The i-DREAMS project focuses on establishing a framework known as the ‘Safety Tolerance Zone (STZ)’ to ensure drivers operate within safe boundaries. This study compares Long-Short-Term-Memory Networks and shallow Neural Networks to assess participants’ safety levels during i-DREAMS on-road trials. Thirty German drivers’ trips and Forty-Three Belgian drivers were analyzed using these methods, revealing factors contributing to risky behavior. Results indicate i-DREAMS interventions significantly enhance driving behavior, with Neural Networks displaying superior performance among the algorithms considered.

The i-DREAMS project has a core objective: to establish a comprehensive framework that defines, develops, and validates a context-aware ‘Safety Tolerance Zone’ (STZ). This zone is crucial for maintaining drivers within safe operational boundaries. The primary focus of this research is to conduct a detailed comparison between two machine learning approaches: long short-term memory networks and shallow neural networks. The goal is to evaluate the safety levels of participants as they engage in natural driving experiences within the i-DREAMS on-road field trials. To accomplish this objective, the study gathered a series of trips from a sample group consisting of 30 German drivers, 43 Belgian drivers, and 26 drivers from the United Kingdom. These trips were then input into the aforementioned machine learning methods to reveal the factors contributing to unsafe driving behaviour across various experiment stages. The results obtained highlight the significant positive impact of i-DREAMS’ real-time interventions and post-trip assessments on enhancing driving behaviour. Furthermore, it is worth noting that neural networks demonstrated superior performance compared to other algorithms considered within this research context.

The i-DREAMS project established a ‘Safety Tolerance Zone (STZ)’ to maintain operators within safe boundaries through real-time and post-trip interventions, based on the crucial role of the human element in driving behavior. This paper aims to model the inter-relationship among driving task complexity, operator and vehicle coping capacity, and crash risk. Towards that aim, data from 80 drivers, who participated in a naturalistic driving experiment carried out in three countries (i.e., Belgium, Germany, and Portugal), resulting in a dataset of approximately 19,000 trips were collected and analyzed. The exploratory analysis included the development of Generalized Linear Models (GLMs) and the choice of the most appropriate variables associated with the latent variables “task complexity” and “coping capacity” that are to be estimated from the various indicators. In addition, Structural Equation Models (SEMs) were used to explore how the model variables were interrelated, allowing for both direct and indirect relationships to be modeled. Comparisons on the performance of such models, as well as a discussion on behaviors and driving patterns across different countries and transport modes, were also provided. The findings revealed a positive relationship between task complexity and coping capacity, indicating that as the difficulty of the driving task increased, the driver’s coping capacity increased accordingly, (i.e., higher ability to manage and adapt to the challenges posed by more complex tasks). The integrated treatment of task complexity, coping capacity, and risk can improve the behavior and safety of all travelers, through the unobtrusive and seamless monitoring of behavior. Thus, authorities should utilize a data system oriented towards collecting key driving insights on population level to plan mobility and safety interventions, develop incentives for road users, optimize enforcement, and enhance community building for safe traveling.

While mobility and safety of drivers are challenged by behavioral changes, the increasingly complex road environment has placed a higher demand on their adaptability. The ultimate goal of this paper was to identify the impact that the balance between task complexity and coping capacity had on crash risk. Towards that aim, an integrated model for understanding the effect of the inter-relationship of task complexity and coping capacity with risk was developed. A vast library of data from a naturalistic driving experiment was created in three countries (i.e., Belgium, UK and Germany) to investigate the most prominent driving behavior indicators available, including speeding, headway, overtaking, duration, distance and harsh events. In order to fulfil the aforementioned objectives, exploratory analysis, such as Generalized Linear Models (GLMs) were developed, and the most appropriate variables associated to the latent variable “task complexity” and “coping capacity” were estimated from the various indicators. Additionally, Structural Equation Models (SEMs) were used to explore how the model variables were inter-related, allowing for both direct and indirect relationships to be modelled. The analyses revealed that higher task complexity levels lead to higher coping capacity by drivers. Additionally, the effect of task complexity on risk was greater than the impact of coping capacity in Belgium and Germany, while mixed results were observed in the UK.

Urban centers are trying to achieve safer and more sustainable mobility but understanding pedestrian behavior at crossings remains a key challenge. In cities where dense traffic and diverse road users intersect, like in the case of Athens, illegal pedestrian crossings are a persistent and hazardous phenomenon. This study explores the issue of pedestrian non-compliance, referring to the red light and jaywalking violations, at designated signalized crossings, in the center of Athens. A Stated Preference (SP) survey was structured in order to examine decision-making scenarios based on varying conditions of travel time, perceived crash risk, and personal comfort. A dataset of 212 responses was collected and analyzed by using binary logistic regression to examine the likelihood of illegal crossing, and generalized linear models to investigate the trade-offs pedestrians make between safety, time, and convenience, given three different scenarios of camera use, a traffic light with a timer, and the baseline scenario to choose from. The models highlight the influence of time pressure, traffic conditions, and social components on individual decisions. The results aim to showcase the importance of pedestrian decisions in preventing road crashes and, in this way, support more behavior-aware policy design, ultimately contributing to safer urban mobility systems.