PRR Project

Summary

1. TASKS TO BE ASSIGNEDThe researcher will collaborate in advanced studies on dynamic bus lane traffic risk mapping and infrastructure-oriented safety intervention management, using a mix of traditional traffic and roadway characteristics data, probe vehicle generated data, and weather and accident data. This comprises the following tasks:• Identify and select the most promising data available on the asset management system of a road operator and the operation support databases of a city bus operator.• Survey existing tools for inobtrusive in-vehicle collection of roadway and road user interaction data using probe vehicles and support their proper installation in vehicles and CANbus data extraction.• Advance road infrastructure safety management methods, broadening existing knowledge on Unprotected Road Users (URU - pedestrian and bicycle users) accident and trauma modelling, risk exposure, and evaluating alternatives for infrastructure interventions and their effects.• Identify the relevant variables to be incorporated in algorithms for categorizing road user interactions and modelling ADAS availability as related to road environment, seasonal, traffic, and weather conditions.• Exploit multivariable information to analyse complex traffic scenarios, evaluate the impact of infrastructure alternatives, and develop a holistic framework for bus lane risk assessment, follow-up risk changes over time, and issue warnings in case of adverse changes.2. SCIENTIFIC PROFILE REQUIREDTo develop the above research, LNEC requires a researcher with background in Civil Engineering, Transportation Engineering, Electrical and Computer Engineering and Human Factors. The following skills are preferred:• Experience in behavioural studies related to traffic and transportation, including road user interaction observation and analysis of accident scenarios.• Transport safety and road design: relations between infrastructure features and accident occurrence or severity, through mathematical/statistical modelling.• Knowledge of machine vision, particularly its application in ADAS vehicle technology and its interaction with road infrastructure.• Background in big data analytics and handling large datasets, with skills in applying AI and machine learning techniques crucial for traffic pattern analysis and infrastructure design.• Familiarity with GIS applications.The candidate should be motivated to develop the identified tasks and participate in joint research projects involving on-site observation of relevant traffic system parameters.3. RATIONALE TO HIRE FOR THE SCIENTIFIC AREAInjuries from traffic accidents are a major health burden in modern societies. By WHO and EU traffic safety goals, Portugal is committed to a 50% reduction in the number of fatalities and serious injuries (MAIS3+) by 2030, as referred to the 2019 numbers of occurrences, corresponding to the projected avoidance of 513 fatalities by 2030 (https://visaozero2030.pt/wp-content/uploads/FASE2-Bases_Nova_Estrategia_VisaoZero2030_Estrutura_Potenciais_Intervencoes.pdf). A comprehensive set of road safety interventions are envisaged in the national strategy VisãoZero2030, several focusing on URU, as they represent 35% of the fatalities and serious injuries.Bus public transport operators can improve the safety of their bus lanes by proactively detecting critical stretches of travelled routes and issuing warnings to concerned drivers, informing road operators and urban public space managers of these higher risk spots, and suggesting adequate infrastructure safety interventions or changing part of the routes.Proactive mapping of higher risk bus lane stretches can be obtained by fusing traffic and roadway characteristics with user interaction data provided by in-vehicle ADAS. Calibration with historical accident data will strengthen methodology’s robustness and its contribution to meeting VisaoZero2030 traffic safety goals regarding to urban areas and URU – two key areas for intervention. A higher share of public transport and safer bus routes contribute to sustainable safe travel and mobility.4. RELEVANCE OF THE SCIENTIFIC PROFILE OF THE PROPOSED POSITION IN THE CONTEXT OF THE STATE-OF-THE-ARTCurrently, area wide URU safety assessments are hindered by lack of robust exposure estimates and the (mathematically) low number of accident occurrences.Modern buses are equipped with sensors and ADAS that generate several types of information and trigger various warning events related to driving manoeuvres and road user. The output of these systems can be stored per vehicle and aggregated at the fleet level.The proposed position is intended to support identifying the most promising in-vehicle data sources and carry out the fusion of their output, developing appropriate data analytics and visualization tools to establish a more efficient, evidence based, and data driven bus route infrastructure safety management, by bus operators and municipal road administrations.