Individual Vehicle Level Research Articles

Field Evaluation of Yellow Change and Red Clearance Intervals with Subsequent Permissive Left-Turn Flashing Yellow Arrow Indication

The flashing yellow arrow (FYA) indication has become commonly used for communicating permissive left-turn operations to road users. A conventional signal phasing communicated to left-turning drivers consists of a leading protected phase followed by a permissive phase using a FYA indication, often referred to as protected-permissive. To transition from the protected to the permissive phase, change (steady yellow arrow) and clearance (steady red arrow, all-red) intervals, or only a change interval, may be used. There is no specific guidance on the use of clearance intervals for left turns with protected-permissive phase and FYA indication. In this study, field data were collected from different geographical regions in the United States to evaluate change and clearance intervals with the objective of developing guidance. Video data were recorded at 37 intersections during peak hours, for approximately 142 h, across eight states. Overall, 4,001 observations of vehicles turning left during the change or clearance interval were analyzed. Field observations were evaluated at the approach level to assess left-turn violation rate and violations per cycle as a function of change and clearance interval configuration. At the individual vehicle level, logistic regression was implemented to evaluate the effect of change interval and clearance interval duration, delayed onset of FYA, and regional variation. The results of the analysis at the approach and individual vehicle level were consistent and indicate that: (i) a clearance interval should be included, (ii) delaying the onset of FYA indication with an extension beyond the all-red clearance interval would reduce left-turn signal violations, and (iii) duration of change intervals has a marginal impact on left-turn signal violations.

Multi-Agent Deep Reinforcement Learning for Multi-Lane Freeways Differential Variable Speed Limit Control in Mixed Traffic Environment

In advanced freeway traffic management systems, variable speed limit control (VSLC) is frequently discussed as one of the control measures. However, in a mixed traffic environment where connected and automated vehicles (CAVs) and human-driven vehicles coexist, the existing VSLC strategies for multi-lane freeways have two major shortcomings: the lack of precise control at the individual vehicle level, and the implementation of uniform VSLC across all lanes. This paper proposes a novel differential variable speed limit control (DVSLC) strategy based on multi-agent reinforcement learning (MARL) in a mixed traffic environment (abbreviated as MARL-DVSLC). The proposed MARL-DVSLC approach utilized a centralized training with decentralized execution paradigm to learn the joint actions of variable speed limit controllers across all lanes, thereby setting different speed limits for each lane. The reward function takes into account the total time spent (TTS) on freeways to improve traffic mobility. Note that MARL-DVSLC disseminates speed limit information to CAVs via infrastructure-to-vehicle (I2V) communication. The effectiveness of MARL-DVSLC is verified under different simulation scenarios. Moreover, its performance is compared with the feedback-based VSLC method, the DVSLC method based on deep deterministic policy gradient (DDPG) (abbreviated as DDPG-DVSLC), and the no-control case in relation to performance. The results indicate that the proposed strategy can effectively improve traffic efficiency and reduce the spatiotemporal range of traffic congestion at a 30% penetration rate of CAVs. Compared with the suboptimal DDPG-DVSLC method, the proposed strategy can improve TTS by 12.88% with stable traffic demand and 10.24% with fluctuating traffic demand.

Road traffic can be predicted by machine learning equally effectively as by complex microscopic model

Since high-quality real data acquired from selected road sections are not always available, a traffic control solution can use data from software traffic simulators working offline. The results show that in contrast to microscopic traffic simulation, the algorithms employing neural networks can work in real-time, so they can be used, among others, to determine the speed displayed on variable message road signs. This paper describes an experiment to develop and test machine learning models, i.e., long short-term memory, gated recurrent unit recurrent networks, and stacked autoencoder networks. It compares their effectiveness with traffic prediction results generated using a widely recognized traffic simulator that analyzes traffic at the level of individual vehicles.

Capacity at All-Way Stop Control Intersections: Case Study

This paper presents an empirical investigation into the capacity of all-way stop-controlled (AWSC) intersections. Video data was collected over four days at an AWSC intersection site in Bozeman, Montana. The site is characterized by single-lane approaches and high level of vehicular and pedestrian traffic. Using strict protocols, video records were processed at the individual vehicle level and several information metrics were extracted for each vehicle in the data set on all approaches. Study results indicate that the total intersection capacity at the study site varied between 400 and 1,400 vehicles per hour. The study suggests that the wide range of capacity observations is largely associated with the pedestrian crossing activity at the study site. Statistical tests confirmed that both pedestrian crossing activity and the level of conflict have significant effects on intersection capacity at the 95% confidence level. For movement type, the right-turn movement was not found to have a significant effect on intersection capacity while left-turn movement was found to negatively affect the intersection capacity. The results presented in this paper offer valuable information on AWSC intersection capacity, given the limited amount of information in the literature and the dated nature of those empirical observations.

Optimal Trajectory Planning of Connected and Automated Vehicles at On-Ramp Merging Area

Cooperative Adaptive Cruise Control (CACC) systems can significantly improve traffic safety and roadway capacity utilizing short following gaps of vehicles enabled by inter-vehicle communications. However, due to merging processes occurring at freeway merging areas, existing CACC operation approaches are generally not applicable and the operation will have to revert back to Adaptive Cruise Control (ACC) or human-driven mode, which in turn will result in a capacity drop. This paper proposes an optimal trajectory optimization strategy for Connected and Automated Vehicles (CAVs) to cooperatively carry out mainline platooning and on-ramp merging. Firstly, a control framework of the CACC is adopted for a longitudinal control of CAVs, which helps individual CAVs to join platoons and to maintain platoon operations. Secondly, to ensure smooth lane-changing executions while achieving stable platoons, an optimal controller that considers lane-changing motivation of merging vehicles and impact of merging on platoons, is proposed. Third, a Legendre pseudo-spectral algorithm is applied to transform the controller into a simpler nonlinear programming problem and to efficiently solve it. Simulation assessments of the proposed method are conducted at both individual vehicle level and traffic-flow level. At the individual vehicle level, the proposed method has the potential to improve the traffic safety without compromising fuel consumption and emissions compared with unoptimized feasible schemes. At a traffic-flow level, an online evaluation platform is implemented, and a typical freeway on-ramp area is studied. The simulation results have demonstrated that the proposed controller provides significant improvements in terms of efficiencies in traffic operations.

Autonomous and Semiautonomous Intersection Management: A Survey

Intersections are a major source of traffic delays and accidents within modern transportation systems. Compared to signalized intersection management, autonomous intersection management (AIM) coordinates the intersection crossing at an individual vehicle level, which provides additional flexibility. AIM can potentially eliminate stopping in intersection crossings due to traffic lights while maintaining a safe separation among conflicting movements. In this article, the state-of-the-art AIM research among various disciplines (e.g., traffic engineering and control engineering) is surveyed from the perspective of three hierarchical layers: corridor coordination layer, intersection management layer, and vehicle control layer. The key aspects of AIM designs are discussed in detail, including conflict detection schemes, priority rules, control centralization, computation complexity, and so on. Potential improvements for AIM evaluation with the emphasis on realistic scenarios are provided. This article serves as a comprehensive review of AIM design and provides promising directions for future research.

Feasibility of Operating a Heavy-Duty Battery Electric Truck Fleet for Drayage Applications

Vehicle fleet electrification is regarded as one major pathway toward achieving energy independence and reducing air pollution and greenhouse gas emissions. Compared to light-duty and medium-duty vehicles, electrification of heavy-duty vehicles, especially Class 8 trucks, is more challenging owing to the battery size required to attain the driving range necessary for their operating goals. As drayage trucks generally have a limited daily mileage, return to a home base every night, and spend a large amount of time creeping and idling, drayage operation has been the first targeted application for Class 8 electric trucks. The feasibility of operating battery electric drayage trucks at the individual vehicle level has recently been demonstrated. However, questions remain as to whether these trucks are capable of meeting the needs of typical drayage operation at the fleet level. This paper presents a feasibility analysis of operating an electric truck fleet based on real-world operation data of a diesel drayage operator in Southern California. Second-by-second activity data collected from 20 trucks in the fleet were used to estimate the corresponding electric energy consumption and the state of charge of the battery using a microscopic electric energy consumption model. An algorithm for generating tours of drayage activity from the collected data was developed and implemented. Multiple scenarios with different battery charging and truck scheduling assumptions were analyzed. The results show that 85% of the tours could be served by electric trucks if there is opportunity for charging at the home base during the time gap between consecutive tours.

A decision-making framework for safe operations of unmanned aerial vehicles in urban scenarios

This paper presents a decision-making scheme at the level of individual unmanned aerial vehicles (UAVs) with the goal of maintaining safe operations for urban mobility. The decision-making approach for a single UAV will consider the risks associated with the current trajectory given the existing environmental conditions and the state of the vehicle. The proposed scheme combines the analysis of system performance, environmental conditions, and mission level parameters for contingency management, i.e., make a determination on: (1) to abort mission and land safely; (2) re-plan current mission in full or abbreviated form; and (3) change mission. A path planning and trajectory optimization algorithm with the goal of minimizing the overall risk of mission failure by considering a number of factors such as the uncertainties in the environment and operating state of the vehicle is proposed. We will consider the mission failure as the loss of control of the vehicle resulting in a collision with other objects or a crash into the ground. An offline part of the framework generates an initial mission plan by considering the state of the vehicle, the environmental, conditions, and the static features of a map of the environment. Once the vehicle takes off, the risk of mission’ failure associated with the remaining trajectory is re-computed in an online framework to assess whether re-planning is required or not. A key challenge that we consider in this paper is to study the effects of multiple interacting subsystems of the UAV on system performance, especially under degraded conditions.

Automated Emergency Vehicle Control Strategy Based on Automated Driving Controls

This study proposes an integrated driving control strategy by taking advantage of the automated driving technology at the individual vehicle level and the traffic signal preemption strategy at the traffic infrastructure level. This aims to facilitate an automated driving-based emergency vehicle control and ultimately to achieve efficient and safe control of emergency vehicles. To this end, this study developed the integrated emergency vehicle control logic, implemented the logic in the microscopic traffic simulation environment using the simulation software’s application programming interface capability, and evaluated the impacts of the proposed emergency vehicle control logic in the aspects of mobility and safety with different driving aggressiveness and preemption initiation settings. The study’s results show that the proposed emergency vehicle control logic achieved benefits on mobility and safety and the benefits of emergency vehicle control strategy can be maximized when the signal preemption and the automated driving control operate in collaboration. Therefore, the proposed integrated approach of automated driving controls and signal preemption will be a great reference for enhancing automated driving technologies supporting a safe and fast mobility solution.

A state-constrained optimal control based trajectory planning strategy for cooperative freeway mainline facilitating and on-ramp merging maneuvers under congested traffic

This paper presents a trajectory planning strategy for connected automated vehicles (CAVs) to cooperatively carry out mainline facilitating (i.e. gap development) and on-ramp merging maneuvers. The trajectory planning tasks of the mainline facilitating vehicle and the merging vehicle are formulated as two related optimal control problems. The motivation behind the proposed strategy is to restrain a facilitating maneuver’s impact on following traffic. To this end, the proposed strategy bounds the speed of the facilitating maneuver from below and meanwhile ensures that the task of gap development can still be fulfilled. Because of the existence of the speed bound, the optimal control problem of the facilitating vehicle becomes constrained in state, in addition to the control constraints. A Pontryagin Maximum Principle (PMP) with state constraints is applied to rigorously derive the analytical solution. The main difficulty of the analytical procedure exists in the fact that the first-order necessary condition on the extremality of the Hamiltonian cannot yield useful information on the property of the optimal control history with regard to making the optimal speed trajectory to satisfy the speed constraint. As a result, additional conditions have to be explored, notably the so-called “jump conditions”, among others. Taking advantage of the analytical solution, the proposed strategy is then implemented under a model predictive control framework.Simulation assessments of the proposed strategy are conducted at two levels – individual vehicle level and traffic flow level. At the individual vehicle level, the proposed strategy shows potential to reduce the risk of rear-end collision between the facilitating vehicle and the following vehicle, the most vulnerable pair of vehicles under the influence of gap development. At the traffic flow level, coupled with Aimsun, the proposed strategy is assessed under mixed traffic flow conditions, with various penetration rates of CAVs. The results show that it has potential to generate lower speed variations of traffic flow, a critical factor in traffic flow safety. Meanwhile it does not show negative impact on traffic efficiency in the simulation, and is likely to improve traffic efficiency in the real world. A sensitivity analysis of the effect of the facilitating maneuver’s lower speed bound is also conducted. Although there exist several limitations with this study, it sheds some light on future research.

Individual Truck Speed Estimation from Advanced Single Inductive Loops

Trucks are an essential element in freight movements, transporting 73% of freight tonnage among all modes. However, they are also associated with severe adverse impacts on roadway congestion, safety, and air pollution. Truck speed by truck body types has been considered as an indicator of traffic conditions and roadway emissions. Even though vehicle speed estimation has been researched for decades, there exists a gap in estimating truck speeds particularly at the individual vehicle level. The wide diversity of vehicle lengths associated with trucks makes it especially challenging to estimate truck speeds from conventional inductive loop detector data. This paper presents a new speed estimation model which uses detailed vehicle signature data from single inductive loop sensors equipped with advanced detectors to provide accurate truck speed estimates. This model uses new inductive signature features that show a strong correlation with truck speed. A modified feature weighting K-means algorithm was used to cluster vehicle length related features into 16 specific groups. Individual vehicle speed regression models were then developed within each cluster. Finally, a multi-layer perceptron neural network model was used to assign single loop signatures to the pre-determined speed related clusters. The new model delivered promising estimation results on both a truck-focused dataset and a general traffic dataset.

Machine Learning Approach to Short-Term Traffic Congestion Prediction in a Connected Environment

Traffic congestion is a complex phenomenon triggered by a combination of multiple interacting factors. One of the main factors is the disturbances caused by individual vehicles, which cannot be identified in aggregate traffic data. Advances in vehicle wireless communications present new opportunities to measure traffic perturbations at the individual vehicle level. The key question is whether it is possible to find the relationship between these perturbations and shockwave formation and utilize this knowledge to improve the identification and prediction of congestion formation. Accordingly, this paper explores the use of three machine learning techniques, logistic regression, random forests, and neural networks, for short-term traffic congestion prediction using vehicle trajectories available through connected vehicles technology. Vehicle trajectories provided by the Next Generation SIMulation (NGSIM) program were utilized in this study. Two types of predictive models were developed in this study: (1) offline models which are calibrated based on historical data and are updated (re-trained) whenever significant changes occur in the system, such as changes/updates to the infrastructure, and (2) online models which are calibrated using historical data and updated regularly using real-time information on prevailing traffic conditions obtained through V2V/V2I communications. Results show that the accuracy of the models built in this study to predict the congested traffic state can reach 97%. The models presented can be used in various potential applications including improving road safety by warning drivers of upcoming traffic slowdowns and improving mobility through integration with traffic control systems.

Assessing crash risk considering vehicle interactions with trucks using point detector data

Trucks have distinct driving characteristics in general traffic streams such as lower speeds and limitations in acceleration and deceleration. As a consequence, vehicles keep longer headways or frequently change lane when they follow a truck, which is expected to increase crash risk. This study introduces several traffic measures at the individual vehicle level to capture vehicle interactions between trucks and non-trucks and analyzed how the measures affect crash risk under different traffic conditions. The traffic measures were developed using headways obtained from Inductive Loop Detectors (ILDs). In addition, a truck detection algorithm using a Gaussian Mixture (GM) model was developed to identify trucks and to estimate truck exposure from ILD data. Using the identified vehicle types from the GM model, vehicle interaction metrics were categorized into three groups based on the combination of leading and following vehicle types. The effects of the proposed traffic measures on crash risk were modeled in two different cases of prior- and non-crash using a case-control approach utilizing a conditional logistic regression. Results showed that the vehicle interactions between the leading and following vehicle types were highly associated with crash risk, and further showed different impacts on crash risk by traffic conditions. Specifically, crashes were more likely to occur when a truck following a non-truck had shorter average headway but greater headway variance in heavy traffic while a non-truck following a truck had greater headway variance in light traffic. This study obtained meaningful conclusions that vehicle interactions involved with trucks were significantly related to the crash likelihood rather than the measures that estimate average traffic condition such as total volume or average headway of the traffic stream.

Assessing rear-end crash potential in urban locations based on vehicle-by-vehicle interactions, geometric characteristics and operational conditions

Rear-end crashes are one of the most frequently occurring crash types, especially in urban networks. An understanding of the contributing factors and their significant association with rear-end crashes is of practical importance and will help in the development of effective countermeasures. The objective of this study is to assess rear-end crash potential at a microscopic level in an urban environment, by investigating vehicle-by-vehicle interactions. To do so, several traffic parameters at the individual vehicle level have been taken into consideration, for capturing car-following characteristics and vehicle interactions, and to investigate their effect on potential rear-end crashes. In this study rear-end crash potential was estimated based on stopping distance between two consecutive vehicles, and four rear-end crash potential cases were developed. The results indicated that 66.4% of the observations were estimated as rear-end crash potentials. It was also shown that rear-end crash potential was presented when traffic flow and speed standard deviation were higher. Also, locational characteristics such as lane of travel and location in the network were found to affect drivers’ car following decisions and additionally, it was shown that speeds were lower and headways higher when Heavy Goods Vehicles lead. Finally, a model-based behavioral analysis based on Multinomial Logit regression was conducted to systematically identify the statistically significant variables in explaining rear-end risk potential. The modeling results highlighted the significance of the explanatory variables associated with rear-end crash potential, however it was shown that their effect varied among different model configurations. The outcome of the results can be of significant value for several purposes, such as real-time monitoring of risk potential, allocating enforcement units in urban networks and designing targeted proactive safety policies.

A simulation-based methodology for quantifying European passenger car fleet CO2 emissions

Common approaches to assess the evolution of CO2 emissions from road vehicles are usually based on (a) estimates of future fleet composition, where most approaches consider vehicles at a rather aggregated level, and (b) emission factors, which are either based on CO2 certification data or statistically-provided functional relationships obtained from real world test data, or a combination of the two. This approach has certain limitations in capturing the effect of new technologies on CO2 emission related policy initiatives. The present study proposes a new method for the detailed calculation of the European light duty vehicle fleet CO2 emissions, which could help to overcome such limitations, achieve better results when making CO2 emissions projections and better support future policies. Simulation at individual vehicle level is combined with fleet composition data, retrieved from the official European CO2 emissions monitoring database, and publicly available data regarding individual vehicle characteristics in order to calculate vehicle CO2 emissions and fuel consumption over different conditions and vehicle configurations. The methodology is applied to analyze and assess the impact of the introduction of the new certification procedure, the Worldwide Light duty vehicle Test Procedure (WLTP), on the European car fleet CO2 emissions. Results show an average WLTP to NEDC CO2 emissions ratio of approximately 1.2. The increases in CO2 emissions are higher for cars exhibiting lower NEDC emission values (additional 29 and 25gCO2/km for vehicles emitting 100 and 119gCO2/km, respectively). At higher emission levels (about 250CO2g/km) WLTP and NEDC present comparable results. Three possible scenarios for the translation of projected NEDC CO2 emissions to WLTP-based ones are quantified.

Changes in speed distribution: Applying aggregated safety effect models to individual vehicle speeds

This study investigated the effect of applying two aggregated models (the Power model and the Exponential model) to individual vehicle speeds instead of mean speeds. This is of particular interest when the measure introduced affects different parts of the speed distribution differently. The aim was to examine how the estimated overall risk was affected when assuming the models are valid on an individual vehicle level. Speed data from two applications of speed measurements were used in the study: an evaluation of movable speed cameras and a national evaluation of new speed limits in Sweden.The results showed that when applied on individual vehicle speed level compared with aggregated level, there was essentially no difference between these for the Power model in the case of injury accidents. However, for fatalities the difference was greater, especially for roads with new cameras where those driving fastest reduced their speed the most. For the case with new speed limits, the individual approach estimated a somewhat smaller effect, reflecting that changes in the 15th percentile (P15) were somewhat larger than changes in P85 in this case. For the Exponential model there was also a clear, although small, difference between applying the model to mean speed changes and individual vehicle speed changes when speed cameras were used. This applied both for injury accidents and fatalities. There were also larger effects for the Exponential model than for the Power model, especially for injury accidents.In conclusion, applying the Power or Exponential model to individual vehicle speeds is an alternative that provides reasonable results in relation to the original Power and Exponential models, but more research is needed to clarify the shape of the individual risk curve. It is not surprising that the impact on severe traffic crashes was larger in situations where those driving fastest reduced their speed the most. Further investigations on use of the Power and/or the Exponential model at individual vehicle level would require more data on the individual level from a range of international studies.

Modelling the Spatio-Temporal Distribution of Ambient Nitrogen Dioxide and Investigating the Effects of Public Transit Policies on Population Exposure

Estimating the future state of air quality associated with transport policies and infrastructure investments is key to the development of meaningful transportation and planning decisions. This paper describes the design of an integrated transportation and air quality modelling framework capable of simulating traffic emissions and air pollution at a refined spatio-temporal scale. For this purpose, emissions of Nitrogen Oxides (NOx) were estimated in the Greater Montreal Region at the level of individual trips and vehicles. In turn, hourly Nitrogen Dioxide (NO2) concentrations were simulated across different seasons and validated against observations. Our validation results reveal a reasonable performance of the modelling chain. The modelling system was used to evaluate the impact of an extensive regional transit improvement strategy revealing reductions in NO2 concentrations across the territory by about 3.6% compared to the base case in addition to a decrease in the frequency and severity of NO2 hot spots. This is associated with a reduction in total NOx emissions of 1.9% compared to the base case; some roads experienced reductions by more than half. Finally, a methodology for assessing individuals’ daily exposure is developed (by tracking activity locations and trajectories) and we observed a reduction of 20.8% in daily exposures compared to the base case. The large difference between reductions in the mean NO2 concentration across the study domain and the mean NO2 exposure across the sample population results from the fact that NO2 concentrations dropped largely in the areas which attract the most individuals. This exercise illustrates that evaluating the air quality impacts of transportation scenarios by solely quantifying reductions in air pollution concentrations across the study domain would lead to an underestimation of the potential health gains.

Connected cruise control: modelling, delay effects, and nonlinear behaviour

ABSTRACTConnected vehicle systems (CVS) are considered in this paper where vehicles exchange information using wireless vehicle-to-vehicle (V2V) communication. The concept of connected cruise control (CCC) is established that allows control design at the level of individual vehicles while exploiting V2V connectivity. Due to its high level of modularity the proposed design can be applied to large heterogeneous traffic systems. The dynamics of a simple CVS is analysed in detail while taking into account nonlinearities in the vehicle dynamics as well as in the controller. Time delays that arise due to intermittencies and packet drops in the communication channels are also incorporated. The results are summarised using stability charts which allow one to select control gains to maintain stability and ensure disturbance attenuation when the delay is below a critical value.

Including congestion effects in urban road traffic CO2 emissions modelling: Do Local Government Authorities have the right options?

Tailpipe emissions from vehicles on urban road networks have damaging impacts, with the problem exacerbated by the common occurrence of congestion. This article focuses on carbon dioxide because it is the largest constituent of road traffic greenhouse gas emissions. Local Government Authorities (LGAs) are typically responsible for facilitating mitigation of these emissions, and critical to this task is the ability to assess the impact of transport interventions on road traffic emissions for a whole network.This article presents a contemporary review of literature concerning road traffic data and its use by LGAs in emissions models (EMs). Emphasis on the practicalities of using data readily available to LGAs to estimate network level emissions and inform effective policy is a relatively new research area, and this article summarises achievements so far. Results of the literature review indicate that readily available data are aggregated at traffic level rather than disaggregated at individual vehicle level. Hence, a hypothesis is put forward that optimal EM complexity is one using traffic variables as inputs, allowing LGAs to capture the influence of congestion whilst avoiding the complexity of detailed EMs that estimate emissions at vehicle level.Existing methodologies for estimating network emissions based on traffic variables typically have limitations. Conclusions are that LGAs do not necessarily have the right options, and that more research in this domain is required, both to quantify accuracy and to further develop EMs that explicitly include congestion, whilst remaining within LGA resource constraints.

Integrated Relocation Model for Free-Floating Carsharing Systems

This paper presents an integrated relocation model for free-floating car-sharing (FFCS) systems. A historical data analysis conducted monthly generates demand indicators. A zone categorization identifies zones with a historical vehicle shortage or surplus and enables the estimation of the optimal vehicle distribution for the target time period. Given the current vehicle positions, the model identifies vehicle imbalances. Five macroscopic and microscopic steps using optimization and rule-based methods recommend relocations and service trips on an individual vehicle level. This process results in detailed staff operation plans. To assess the impact of relocations on the system's operation, the model was applied to an FFCS system in Munich, Germany. Three real-world field tests conducted at stages of the model's development exhibited different degrees of automation. The evaluation showed promising results. All tests had positive impacts on the main measures of success. The earnings per vehicle were increased by up to 18%. The mean idle time per trip end of all vehicles in the feet was decreased by up to 18%. The difference in idle times between the relocated vehicles and neighboring vehicles that were not relocated was decreased by up to 31%. Most important, the profit of the operator was between 4.7% and 5.8% higher than without relocations. The automation of the whole model in the last field test led to slightly lower but still very positive impacts while facilitating the whole relocation process for the operator.