Evaluating and promoting sustainability of micro-shared mobility system leveraging big data and machine learning

Shared micro-mobility systems (SMMS), such as bike/e-bike and e-scooter/e-mopeds sharing, are potential sustainable alternatives to fossil-powered transportation and are being embraced by Swedish cities to promote sustainable mobility, especially during the COVID-19 pandemic. SMMS, as new drivers of sustainable mobility, are operating in over 100 EU cities. For urban planners and policymakers, the main attraction of SMMS is promoting active mobility for reducing energy consumption and greenhouse gases (GHG) emissions (called environmental benefits). For users, the benefits of SMMS are reducing travel expenses (e.g., by reducing travel time, costs and increasing convenience). Despite many qualitative discussions about the potential benefits of SMMS, quantitative assessments about societal effects (i.e., environmental influences and user benefits) and keys to promoting sustainability of SMMS in different urban contexts are lacking. Urban policymaker, planner and SMMS operators need accurate assessments of SMMS to support cost-benefit analysis and policymaking in different urban contexts. Meanwhile, identifying key drivers of promoting usages and sustainable benefits of SMMS are crucial foundations for scientific planning, operational optimizations, and countermeasures.

Emerging big-data resources and machine learning methods provide great potential solutions for the above-mentioned challenges.The research team obtains operation data from SMMS (including bike, e-bike, e-scooter, and e-mopeds) over 50 cities around the world such as EU (e.g., Sweden, UK, Germany, Italy, France, Netherlands), USA, China, Singapore, and Australia. Based on the massive SMMS data, the main objectives of this proposal include:

§ O1: Develop a methodological framework for high-resolution assessments of the societal effects of SMMS, leveraging big data. This objective aims to propose a novel approach at the trip level to quantitatively estimate the environmental influence and user benefit derived from SMMS in different urban contexts with a high level of accuracy considering life-cycle assessments and leveraging big data.

§ O2: Identify and model the determinants for improving usages and societal benefits of SMMS based on an international meta-analysis. By capitalizing on the available data from over 50 cities, this objective aims to comprehensively decipher the determinants (both objective and subjective factors) and efficient countermeasures for promoting SMMS usage and sustainability in different urban and cultural contexts.

§ O3: Develop a spatiotemporal short-term demand prediction model for SMMS using deep learning algorithms. The focus is to accurately predict the short-term spatiotemporal demand of SMMS for operational optimizations of SMMS to realize dynamic demand-supply matchups and promote utilization rates.

 

The project is funded by Area of Advance Transport for two years with a total amount of 2.8 million SEK. The project is an interdisciplinary collaboration between three departments: ACE Chalmers (Kun Gao, mobility analysis using big data; Jiaming Wu, machine learning and big data; Xiaobo Qu, transport planning and optimization), M2 Chalmers (Jelena Andric and Majid Astaneh, life-cycle assessment and machine learning) and Department of Psychology University of Gothenburg (Lars-Olof Johansson, policymaking and social science).

Samarbetande organisationer

  • Göteborgs universitet (Akademisk, Sweden)
  • Göteborgs universitet (Utgivare, Sweden)
Startdatum 2022-01-01
Slutdatum 2023-12-31

Sidansvarig Publicerad: sö 25 jul 2021.