The dramatic COVID-19 epidemic is highlighting how health aspects are crucial for people's lives. It is also showing as controlling mobility is perhaps equally relevant for preventing the infection of other people and avoiding an exponential spread of the virus.
In epidemic conditions, monitoring people's mobility is fundamental to prevent the spread of the virus; however, it is equally important to analyze the mobility patterns that are profoundly modified in the spatial and temporal distribution, in the choice of the transport mode and the mobility behavior within the mode. Many operators are making available extensive sets of data on the people's mobility, and so enable systematic analyses of people's behavior and transport system performances.
However, the pandemic outbreak is imposing strict reductions to the capacity of the Public Transport to comply with the social distancing requirements. Such a radical change of paradigm requires studying and developing new methods to control the demand for public transport that guarantee accessibility, reduce delays, and ensure safe conditions by limiting the risk of contagion.
Indeed, the consequent substantial reduction of patronage is critical for the financial sustainability of the public transport companies. Thus, it is fundamental that public transport provides users perception of a safe, modern, and efficient system, along with other policy measures aimed at shaving the peaks of demand for transport.
The specific objective of the proposal is to:
- Analyze the effects of the lockdown process and the subsequent reopening stage on the people's mobility and the transport system performances, exploiting all available data.
- Design a monitoring system based on an individual telecommunication system, based, e.g., on Bluetooth.
- Design a control system for public transport services, able to ensure the social distancing of passengers that will be required in the different phases of pandemic spread.
The pandemic outbreak poses unprecedented issues on people's mobility and transport system control that are faced for the first time this century. The literature on these issues is rapidly arising but is still at a seminal level.
Planning, monitoring, and control of mobility and transport systems in the post-COVID era require solving many issues that will be faced in the research project and identify research topics.
New problems arise for public transport, which in regular times was not affected by congestion effect unless particular and limited cases.
ANALYSIS OF MOBILITY IN SANITARY EMERGENCY CONDITIONS
As the COVID pandemic outbreak spread out, the major data providers are supplying reports that illustrate data analytics on people's mobility in most Countries in the world. Still, they do not present the higher resolution level required to analyze the mobility within cities. The research project focuses on the urban level and will provide an analysis of mobility data at the traffic zoning level. This is possible thanks to the cooperation agreement with the Rome Mobility Agency, aimed at studying the application of decision support systems to predict the impacts of future actions on the post-COVID mobility and control the mobility patterns on the transport network. This cooperation agreement will also enable to test ad-hoc telecommunication networks and specific control strategies for the transit system.
PUBLIC TRANSPORT CONTROL PROBLEM
From the operational point of view, a control system that optimizes the control of queues to access at the stations of the transit line is absolutely new.
It can exploit the modeling structure of freeway ramp metering systems, with some differences. Transit is not affected by a real congestion phenomenon, because the travel time does not increase when the social distancing is violated. However, the problem has a peculiar complexity due to equity issues and needs to favor some users depending on their origins and destinations, as well as the lengths of their trip.
The control design problem will be first faced for a metro network, which is a close system, and so it is easier to be formulated and implemented. Also, possible extensions of the problem to bus transit networks will be considered, which have a large number of interconnections and require to formulate the problem by following a network approach.
AD-HOC TELECOMMUNICATION NETWORK DESIGN
The project is also relevant for telecommunication aspects. The first issue is related to the choice of the most appropriate technology for monitoring transit users in the metro system and on the other hand be suitable for an extension to bus networks. The required features of this system will be to work in an underground environment, to be so cheap to be available and used by as larger as possible fraction of users. An innovative result with respect to the state-of-art will be the design of multi-hop protocols that allow the development of a mesh network specifically conceived for monitoring the transit users in the presence of different scenarios of social distancing.
TRANSPORT DEMAND ESTIMATION
Transport system modeling requires methods to estimate dynamic origin-destination matrices exploiting many data sources: turnstiles at the metro station entrance, individual signals provided by the ad hoc telecommunication network (e.g. Bluetooth), socio-economic data based on land use representation, needed to define specific scenarios complying with demand control policies.
This is the case, for example, of scenarios corresponding to partial lockdown policies that ban some specific activities in limited areas of the town or change the times for different kinds of activities (shops, schools, offices), that encourage or reduce the fraction of smart working as well as the effect of general conditions that affect the modal shift toward other transport modes.
The time-dependent estimation of the Origin-Destination matrix is a largely debated problem in the scientific literature. It was already dealt with by the applicant team regarding the road traffic, where many monitoring stations count the traffic on road links. In the proposed research, the time-dependent origin-destination estimation problem will be extended to the new case of a capacity-constrained transit system with possible queues at stations, which imply both a lag between the arrivals and entrance of the users in the system and possible transit line changes.
This problem was never studied in the past.