At the beginning of the summer, we organised three Advisory Board (AB) meetings to gather expertise and advice on specific questions linked to our technical activities.
Session 1 – Advancing Mode Choice Predictions: Exploring State-of-the-Art Methods and Emerging Data Sources
In TANGENT, NTUA is working on identifying factors affecting travel decision-making, mobility behavioural shifts linked to unexpected events, and more widely travel choices. To do so, they are developing a set of models to describe travel choices and providing recommendations to incorporate behavioural models in traffic management systems. These activities are based on the collection of data on users’ stated preferences (surveys) and revealed preferences (google timeline data) in all four TANGENT pilots, namely Lisbon, Rennes, Greater Manchester and Athens.
Key insights from the discussion with AB:
– Mode choice models are essential to analyse mobility behaviours. They can also be helpful when performed in a dynamic way to analyse disruptions or mode changes for a particular incident.
– Collecting reliable, transparent and relevant data is necessary for smart traffic management and well-synchronised public transport.
Session 2 – Advancing Mobility Strategies: State-of-the-Art Traffic Forecasting and Generation of Transport Policy and Response Plans Using Automatic Design Concept
This session was divided into two sub-sections:
1) Transport Prediction and Simulation Models
The first topic was led by IMEC and AIMSUN. They aim to create a framework architecture to analyse the interoperability of demand and supply models using simulations. This framework will be created with OD matrices, data collection and preparation for demand and supply.
Key insights from the discussion:
– The importance of safety and accessibility is to be considered as essential in Demand Responsive Transport and transit modes as well as intermodality linking public transport with rural areas.
– In regard to the synchronisation of public transport and traffic control, the emission and costs should be included as performance indicators.
– Adjustments like changes in frequencies, phase adjustments, and changes in time are options that are being considered to synchronise public transport in the TANGENT project.
– The unpredictability of human behaviour and understanding how it influences transport models even with state-of-the-art road networks, known traffic flow etc. is a major concern.
– Governance of cities for transport infrastructure should base on a simulation model to have a homogenous transport model.
2) Transport policy and response plans using COD
The second topic was led by Deusto and Aimsun. The project seeks to create a generation of transport policy and response plans using Computer Optimized Design (COD) which takes into consideration the current and future challenges in the transport sector. Indeed, decision-makers need new tools to manage the increasingly complex and multimodal sector. In regard to CODs, for example, the project is developing a Consensus Reaching Mechanism based on preference matrices. Other use cases linked to COD in the project include dynamic congestion pricing, the integration of demand-responsive transport and transit modes, the synchronisation of public transport and traffic control, and signal vehicle coupled control with CAVs.
Key insights from the discussion focused on Dynamic Congestion Pricing:
– Dynamic Congestion Pricing should be tailored to the passengers and adapted to the local context.
– Other measures need to be included to optimise DCP such as reducing public transport fares and increasing toll prices.
Session 3 – Unlocking Data Collaboration and Assessing Smart Infrastructure Indexes
This session was divided into two sub-sections:
1) Fostering a data-sharing culture
The first topic was led by Cefriel and aims to foster strong data-sharing cultures. To create TANGENT’s solutions to optimise traffic management, data is required coming from intermodal mobility, users and sensors. However, collected data is often heterogeneous in formats which requires harmonisation through data conversion techniques. The project created a TANGENT Data Catalogue, a digital platform supporting the sharing, findability, and accessibility of the distributed digital assets needed by the project.
Key insights from the discussion focused on barriers for data sharing:
– Identified barriers include data sharing being perceived as a mandate, costs associated with creating and sharing data, security concerns with data sharing, data sovereignty is often not ensured, fears of misuse of data, ensuring data quality and liability, etc.
– Solutions were discussed including standardising mechanisms, data quality metrics, ensuring documentation, etc.
2) Smart Infrastructure Classification
The second topic was led by A-to-Be and focused on the Smart Infrastructure Classification. Automated driving (AD) implies that the vehicle control gradually moves from human perception and control to partial or full control by computer systems. Therefore, AD requires certain prerequisites. TANGENT will focus on developing an integrated framework for supporting transport, operational management and strategic decision-making where different transport data sets, AI techniques, analytical models and simulation tools are integrated to provide real-time and forecast in dashboards to different stakeholders.
Key insights from the discussion on the relevancy of classifying infrastructures to facilitate AD:
– Classifying infrastructure to provide safety and security to the vehicle does not provide any advantage when the vehicle does not want to use it or wants to be fully integrated rather it should be about systems controlled by the vehicles. In addition, the level of service provided by the infrastructure depends on the vehicles and not on the category of automation.
– Traffic management should also influence the operation of the vehicles in the network and a corresponding level of service should be provided.
– Care should be taken to involve C-ITS more with autonomous vehicles.