Abstract

Sharing mobility has become one of the emerging transport modes, but seldom is its planning entrusted to private operators and not seen as a major vector of urban mobility. 

The vehicle allocation greatly influences the benefits of the whole system, since it is linked to the potential accessibility and the opportunity to substitute and complement other less sustainable modes.

In this framework, the role of planning is paramount, and the operational design behind a free-flow system has the opportunity to be assessed at the intersection between the utilization patterns of the vehicles and the characteristics of the urban environment.

Project Description & Overview

The understanding on how the introduction of sustainable and shared vehicles has an impact on the overall movement pattern of a city, is still an open conversation. Both temporal and spatial phenomena are influencing the carsharing demand, time-invariant factors, such population and workplace density, and time-variant factors, like public transport offer and points of interest, need to be taken into consideration and analyzed in relation to the recorded utilization pattern of the vehicles.

At the same time, the availability of vehicles in specific places, at specific times, is shaping the potential accessibility and the impact of these services. Re-allocation strategies are implemented to distribute vehicles according to the expected demand, and this process has a logistic cost of maintenance, recharging and moving the vehicles around the city, ultimately allowing the access to the system in relation to the effectiveness of the vehicles allocation strategy.

This process has the potentiality to be streamlined and refined by looking at the relation between the city, with its site-specific defining characteristics, and the service, with patterns that can be analyzed and predicted. The case of Zity (Car Sharing Mobility Services Italy Srl), a newly introduced full-electric free-flow system in Milan, is the perfect example to face this problem by developing a Vehicle Allocation Model (V.A.M.) to be integrated in the management system of the service. The objective is to maximize the potential accessibility while minimizing the logistics, ultimately improving the effectiveness of the sharing modal shift and the electrification of urban mobility.

Datasets

Trip data of the selected free-flow carsharing system (Zity) of an extended period of time (3 months or more) with information on: i) starting and ending points, with timestamp; ii) intermediate stop points (standby parking mode), with timestamp; iii) persistent ID for each user. Additional data will focus on: iv) demographics (census and workforce statistical geodata); v) POI location and typologies; vi) other relevant open-data able to describe the city of Milan.

Competencies

• Quantitative research and analysis
• Collection, processing, filtering and analysis of location-based datasets and statistical datasets
• Advanced knowledge of GIS platform of choice (QGIS or ArcGIS, GeoDa)
• Python data science kit (pandas, geopandas, numpy, etc.)
• Knowledge about data visualization techniques

Learning Outcomes & Deliverables

• Data visualization about input trip data datasets – static or interactive web visualization 
• Vehicle Allocation Model (V.A.M.) developed in GIS platform

Students

Zhuohao Deng, Hanbyul Jo, Azaria Laras, and Junru Song