Use information and prediction to reduce disorderly parking

How to cite

Zhan, Zilin; Wang, Yacan; Su, Duan; Yang, Nan; He, Haoran. 2021. "Use information and prediction to reduce disorderly parking." AEA RCT Registry. AEARCTR-0007952. https://doi.org/10.1257/rct.7952-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Zilin Zhan* , Yacan Wang , Duan Su , Nan Yang , Haoran He

* Primary investigator

Registry ID
AEARCTR-0007952
DOI
https://doi.org/10.1257/rct.7952-1.0
First registered
Last updated
Status
On going
Trial dates
2021-04-28 – 2021-10-28
Countries
Private
JEL codes
Public economics, Transportation economics
Secondary IDs
National Natural Science Foundation of China and the Joint Programming Initiative Urban Europe, National Natural Science Foundation of China
Registry record
http://www.socialscienceregistry.org/trials/7952

Abstract

Dockless bike-sharing is popular and provides the commuting service for "the last mile of trip". Users often make improper parking and influence the public environment and bicycle circularity. Even some parking locations are already very crowded, users still park shared bicycles here. The company needs to take manpower to carry these bicycles. In our experiment, we use machine learning to predict the destination of users' trips. If users will ride to very busy areas, we send information through app to remind them to avoid parking bicycles here.

Intervention

Reminder: Use app information to remind users to reduce parking in crowded areas.
Social norm information: Use app information to emphasize the proportion of users who parking orderly.
Altruism information: Use app information to emphasize the positive externality that parking orderly.
Social norm and altruism information: Use app information to emphasize the proportion of users who parking orderly and the positive externality that parking orderly.

Experimental design

During treatment period, when a subject unlock a shared bicycle mobile, the company will predict the destination of his or her riding. If we forecast that the subject will ride to a very crowded area, the app sends treatment information to them.

Design details

Randomization method
Completed randomization
Randomization unit
times
Clusters
One city
Observations
Expected over 5000 bike-sharing users
Arms
1000 users for control, 1000 users for reminder group, 1000 users for social norm group, 1000 users for altruism group, 1000 users for Social norm and altruism group.
Primary outcomes
Whether to park bicycles in crowded areas

Cited by

No citation data available for this registration yet.

Keywords