The TANGENT project started one year ago. Let’s take this opportunity to look back at three key research deliverables published since the beginning of the project.
D3.1 Travel behaviour: state-of-the-art, current and future mobility patterns
This deliverable aims to identify factors affecting travel decision-making through a critical academic review. It looks at three major axes:
First, it focuses on approaches to collect data on travellers’ choices
The review reveals that travel data collection is most representative and accurate when it combines stated preferences (declared mobility patterns, preferences between different modes, etc.) and real preferences (real travel patterns of users recorded usually through location-based applications).
Second, it looks at state-of-the-art approaches to analyse and model travel decisions
From a modelling perspective, the literature review highlighted that Machine Learning techniques have been widely used during the last decade to analyse transportation sector and travel behaviour. In general, predicting accuracy is better when feeding the model with new and unseen data compared to simple MultiNomial Logit (MNL) models. However, classic MNL models maintain their importance against Machine Learning methods due to the concept of interpretability.
Third, it looks at factors that may affect the decision-making process of travellers with a special emphasis on infrastructure- and system-related factors (e.g., the introduction of new services and the occurrence of network disruptions).
For this analysis, three major categories are considered: user-related, trip-related and service-related factors, together with the various choices travellers make before and during a journey. Specifically, three choice dimensions were considered: (i) travel mode (e.g., car, public transport, bicycle), (ii) departure time, and (iii) route choice.
Finally, the review looks at travel behaviour changes due to system-level disruptions. These can be caused by altered by extreme disruptions, the introduction of new services, Transportation Management Strategies measures and the circumstances where non-recurrent events occur (i.e., hazardous events and network disruptions).
Based on the findings and TANGENT’s goals, the foundations for the mobility data collection plan were set. The plan will help efficiently design the data collection process which will include a questionnaire survey and trajectory gathering through Google Timeline, therefore using both stated preferences and revealed preferences data.
D4.1 Report on the relevant state-of-the-art approaches for traffic predictions and simulations
The goal of this project chapter (Work Package 4) is to develop a framework for real-time traffic monitoring and forecasting under various circumstances (e.g. large/sport event, roadworks, accidents) while incorporating novel traffic sensing technologies from smart infrastructure and sensors. It will extend the state-of-the-art regarding traffic forecasting approaches with a focus on modern and future mobility.
To do so, this deliverable explores the state-of-the-art prediction methodologies, such as data-driven and simulation-based estimation and prediction approaches, both on network supply and demand sides. The focus is deep learning approaches for efficient numerical representations of traffic networks, data availability and granularity, new data sources and graph theory. The analysis is widened with aspects such as transferability and generalizability of the different methods.
The goal of the deliverable is to have a thorough analysis of existing approaches in order to set up the requirements and limitations for the future network flow predictions. It provides an overview of what has been done in the field of traffic state predictions and simulations and what should be further explored. The results of this exploration will provide the base for the development of the techniques for the remaining tasks including the development of traffic supply and demand predictions, the identification of critical conditions and congestion duration prediction and more.
D5.1. Analysis of Current Approaches in Optimization of Transport Network Management
This deliverable analyses scientific literature related to current approaches to transport network management optimization within a multi-actor setting. It focuses on optimizations models and techniques of Signal Vehicle Couple Control with Connected and Automated Vehicles (CAVs), Synchronization of shared and on-demand mobility with transit modes, and Dynamic Congestion Pricing.
Based on this literature review and the case studies’ priorities, the deliverable describes and justifies why TANGENT focuses on the following optimization problems in transport network management:
- Coupled traffic signal and route planning optimization for CAVs
- Optimization of integration DRT systems with public transit modes
- Synchronization of public transport and Traffic control
- Optimization of Dynamic Congestion Pricing schemes
Going further, D5.1 reviews literature related to negotiation and arbitration models for transport network management. Specifically, it focuses on integrated decision-making to integrate a variety of stakeholders objectives, needs and priorities. Finally, the deliverable provides practical guidelines that delineate the development of a consensus-reaching mechanism, discussing its objective, scope, domain of application, stakeholders, data needs and modelling approach, and how they will be applied in the context of the TANGENT project.