and duration of idling, stop-and-go traffic, and deceleration events are crucial for micro-level emissions modeling as they significantly affect pollutants such as NOx and PM [75].
Traffic Signal Timing and Control: Real-time data on traffic signal operations, such as red-light duration and vehicle queuing, are crucial for calculating emissions during frequent stops and starts in urban traffic [76].
Several emissions models require micro-level traffic parameters to estimate emissions at fine temporal resolution and focus on second-by-second data. These models include VT-Micro, CMEM, MODEM, PHEM, VeTESS, and MOVES. At the micro level, these models rely on parameters such as second-by-second speed, acceleration, engine load, and VSP to capture detailed vehicle behavior, including acceleration, deceleration, and idling. VT-Micro and CMEM specifically use second-by-second speed and acceleration to model emissions at a highly granular level, making them ideal for urban traffic conditions under which driving patterns frequently change. MODEM also uses second-by-second driving data to provide highly detailed emissions calculations. PHEM and VeTESS simulate emissions based on transient driving events and require inputs such as speed, throttle position, and gear shifts. MOVES, although scalable across different levels, uses micro-level data for project-level analysis and incorporates second-by-second vehicle dynamics for precise emissions estimates. These models are crucial for detailed simulations of traffic in urban environments, intersections, or heavily congested areas, where vehicle behavior significantly affects emissions.

6. Development Trends in Traffic Emissions Models

6.1. Improving the Coupling Between Emissions Models and Traffic Models

To accurately assess the effects of traffic control strategies, enhancing the coupling between emissions models and traffic simulation models is crucial. This improvement unfolds along two main dimensions.
On the emissions modeling side, progress lies in developing VSP distribution maps that are sensitive to traffic control measures and aligned with the physical layout of road infrastructure—such as intersections, toll booths, and merging areas [77,78]. These VSP maps improve the model’s ability to detect emissions hotspots and evaluate exposure risks in areas with concentrated traffic activity.
On the traffic modeling side, it is essential to refine the compatibility of simulation outputs with emissions model requirements [79]. This involves improving the spatial and temporal detail of key variables such as speed, acceleration, traffic density, and queue length, ensuring these outputs can be directly applied to micro- or meso-scale emissions models [80]. Strengthening this integration will enable more accurate, fine-grained evaluations of traffic control strategies and support the development of environmentally responsive traffic management approaches [81].

6.2. Improve the Applicability of Emissions Models to New Technologies

The rapid evolution of automotive technologies—marked by advances in vehicle intelligence, the proliferation of connected and CAVs, and the rise in AI-driven large-scale models—has become an inevitable trajectory for future transportation systems [82]. To keep pace with these developments, emissions models must be adapted to more accurately reflect the environmental impacts of intelligent and automated vehicle operations. Nonetheless, current research in emissions modeling reveals several significant shortcomings that must be addressed [83].
First, there is a notable lack of comprehensive, systematic evaluations of CAV algorithms under complex traffic conditions [84,85]. Existing studies often oversimplify variables such as fleet composition, roadway types, and congestion levels, limiting their ability to fully assess the potential energy savings and emissions reductions associated with different CAV deployment scenarios.
Second, the validation of CAV-specific simulation models for emissions estimation remains insufficient [86,87]. Few investigations have rigorously compared real-world and simulated data for key operational parameters—such as speed, acceleration, and energy consumption—or systematically analyzed the sources and impacts of simulation errors on emissions predictions.
Third, much of the current work merely applies minor recalibrations to existing emissions models without probing the underlying shifts in key emissions-related parameters under automated driving conditions [88,89]. A deeper mechanistic understanding of how CAV technologies alter vehicle dynamics, along with the development of integrated modeling frameworks that bridge traditional and CAV-specific factors, is urgently needed.
Fourth, the integration of AI technologies into emissions modeling remains largely unexplored. While machine learning and deep learning offer promising tools for capturing complex traffic–emissions interactions, improving real-time prediction accuracy, and fusing data from diverse sources, their systematic application within emissions models is still in its early stages [90,91].

6.3. Enhancing the Use of Emissions Models in the Era of Traffic Big Data

Recent advancements in traffic big data offer significant opportunities to enhance the accuracy and adaptability of emissions models by incorporating real-time information from a wide range of sources. Notably, recent work by Ghaffarpasand et al. introduced an innovative methodology for generating high-resolution urban emissions inventories, achieving a spatial resolution of 15 m and a temporal resolution of two hours [92,93]. Their approach combines traffic flow modeling with atmospheric dispersion techniques, allowing for a much finer representation of spatial and temporal variations in emissions. Integrating such detailed datasets into emissions models can greatly improve their responsiveness to rapid changes in urban traffic environments, addressing persistent challenges in dynamic emissions assessment under real-world conditions.
While models based on vehicle activity and traffic flow remain indispensable tools for environmental evaluation, the growing availability of real-time air quality monitoring technologies—such as satellite-based observations and dense ground-level sensor networks—is reshaping the traditional modeling landscape. These tools enable direct measurements of pollutant concentrations at high spatiotemporal resolutions and, in some cases, may reduce the dependence on conventional traffic-based emissions estimation methods. To maintain the relevance of traffic emissions models within this evolving context, it is essential to acknowledge and adapt to these technological shifts.
However, two key challenges currently limit the full exploitation of traffic big data:
(1) Data heterogeneity challenges: The growing availability of traffic data from GPS devices, vehicle sensors, traffic cameras, and public monitoring platforms presents significant opportunities for improving emissions modeling. However, the diversity in data formats, accuracy levels, and spatial–temporal resolutions introduces substantial integration challenges. Current emissions models often struggle with inconsistencies among datasets, resulting in fragmented and less reliable outputs [94]. To address this, robust data standardization frameworks and unified cleaning protocols are urgently needed to harmonize disparate data streams into validated, coherent inputs [95]. Additionally, dynamic data validation techniques, such as machine learning-based anomaly detection, should be systematically employed to ensure the integrity of real-time data inputs [96].
(2) Inflexibility in model input frameworks: despite the proliferation of diverse traffic datasets, most existing emissions models lack flexible input frameworks capable of dynamically integrating multisource, real-time data. This constraint hampers the models’ ability to accurately capture rapidly changing traffic and environmental conditions [97]. Future models must be designed with dynamic, modular input architecture that can seamlessly ingest vehicle telematics, traffic control data, environmental sensor readings, and weather information [98,99].

7. Conclusions

This study presents a comprehensive and critical review of the integration between traffic dynamics and emissions modeling, examining the methodological evolution, application challenges, and future development trends across macro-, meso-, and micro-scales.
A contribution of this work lies in its detailed classification and comparative evaluation of emissions modeling approaches—driving cycle-based, speed–acceleration matrix-based, engine power-based, and vehicle-specific power (VSP)-based models. By analyzing critical parameters such as second-by-second vehicle speeds, acceleration profiles, VSP distributions, and fleet composition, this study extracts essential elements for strengthening traffic–emissions coupling. It highlights the need for models to better accommodate dynamic, heterogeneous traffic flows while maintaining computational feasibility and predictive accuracy across different spatial and temporal scales.
This review further analyzes the challenges introduced by emerging technologies, notably connected and CAVs and AI-driven modeling techniques. Current emissions models remain inadequate in representing the behavioral complexities of intelligent vehicles and lack rigorous validation against empirical data. It stresses the urgent need for CAV-adaptive emissions frameworks, enhanced real-world calibration methods, and systematic integration of real-time traffic big data to improve the predictive reliability and adaptability of emissions models under rapidly evolving traffic environments.
Finally, the study provides structured guidance for policymakers, traffic management authorities, and researchers. The findings support policymakers and traffic management authorities in selecting modeling approaches that balance resolution, adaptability, and practical implementation demands. For researchers and model developers, the review highlights key priorities for future work, including the integration of connected vehicle behaviors, real-time data fusion, and AI-driven predictive modeling. Strengthening dynamic, traffic-based emissions frameworks remains crucial for improving the reliability, precision, and applicability of road pollution assessments amid increasingly complex and intelligent transportation systems.

Author Contributions

Conceptualization, X.W.; methodology, X.W. and X.Y.; formal analysis, X.W. and J.H.; investigation, X.W. and J.H.; resources, X.W., X.Y. and S.L.; writing—original draft preparation, X.W.; writing—review and editing, X.W., X.Y., J.H. and S.L.; project administration, X.Y. and S.L.; funding acquisition, S.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by National Natural Science Foundation of China (NSFC), grant number 71871130.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

No new data were created.

Acknowledgments

During the preparation of this manuscript, the authors used ChatGPT (OpenAI, GPT-4) solely for the purpose of language refinement. The authors have reviewed and edited the output and take full responsibility for the content of this publication.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Yang, Q.; Hu, X.; Wang, Y.; Liu, Y.; Liu, J.; Ma, J.; Wang, X.; Wan, Y.; Hu, J.; Zhang, Z.; et al. Comparison of the impact of China’s railway investment and road investment on the economy and air pollution emissions. J. Clean. Prod. 2021, 293, 126100. [Google Scholar] [CrossRef]
  2. Lv, Y.; Wang, S.; Gao, Z.; Li, X.; Sun, W. Design of a heuristic environment-friendly road pricing scheme for traffic emission control under uncertainty. J. Environ. Manag. 2019, 236, 455–465. [Google Scholar] [CrossRef] [PubMed]
  3. Sun, D.J.; Wu, S.; Shen, S.; Xu, T. Simulation and assessment of traffic pollutant dispersion at an urban signalized intersection using multiple platforms. Atmos. Pollut. Res. 2021, 12, 101087. [Google Scholar] [CrossRef]
  4. Alam, M.J.; Shahrier, H.; Anik, M.A.H.; Habib, M.A. Activity-based integrated modelling for assessing COVID-19 impacts on transport operations and emissions. Transp. Lett. 2023, 15, 1011–1024. [Google Scholar] [CrossRef]
  5. Jia, T.; Zhang, P.; Chen, B. A microscopic model of vehicle CO2 emissions based on deep learning—A spatiotemporal analysis of taxicabs in Wuhan, China. IEEE Trans. Intell. Transp. Syst. 2022, 23, 18446–18455. [Google Scholar] [CrossRef]
  6. Chen, Y.; Wen, C.; Jiang, C.; Jiang, X. Global sensitivity analysis of VISSIM parameters for project-level traffic emissions: A case study at a signalized intersection. Environ. Technol. 2022, 43, 3801–3820. [Google Scholar] [CrossRef] [PubMed]
  7. Lin, H.; Han, Y.; Cai, W.; Jin, B. Traffic signal optimization based on fuzzy control and differential evolution algorithm. IEEE Trans. Intell. Transp. Syst. 2022, 24, 8555–8566. [Google Scholar] [CrossRef]
  8. Zhang, H.; Guo, S.; Long, X.; Hao, Y. Combined dynamic route guidance and signal timing optimization for urban traffic congestion caused by accidents. J. Adv. Transp. 2023, 2023, 5858614. [Google Scholar] [CrossRef]
  9. Rodriguez-Rey, D.; Guevara, M.; Linares, M.P.; Casanovas, J.; Salmerón, J.; Soret, A.; Jorba, O.; Tena, C.; García-Pando, C.P. A coupled macroscopic traffic and pollutant emission modelling system for Barcelona. Transp. Res. Part D Transp. Environ. 2021, 92, 102725. [Google Scholar] [CrossRef]
  10. Harley, R.A.; Russell, A.G.; McRae, G.J.; Cass, G.R.; Seinfeld, J.H. Photochemical modeling of the Southern California air quality study. Environ. Sci. Technol. 1993, 27, 378–388. [Google Scholar] [CrossRef]
  11. Bishop, G.A. Does California’s EMFAC2017 vehicle emissions model underpredict California light-duty gasoline vehicle NOx emissions? J. Air Waste Manag. Assoc. 2021, 71, 597–606. [Google Scholar] [CrossRef]
  12. Fujita, E.M.; Campbell, D.E.; Zielinska, B.; Chow, J.C.; Lindhjem, C.E.; DenBleyker, A. Comparison of the MOVES2010a, MOBILE6.2, and EMFAC2007 mobile source emission models with on-road traffic tunnel and remote sensing measurements. J. Air Waste Manag. Assoc. 2012, 62, 1134–1149. [Google Scholar] [CrossRef]
  13. Gerard, D.; Lave, L.B. Implementing technology-forcing policies: The 1970 Clean Air Act Amendments and the introduction of advanced automotive emissions controls in the United States. Technol. Forecast. Soc. Chang. 2005, 72, 761–778. [Google Scholar] [CrossRef]
  14. Wolf, J.F.; Beth Farquhar, M. Assessing progress: The state of metropolitan planning organizations under ISTEA and TEA-21. Int. J. Public Adm. 2005, 28, 1057–1079. [Google Scholar] [CrossRef]
  15. Macpherson, A.J.; Simon, H.; Langdon, R.; Misenheimer, D. A mixed integer programming model for National Ambient Air Quality Standards (NAAQS) attainment strategy analysis. Environ. Model. Softw. 2017, 91, 13–27. [Google Scholar] [CrossRef]
  16. Eisinger, D.; Niemeier, D.A. Transportation control measures: Federal requirements and state implementation plan development considerations. Transp. Res. Rec. 2004, 1880, 59–70. [Google Scholar] [CrossRef]
  17. Pokharel, S.S.; Bishop, G.A.; Stedman, D.H. An on-road motor vehicle emissions inventory for Denver: An efficient alternative to modeling. Atmos. Environ. 2002, 36, 5177–5184. [Google Scholar] [CrossRef]
  18. Park, H.; Qi, Y.; Smith, B.L.; O’Meara, K. New methodology for development of regional vehicle-mapping tables for MOBILE emissions model. Transp. Res. Rec. 2006, 1987, 92–102. [Google Scholar] [CrossRef]
  19. Fontaras, G.; Franco, V.; Dilara, P.; Martini, G.; Manfredi, U. Development and review of Euro 5 passenger car emission factors based on experimental results over various driving cycles. Sci. Total Environ. 2014, 468, 1034–1042. [Google Scholar] [CrossRef]
  20. Jia, G.; Huang, Z.; Xu, Y.; Zhong, Z.; Sha, Q.; Huang, X.; Yang, J.; Zheng, J. A dynamic dust emission allocation method and holiday profiles applied to emission processing for improving air quality model performance. Aerosol Air Qual. Res. 2019, 19, 2531–2542. [Google Scholar] [CrossRef]
  21. Jamshidnejad, A.; Papamichail, I.; Papageorgiou, M.; De Schutter, B. A mesoscopic integrated urban traffic flow-emission model. Transp. Res. Part C Emerg. Technol. 2017, 75, 45–83. [Google Scholar] [CrossRef]
  22. Andrade, G.M.S.D.; Araújo, F.W.C.D.; Santos, M.P.M.D.N.; Magnani, F.S. Standardized comparison of 40 local driving cycles: Energy and kinematics. Energies 2020, 13, 5434. [Google Scholar] [CrossRef]
  23. Pacheco, A.F.; Martins, M.E.; Zhao, H. New European drive cycle (NEDC) simulation of a passenger car with a HCCI engine: Emissions and fuel consumption results. Fuel 2013, 111, 733–739. [Google Scholar] [CrossRef]
  24. Tutuianu, M.; Bonnel, P.; Ciuffo, B.; Haniu, T.; Ichikawa, N.; Marotta, A.; Pavlovic, J.; Steven, H. Development of the World-wide harmonized light duty test cycle (WLTC) and a possible pathway for its introduction in the European legislation. Transp. Res. Part D Transp. Environ. 2015, 40, 61–75. [Google Scholar] [CrossRef]
  25. Oh, C.; Cha, G. Impact of fuel, injection type and after-treatment system on particulate emissions of light-duty vehicles using different fuels on FTP-75 and HWFET test cycles. Int. J. Automot. Technol. 2015, 16, 895–901. [Google Scholar] [CrossRef]
  26. Tsai, J.H.; Chiang, H.L.; Hsu, Y.C.; Peng, B.J.; Hung, R.F. Development of a local real world driving cycle for motorcycles for emission factor measurements. Atmos. Environ. 2005, 39, 6631–6641. [Google Scholar] [CrossRef]
  27. US Environmental Protection Agency. User’s Guide to MOBILE6.1 and MOBILE6.2: Mobile Source Emission Factor Model; EPA Report EPA-420-R-03-010; Office of Transportation and Air Quality: Ann Arbor, MI, USA, 2003. Available online: https://www.nrc.gov/docs/ML1118/ML11182C063.pdf (accessed on 5 June 2025).
  28. Ekström, M.; Sjödin, Å.; Andreasson, K. Evaluation of the COPERT III emission model with on-road optical remote sensing measurements. Atmos. Environ. 2004, 38, 6631–6641. [Google Scholar] [CrossRef]
  29. The California Air Resources Board. EMFAC2021 Volume I—User’s Guide. 2021. Available online: https://ww2.arb.ca.gov/sites/default/files/2021-06/emfac202x_users_guide_01112021_final_ada.pdf (accessed on 5 June 2025).
  30. Notter, B.; Cox, B.; Althaus, H. Handbook of Emission Factors for Road Transport (HBEFA) 4.1. ERMES Meeting, Thessaloniki, May 2019. Available online: https://ermes-group.eu/sites/default/files/ERMES_2019_plenary_meeting/7309a_HB41_ERMES_2019-05-14.pdf (accessed on 5 June 2025).
  31. Tang, Y.; Yin, H.; Huang, Z.; Yu, L.; Song, G. Application of CVEM emission model in comprehensive transportation planning: A case study on Chengdu-Chongqing urban agglomeration. Transp. Res. 2018, 4, 31–40. (In Chinese) [Google Scholar]
  32. Lejri, D.; Can, A.; Schiper, N.; Leclercq, L. Accounting for traffic speed dynamics when calculating COPERT and PHEM pollutant emissions at the urban scale. Transp. Res. Part D Transp. Environ. 2018, 63, 588–603. [Google Scholar] [CrossRef]
  33. Yang, Y.; Li, T.; Hu, H.; Zhang, T.; Cai, X.; Chen, S.; Qiao, F. Development and emissions performance analysis of local driving cycle for small-sized passenger cars in Nanjing, China. Atmos. Pollut. Res. 2019, 10, 1514–1523. [Google Scholar] [CrossRef]
  34. Li, Y.; Lv, C.; Yang, N.; Liu, H.; Liu, Z. A study of high temporal-spatial resolution greenhouse gas emissions inventory for on-road vehicles based on traffic speed-flow model: A case of Beijing. J. Clean. Prod. 2020, 277, 122419. [Google Scholar] [CrossRef]
  35. Jung, S.; Kim, J.; Kim, J.; Hong, D.; Park, D. An estimation of vehicle kilometer traveled and on-road emissions using the traffic volume and travel speed on road links in Incheon City. J. Environ. Sci. 2017, 54, 90–100. [Google Scholar] [CrossRef]
  36. Amirjamshidi, G.; Roorda, M.J. Development of simulated driving cycles for light, medium, and heavy duty trucks: Case of the Toronto Waterfront Area. Transp. Res. Part D Transp. Environ. 2015, 34, 255–266. [Google Scholar] [CrossRef]
  37. Song, G.; Yu, L.; Wu, Y. Development of speed correction factors based on speed-specific distributions of vehicle specific power for urban restricted-access roadways. J. Transp. Eng. 2016, 142, 04016001. [Google Scholar] [CrossRef]
  38. Ryu, B.Y.; Jung, H.J.; Bae, S.H. Development of a corrected average speed model for calculating carbon dioxide emissions per link unit on urban roads. Transp. Res. Part D Transp. Environ. 2015, 34, 245–254. [Google Scholar] [CrossRef]
  39. Zhu, S.; Kim, I.; Choi, K. High-resolution simulation-based analysis of leading vehicle acceleration profiles at signalized intersections for emission modeling. Int. J. Sustain. Transp. 2021, 15, 375–385. [Google Scholar] [CrossRef]
  40. Aho, K. Modeling Light Duty Vehicle Emissions Based on Instantaneous Speed and Acceleration Levels. Ph.D. Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA, 2002. Available online: https://vtechworks.lib.vt.edu/server/api/core/bitstreams/5301ce13-7275-4871-953b-d1d60b7a6f18/content (accessed on 5 June 2025).
  41. Boulter, P.G.; Hickman, A.J.; Latham, S.; Layfield, R.; Davison, P.; Whiteman, P. The Impacts of Traffic Calming Measures on Vehicle Exhaust Emissions. Ph.D. Thesis, Middlesex University, London, UK, 2001. Available online: https://trl.co.uk/uploads/trl/documents/TRL482.pdf (accessed on 5 June 2025).
  42. Guo, D.; Zhao, J.; Xu, Y.; Sun, F.; Li, K.; Wang, J.; Sun, Y. The impact of driving conditions on light-duty vehicle emissions in real-world driving. Transport 2020, 35, 379–388. [Google Scholar] [CrossRef]
  43. Xu, Z.; Wei, T.; Easa, S.; Zhao, X.; Qu, X. Modeling relationship between truck fuel consumption and driving behavior using data from internet of vehicles. Comput.-Aided Civ. Infrastruct. Eng. 2018, 33, 209–219. [Google Scholar] [CrossRef]
  44. Panis, L.I.; Broekx, S.; Liu, R. Modelling instantaneous traffic emission and the influence of traffic speed limits. Sci. Total Environ. 2006, 371, 270–285. [Google Scholar] [CrossRef]
  45. Rakha, H.; Ahn, K.; Trani, A. Development of VT-Micro model for estimating hot stabilized light duty vehicle and truck emissions. Transp. Res. Part D Transp. Environ. 2004, 9, 49–74. [Google Scholar] [CrossRef]
  46. Yoon, S.; Li, H.; Jun, J.; Ogle, J.H.; Guensler, R.L.; Rodgers, M.O. Methodology for developing transit bus speed–acceleration matrices for load-based mobile source emissions models. Transp. Res. Rec. 2005, 1941, 26–33. [Google Scholar] [CrossRef]
  47. Scora, G.; Barth, M. Comprehensive Modal Emissions Model (CMEM), version 3.01; User guide; Centre for Environmental Research and Technology, University of California: Riverside, CA, USA, 2006; Available online: https://www.cert.ucr.edu/sites/default/files/2019-07/CMEM_User_Guide_v3.01d.pdf (accessed on 5 June 2025).
  48. Cappiello, A.; Chabini, I.; Nam, E.; Lué, A.; Abou Zeid, M. A statistical model of vehicle emissions and fuel consumption. In Proceedings of the IEEE 5th International Conference on Intelligent Transportation Systems, Singapore, 6 September 2002; IEEE: New York, NY, USA, 2002. [Google Scholar] [CrossRef]
  49. Zamboni, G.; André, M.; Roveda, A.; Capobianco, M. Experimental evaluation of Heavy Duty Vehicle speed patterns in urban and port areas and estimation of their fuel consumption and exhaust emissions. Transp. Res. Part D Transp. Environ. 2015, 35, 1–10. [Google Scholar] [CrossRef]
  50. Pelkmans, L.; Debal, P.; Hood, T.; Hauser, G.; Delgado, M. Development of a simulation tool to calculate fuel consumption and emissions of vehicles operating in dynamic conditions. In SAE Technical Paper 2004, No. 2004-01-1873; SAE International: Warrendale, PA, USA, 2004. [Google Scholar]
  51. Li, Q.; Jia, T. Correction of measured taxicab exhaust emission data based on CMEM model. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2017, 42, 1245–1250. [Google Scholar] [CrossRef]
  52. Jimenez-Palacios, J.L. Understanding and Quantifying Motor Vehicle Emissions with Vehicle Specific Power and TILDAS Remote Sensing. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA, USA, 1999. Available online: https://dspace.mit.edu/handle/1721.1/44505 (accessed on 5 June 2025).
  53. Zhang, Z.; Song, G.; Chen, J.; Zhai, Z.; Yu, L. Development of a simplified model of speed-specific vehicle-specific power distribution based on vehicle weight for fuel consumption estimates. Transp. Res. Rec. 2020, 2674, 52–67. [Google Scholar] [CrossRef]
  54. Li, J.; He, H.; Peng, B. Development of operating mode distribution models for light-duty vehicles on unrestricted access roadways. Transp. Res. Rec. 2020, 2674, 416–428. [Google Scholar] [CrossRef]
  55. US Environmental Protection Agency. Motor Vehicle Emission Simulator. MOVES 2010 User Guide; EPA-420-B-09-041; Office of Transportation and Air Quality: Ann Arbor, MI, USA, 2009. Available online: https://www.nrc.gov/docs/ML1127/ML112720505.pdf (accessed on 5 June 2025).
  56. Hui, G.; Zhang, Q.Y.; Yao, S.; Wang, D.H. Evaluation of the International Vehicle Emission (IVE) model with on-road remote sensing measurements. J. Environ. Sci. 2007, 19, 818–826. [Google Scholar] [CrossRef]
  57. US Environmental Protection Agency. MOVES4 Technical Guidance: Using MOVES to Prepare Emission Inventories for State Implementation Plans and Transportation Conformity; EPA-420-B-23-011; Office of Transportation and Air Quality: Ann Arbor, MI, USA, 2023. Available online: https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P101862T.pdf (accessed on 5 June 2025).
  58. Peng, F.; Zhang, Y.; Song, G.; Huang, J.; Zhai, Z.; Yu, L. Evaluation of real-world fuel consumption of hybrid-electric passenger car based on speed-specific vehicle power distributions. J. Adv. Transp. 2023, 2023, 9016510. [Google Scholar] [CrossRef]
  59. Chen, Y.; Li, S.; Li, Y. A review on quantitative energy consumption models from road transportation. Energies 2023, 17, 2. [Google Scholar] [CrossRef]
  60. Alberini, A.; Burra, L.T.; Cirillo, C.; Shen, C. Counting vehicle miles traveled: What can we learn from the NHTS? Transp. Res. Part D Transp. Environ. 2021, 98, 102984. [Google Scholar] [CrossRef]
  61. Liu, S.; Hellendoorn, H.; De Schutter, B. Model predictive control for freeway networks based on multi-class traffic flow and emission models. IEEE Trans. Intell. Transp. Syst. 2016, 18, 306–320. [Google Scholar] [CrossRef]
  62. Martinet, S.; Liu, Y.; Jean-Soro, L.; Goriaux, M.; Andre, M. In-situ estimation of non-regulated pollutant emission factors in an urban area of Nantes, France, with fleet composition characterization. Transp. Res. Part D Transp. Environ. 2019, 76, 193–210. [Google Scholar] [CrossRef]
  63. Ni, W.; Li, H.; Zhang, Y.; Zou, Z. Effects of fuel type and operation parameters on combustion and NOx emission of the iron ore sintering process. Energies 2019, 12, 213. [Google Scholar] [CrossRef]
  64. Zhai, Z.; Song, G.; Liu, Y.; Cheng, Y.; He, W.; Yu, L. Characteristics of operating mode distributions of light duty vehicles by road type, average speed, and driver type for estimating on-road emissions: Case study of Beijing. J. Intell. Transp. Syst. 2019, 23, 191–202. [Google Scholar] [CrossRef]
  65. Tu, R.; Kamel, I.; Wang, A.; Abdulhai, B.; Hatzopoulou, M. Development of a hybrid modelling approach for the generation of an urban on-road transportation emission inventory. Transp. Res. Part D Transp. Environ. 2018, 62, 604–618. [Google Scholar] [CrossRef]
  66. Wang, S.; Xie, X.; Ju, R. A mesoscopic traffic data assimilation framework for vehicle density estimation on urban traffic networks based on particle filters. Entropy 2019, 21, 358. [Google Scholar] [CrossRef]
  67. Tsiakmakis, S.; Fontaras, G.; Anagnostopoulos, K.; Ciuffo, B.; Pavlovic, J.; Marotta, A. A simulation-based approach for quantifying CO2 emissions of light-duty vehicle fleets. A case study on WLTP introduction. Transp. Res. Procedia 2017, 25, 3898–3908. [Google Scholar] [CrossRef]
  68. Song, G.; Zhou, X.; Yu, L. Delay correction model for estimating bus emissions at signalized intersections based on vehicle specific power distributions. Sci. Total Environ. 2015, 514, 108–118. [Google Scholar] [CrossRef]
  69. Shang, R.; Zhang, Y.; Shen, Z.M.; Zhang, Y. Analyzing the effects of road type and rainy weather on fuel consumption and emissions: A mesoscopic model based on big traffic data. IEEE Access 2021, 9, 62298–62315. [Google Scholar] [CrossRef]
  70. Zhao, H.; He, R.; Yin, N. Modeling of vehicle CO2 emissions and signal timing analysis at a signalized intersection considering fuel vehicles and electric vehicles. Eur. Transp. Res. Rev. 2021, 13, 5. [Google Scholar] [CrossRef]
  71. Hu, H.; Yoon, C.; Yang, C.; Kim, J. Application of an emission estimation methodology to reflect microscale road driving conditions. Int. J. Highw. Eng. 2016, 18, 115–125. [Google Scholar] [CrossRef]
  72. Zhai, Z.; Song, G.; Lu, H.; He, W.; Yu, L. Validation of temporal and spatial consistency of facility- and speed-specific vehicle-specific power distributions for emission estimation: A case study in Beijing, China. J. Air Waste Manag. Assoc. 2017, 67, 949–957. [Google Scholar] [CrossRef] [PubMed]
  73. Wang, X.; Song, G.; Zhai, Z.; Wu, Y.; Yin, H.; Yu, L. Effects of vehicle load on emissions of heavy-duty diesel trucks: A study based on real-world data. Int. J. Environ. Res. Public Health 2021, 18, 3877. [Google Scholar] [CrossRef] [PubMed]
  74. Wang, X.; Song, G.; Wu, Y.; Zhai, Z. Road grade estimation based on power demand difference of heavy-duty diesel trucks for emission estimation. Transp. Res. Rec. 2023, 2677, 129–150. [Google Scholar] [CrossRef]
  75. Xu, J.; Saleh, M.; Hatzopoulou, M. A machine learning approach capturing the effects of driving behaviour and driver characteristics on trip-level emissions. Atmos. Environ. 2020, 224, 117311. [Google Scholar] [CrossRef]
  76. Zhou, S.; Li, H.; Zhang, Q.; Gao, M.; Ran, B. Traffic signal coordination control optimization considering vehicle emissions on urban arterial road. J. Comput. Methods Sci. Eng. 2021, 21, 233–239. [Google Scholar] [CrossRef]
  77. Zhai, Z.; Tu, R.; Xu, J.; Wang, A.; Hatzopoulou, M. Capturing the variability in instantaneous vehicle emissions based on field test data. Atmosphere 2020, 11, 765. [Google Scholar] [CrossRef]
  78. Xu, J.; Saleh, M.; Wang, A.; Tu, R.; Hatzopoulou, M. Embedding local driving behaviour in regional emission models to increase the robustness of on-road emission inventories. Transp. Res. Part D Transp. Environ. 2019, 73, 1–14. [Google Scholar] [CrossRef]
  79. Xu, J.; Hilker, N.; Turchet, M.; Al-Rijleh, M.; Tu, R.; Wang, A.; Fallah-Shorshani, M.; Evans, G.; Hatzopoulou, M. Contrasting the direct use of data from traffic radars and video-cameras with traffic simulation in the estimation of road emissions and PM hotspot analysis. Transp. Res. Part D Transp. Environ. 2018, 62, 90–101. [Google Scholar] [CrossRef]
  80. Shan, X.; Hao, P.; Chen, X.; Boriboonsomsin, K.; Wu, G.; Barth, M.J. Vehicle energy/emissions estimation based on vehicle trajectory reconstruction using sparse mobile sensor data. IEEE Trans. Intell. Transp. Syst. 2018, 20, 716–726. [Google Scholar] [CrossRef]
  81. Jamshidnejad, A.; Papamichail, I.; Papageorgiou, M.; Schutter, B. Sustainable model-predictive control in urban traffic networks: Efficient solution based on general smoothening methods. IEEE Trans. Control Syst. Technol. 2018, 26, 813–827. [Google Scholar] [CrossRef]
  82. Kemp, N.; Keoleian, G.; He, X.; Kasliwal, A. Life cycle greenhouse gas impacts of a connected and automated SUV and van. Transp. Res. Part D Transp. Environ. 2020, 83, 102375. [Google Scholar] [CrossRef]
  83. Manjunatha, P.; Roy, S.; Elefteriadou, L.; Guin, A.; Hunter, M. Evaluation of the operational effects of autonomous and connected vehicles through microsimulation. Transp. Res. Rec. 2022, 2676, 69–84. [Google Scholar] [CrossRef]
  84. Peng, J.; Wei, S.; Chai, L. Strategy of lane-changing coupling process for connected and automated vehicles in mixed traffic environment. Transp. B Transp. Dyn. 2022, 11, 979–995. [Google Scholar] [CrossRef]
  85. Fang, X.; Li, H.; Tettamanti, T.; Eichberger, A.; Fellendorf, M. Effects of automated vehicle models at the mixed traffic situation on a motorway scenario. Energies 2022, 15, 2008. [Google Scholar] [CrossRef]
  86. Le Rhun, A.; Bonnans, F.; De Nunzio, G.; Leroy, T.; Martinon, P. A stochastic data-based traffic model applied to vehicles energy consumption estimation. IEEE Trans. Intell. Transp. Syst. 2019, 21, 3025–3034. [Google Scholar] [CrossRef]
  87. Huang, C.; Hu, P.; Lian, J. Online optimum velocity calculation under V2X for smart new energy vehicles. Trans. Inst. Meas. Control 2021, 43, 2368–2377. [Google Scholar] [CrossRef]
  88. Xing, Y.; Lv, C.; Mo, X.; Hu, Z.; Huang, C.; Hang, P. Toward safe and smart mobility: Energy-aware deep learning for driving behavior analysis and prediction of connected vehicles. IEEE Trans. Intell. Transp. Syst. 2021, 22, 4267–4280. [Google Scholar] [CrossRef]
  89. Zhao, B.; Lin, Y.; Hao, H.; Yao, Z. Fuel consumption and traffic emissions evaluation of mixed traffic flow with connected automated vehicles at multiple traffic scenarios. J. Adv. Transp. 2022, 2022, 6345404. [Google Scholar] [CrossRef]
  90. Zhang, Y.; Wang, L.; Sun, T.; Huang, B. Vehicle-based monitoring and AI unravel patterns of on-road carbon and pollutant emissions. Innov. Geosci. 2024, 2, 100085. [Google Scholar] [CrossRef]
  91. Zhong, W.; Liu, Y.; Dong, K.; Ni, G. Assessing the synergistic effects of artificial intelligence on pollutant and carbon emission mitigation in China. Energy Econ. 2024, 138, 107829. [Google Scholar] [CrossRef]
  92. Ghaffarpasand, O.; Pope, F.D. Telematics data for geospatial and temporal mapping of urban mobility: Fuel consumption, and air pollutant and climate-forcing emissions of passenger cars. Sci. Total Environ. 2023, 894, 164940. [Google Scholar] [CrossRef]
  93. Ghaffarpasand, O.; Pope, F.D. Telematics data for geospatial and temporal mapping of urban mobility: New insights into travel characteristics and vehicle specific power. J. Transp. Geogr. 2024, 115, 103815. [Google Scholar] [CrossRef]
  94. Liu, J.; Han, K.; Chen, X.M.; Ong, G.P. Spatial-temporal inference of urban traffic emissions based on taxi trajectories and multi-source urban data. Transp. Res. Part C Emerg. Technol. 2019, 106, 145–165. [Google Scholar] [CrossRef]
  95. Zhang, S.; Niu, T.; Wu, Y.; Zhang, K.M.; Wallington, T.J.; Xie, Q.; Wu, X.; Xu, H. Fine-grained vehicle emission management using intelligent transportation system data. Environ. Pollut. 2018, 241, 1027–1037. [Google Scholar] [CrossRef] [PubMed]
  96. Le Cornec, C.M.; Molden, N.; van Reeuwijk, M.; Stettler, M.E. Modelling of instantaneous emissions from diesel vehicles with a special focus on NOx: Insights from machine learning techniques. Sci. Total Environ. 2020, 737, 139625. [Google Scholar] [CrossRef]
  97. Koupal, J.W.; DenBleyker, A.; Manne, G.; Batista, M.H.; Schmitt, T. Capabilities and limitations of telematics for vehicle emissions inventories. Transp. Res. Rec. 2022, 2676, 49–57. [Google Scholar] [CrossRef]
  98. Ding, H.; Cai, M.; Lin, X.; Chen, T.; Li, L.; Liu, Y. RTVEMVS: Real-time modeling and visualization system for vehicle emissions on an urban road network. J. Clean. Prod. 2021, 309, 127166. [Google Scholar] [CrossRef]
  99. Xu, Z.; Wang, R.; Kang, Y.; Zhang, Y.; Xia, X.; Wang, R. A deep transfer NOx emission inversion model of diesel vehicles with multisource external influence. J. Adv. Transp. 2021, 2021, 4892855. [Google Scholar] [CrossRef]
Table 1. Classification of emissions models and related information about the models.
Table 1. Classification of emissions models and related information about the models.
Model TypeModel NameFull NameDeveloper, PublisherCountries/StatesReleased inUpdated inHomepage
(accessed on 5 June 2025)
Driving Cycle CategoryMOBILEMobile Source Emission FactorU.S. EPAU.S. states 19782002https://www.epa.gov/moves/description-and-history-mobile-highway-vehicle-emission-factor-model
EMFACEmission Factor ModelCalifornia Air Resources BoardCalifornia19602025https://ww2.arb.ca.gov/our-work/programs/msei/on-road-emfac
COPERTComputer Program to Calculate Emissions from Road TransportEuropean Environment AgencyMany countries in Europe19902024https://copert.emisia.com/
HBEFAHandbook of Emission Factors for Road TransportEnvironmental Protection Agencies of Germany, Switzerland, and AustriaGermany, Austria, Switzerland, Sweden, Norway, and France19952022https://www.hbefa.net/
CVEMChina Vehicle Emissions ModelMinistry of Ecology and Environment of the People’s Republic of ChinaChina20092019https://www.vecc.org.cn/front/channel/7678
Speed–Acceleration CategoryVT-MicroVirginia Tech Microscopic ModelVirginia TechNorth America2002-https://www.vtti.vt.edu/give/environment/csm.html
MODEMModal Emissions ModelEuropean DRIVEEurope19922001https://cordis.europa.eu/project/id/V1053
Engine Power CategoryCMEMComprehensive Modal Emissions ModelUniversity of California RiversideNorth America19992005https://www.cert.ucr.edu/cmem
PHEMPassenger Car and Heavy Duty Emissions ModelGraz University of TechnologyNorth America19992016https://www.itna.tugraz.at/en/research/areas/em/simulation/phem.html
VeTESSVehicle Transient Emissions Simulation SoftwareMotor Industry Research AssociationEurope2003-https://doi.org/10.4271/2004-01-1873
EMITEmissions from Traffic ModelMassachusetts Institute of TechnologyNorth America2002-https://www.cerc.co.uk/environmental-software/EMIT-tool.html
Specific Power Distribution CategoryMOVESMotor Vehicle Emission SimulatorU.S. EPAU.S. states20042024https://www.epa.gov/moves
IVEInternational Vehicle Emissions ModelU.S. EPADeveloping countries20042008https://ndcpartnership.org/knowledge-portal/climate-toolbox/international-vehicle-emissions-ive-model
Table 2. Comparative overview of major traffic emissions models.
Table 2. Comparative overview of major traffic emissions models.
Model NamesRequired InputsComputational ComplexitiesSpatial/Temporal ResolutionsKey StrengthsMain Limitations
MOBILEAvg speed, VMT, fleet dataLowMacroSimple, historically usedLow sensitivity to dynamic changes
EMFACTraffic flow, speed, fleet compositionModerateMacro/MesoRegularly updated fleet dataRegion-specific calibration
COPERTSpeed distributions, traffic dataModerateMacro/MesoFlexible, good for EU contextLimited micro-scale detail
HBEFADetailed traffic situationsModerateMeso/MicroHigh-resolution scenariosComplex input requirements
CVEMVehicle type, speed, gradientLowMacroLocalized for UKLimited temporal resolution
VT-MicroSpeed, accelerationHighMicroCaptures second-by-second variationRequires high-resolution input
MODEMSpeed, road typeModerateMicroRoad type adaptabilityCalibration-intensive
CMEMEngine power, speed, accelerationHighMicroPhysics-based, real-time capableData-intensive and complex
PHEMInstantaneous speed, engine loadHighMicroDetailed engine modelingLab-level calibration required
VeTESSVehicle trajectory, behaviorHighMicroBehaviorally realisticLimited to behavioral assumptions
EMITInstantaneous operating modeHighMicroHigh temporal resolutionRequires extensive sensor data
MOVESSpeed, acceleration, road grade, VSPHighMeso/MicroHighly detailed, EPA standardHigh data and computational demand
IVEFuel type, maintenance, local dataModerateMacro/MesoDesigned for local conditionsLimited by self-reporting accuracy
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wang, X.; Yue, X.; Huang, J.; Li, S. Integrating Traffic Dynamics and Emissions Modeling: From Classical Approaches to Data-Driven Futures. Atmosphere 2025, 16, 695. https://doi.org/10.3390/atmos16060695

AMA Style

Wang X, Yue X, Huang J, Li S. Integrating Traffic Dynamics and Emissions Modeling: From Classical Approaches to Data-Driven Futures. Atmosphere. 2025; 16(6):695. https://doi.org/10.3390/atmos16060695

Chicago/Turabian Style

Wang, Xin, Xianfei Yue, Jianchang Huang, and Shubin Li. 2025. "Integrating Traffic Dynamics and Emissions Modeling: From Classical Approaches to Data-Driven Futures" Atmosphere 16, no. 6: 695. https://doi.org/10.3390/atmos16060695

APA Style

Wang, X., Yue, X., Huang, J., & Li, S. (2025). Integrating Traffic Dynamics and Emissions Modeling: From Classical Approaches to Data-Driven Futures. Atmosphere, 16(6), 695. https://doi.org/10.3390/atmos16060695

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop