Dipartimento di Economia e Statistica "Cognetti de Martiis"

Descrizione del progetto

1. Introduction

The University of Turin (UniTo) is a large athenaeum (around 85,000 individuals) scattered among over one hundred locations in the metropolitan area. The residential area covered by its community is also very extensive, since the Piedmont region has only one additional university center (UPO), smaller and offering a more limited range of fields of study. In 2018/19, 41% of UniTo students came from outside the province of Turin, while 34% were domiciled in the province but outside the capital.

UniTo aims to improve its overall environmental sustainability, within the framework of the Italian University Network for Sustainable Development (RUS), and in 2016 created a specific branch, the Green Office, dealing with issues concerning energy, food, public procurement, waste, climate change and mobility. Its home city of Turin, moreover, is a critical context, lying in the Po valley, one of the most polluted areas in Europe (Horálek et al., 2016), and is among the worst cities in Italy in terms of air quality, with a higher car modal share than its neighbors Milan and Genoa (ISFORT, 2021).

Due to its sheer size, the impact of daily home-university journeys at UniTo strongly contributes to shaping the mobility of Turin and the metropolitan area, with significant repercussions on road congestion, air pollution and use of public space. One of the main problems concerns the parking of arriving private cars. While UniTo has no direct competence on the management of the road network and public transport services, it does have its own parking spaces whose management can potentially be used to implement policies to discourage the use of private cars when there are efficient transport alternatives. The demand for internal parking spaces is also growing due to the progressive extension of public areas with parking fees, which make parking inside university structures more attractive. In this context, the SUSTAIN project (Specific User Sustainability Through Accurate Index Number) aims to encourage sustainable mobility choices in a targeted and innovative way, leveraging the allocation of parking permits for internal parking spaces. Obviously, the availability of a safe and cheap parking space is a significant incentive to choose the car as unimodal means of transport, especially in high-density urban contexts.

2. The new “Luigi Einaudi” Campus: a case study

In 2012, the new "Luigi Einaudi" Campus (CLE) was activated; it is today one of the main seat of the University aggregating many thousands of people. To comply with current legislation, CLE includes a large car park (500 units), to avoid overcrowding the neighborhood with parked cars of people coming to the campus. This remained largely empty for years (also due to the monthly fee of € 10 - low but not free); things changed considerably in 2017 with the introduction of parking fees (more than 6 times the cost of the internal one) in the surrounding area. It became therefore necessary to define how to manage the allocation of permits selectively, based on criteria linked to sustainability as far as possible. This case is not only relevant in itself, as the creation of already planned new structures with attached parking will create similar contexts, as will do the reorganization of the parking spaces of existing offices, now managed with heterogeneous and often inconsistent rules.

Finally, sustainable innovation regarding car parking management has a strong potential outside university as well, as it could be adopted and used for the management of public - or even private - spaces in contexts where this is a scarce resource, such as dense metropolitan ones.

3. Project SUSTAIN

Starting from these premises, the project aims to establish innovative and rational criteria to promote sustainable mobility in the university, recognizing the diversified mobility needs of individuals, with heterogeneous levels of actually achievable sustainability. The best results will derive from a combination of choices and opportunities based on the use of all available data, optimizing overall sustainability without excessively penalizing less eco-friendly travel arrangements in the most difficult cases. The guidelines of the project are therefore:

• Equity: allow access to parking to those who have a real need to travel by car, discouraging others;

• Efficiency: obtain the needed information in a simple, fast and convenient way from a web platform;

• Extensibility: define and implement criteria that can be easily extended to different contexts.

The goal is therefore to create a system of disincentives to use the car "door to door" based on a sustainable accessibility index that involves all the relevant information (the origin-destination matrix and the travel options available) and can be obtained using web resources largely available.

For any origin-destination, the index will summarize the comparative advantage of traveling by car compared to the best multimodal solution combining active mobility, public transport and sharing. It will be computed by a customized version of the existing open source routing engine (OpenTripPlanner) that the Piedmont Region uses for www.muoversinpiemonte.it, its info-mobility platform, that delivers the best public transport solutions available for any origin-destination.

The algorithm should not be limited to comparing travel times, involving all features of the sustainable travel options in term of comfort and commitment: e.g. the length and mode of the first and last legs of the journey as well as the number and duration of intermediate connections with respect to a reference "ideal" connection time (e.g. too risky if less 10 minutes, too long if more than 15 minutes). In addition, the frequency during the whole day of similar solutions could be considered, as well as their average characterization such as speed. In conclusion, the index will therefore represent a measure of the overall comparative quality of the journey from home to university with the least possible use of cars, versus one made entirely with one's own car.

Parking permits will normally be awarded to those with a low value of the index (i.e. a low level of sustainable accessibility), whose car journey is significantly faster and more comfortable than the best sustainable option. Those who, on the other hand, can travel by active mobility plus public transport as fast - or even faster - than by car, will generally not be granted the permit. Obviously, the lack of parking permit does not imply that you cannot come to the campus by car; however, it does entails some disincentives: paying the higher fees of public parking lots, a possibly bigger distance between the car park location and the campus, as well as a longer time to find a place for your car.

4. Customizing the routing engine

To fulfill the needs involved by the computation of the sustainable accessibility index, the routing engine should provide uni-modal or multi-modal travel solutions between any two points in the region combining walking, cycling, rail, bus/tramway/metro services, or even limited car use: in some cases a fully sustainable travel option may not exist, but allowing car use for the first 5-10 kilometers as the first leg could allow you to connect to the public transport network and complete the journey in a sustainable way (as long car is limited to the first leg, no car parking at university is required).

It is also necessary to allow calibration of relevant parameters, such as the maximum distance covered on foot, by bike and by car, and the waiting time - as well as the distance to be covered on foot - for connections between public transport services. The number of travel solutions provided by the engine, referred to a user defined arrival time, should be adjustable if more than one is available.

For each proposed path, the output will provide the following data: total estimated travel time; number of sections; mode used in each section; connection times; length of active mobility stretches.

In addition, it will produce an assessment of how frequent are similar solutions throughout the day, and the average values of the same parameters for all solutions throughout the day.

The customized routing engine will interact with users applying for a parking permit through a web-based layer that will allow them to precisely specify their home location on the map; it will also perform the geo-referencing of the point in space, providing the input for the routing engine.

5. Computation and calibration of the sustainable accessibility index

The final version of the index will involve the exact timings for both travel alternatives, and the customization of the routing engine should avoid cases where no solutions exists due to a too long first leg: in such instances the sustainable solution will include an initial car leg, ending at the nearest public transport node. The index definition however will be more complex: moving from the sole information on journey times to the whole array of data listed in Sec. 4 implies combining heterogeneous values, with no natural metric, to be weighted subjectively. Also, its extreme values (i.e. maximum and minimum sustainable accessibility) are undefined. Some degree of subjectivity is even implied in the routing engine customization: for example, on a certain journey a long first leg by car could result in shorter times when compared with a solution where a short car leg is in itself more sustainable, but involves much longer travel times: where and how do you set the limit - if any - for the car part of the journey when evaluating the most sustainable solution?

After having defined a reasonable weighting system, to establish the final metric and set the extreme values a preliminary calibration process will be performed, using a pilot survey on a small sample (some 200 cases), starting with an arbitrary metric, checking the actual extreme values for the sample, as well as the variability induced by the involved factors, possibly adjusting their relative weights, and finally rescaling the metric so that extreme values are close to a normalized (e.g. 0 to 1) range. The chosen metric will have to be examined also with regard to its discriminating power: an index ranging from zero to one, but with actual values in the population concentrated between 0.4 and 0.6 in 95% of the cases will not help in distinguishing significantly most part of the population.

6. Further considerations

One of the strengths of the project lies in its scalability: the initial reference to the Luigi Einaudi Campus is in fact coincidental, as the use of the accessibility index can be extended to other branches where UniTO owns parking spaces operated in makeshift ways lacking in terms of equity and/or efficiency. In addition, the involvement of local authorities (both the Metropolitan City of Turin and the Piedmont Region are partners) in the project reflects the interest in the definition of a tool that can be shared on an even larger scale in any context in which public parking spaces have to be managed. In fact, this is a tool that, combined with other criteria deemed suitable in the specific case, allows for the inclusion of sustainability in an effective and targeted way in the governance of mobility, also helping local authorities that have responsibility for the actual governance of public transport systems and mobility in general to identify the priority framework in transport policies.

A further potential use of the technology lies in an assessment of accessibility that is no longer individual (how easy it is for the individual to reach a given location), but structural, i.e. how much is a given location reachable in a sustainable, but also efficient, way by the whole community that gravitates around it. In fact, having computed the accessibility index for the members of such community - or for a representative sample - the average value of the individual indices (possibly weighted in relation to various possible factors, such as periodic frequency and travel times), it will constitute a highly realistic parameter of accessibility of the structure due to it being based on the situation of the actual community that converges there, rather than on a generic evaluation of the transport services that serve it. In the same line, other statistical indicators regarding the distribution of the index values among members of the same community could offer further insights on the sustainable accessibility of the location: the standard deviation, for example, would suggest how heterogeneous - or, on the contrary, how similar - its level is in the community it serves.

1. ISFORT (2021), 18° Rapporto sulla mobilità degli italiani - Governare le transizioni per una ripresa sostenibile, Istituto Superiore di Formazione e Ricerca per i Trasporti, Roma

2. Horálek J., Schreiberová M., de Leeuw F., Kurfürst P., de Smet P., Schovánková J., (2016), European air quality maps for 2016, European Environment Agency