Machine Learning Approach to Quantity Management for Long-Term Sustainable Development of Dockless Public Bike: Case of Shenzhen in China
Bibliographic record
Abstract
Since the number of bicycles is critical to the sustainable development of dockless PBS, this research practiced the introduction of a machine learning approach to quantity management using OFO bike operation data in Shenzhen. First, two clustering algorithms were used to identify the bicycle gathering area, and the available bike number and coefficient of available bike number variation were analyzed in each bicycle gathering area’s type. Second, five classification algorithms were compared in the accuracy of distinguishing the type of bicycle gathering areas using 25 impact factors. Finally, the application of the knowledge gained from the existing dockless bicycle operation data to guide the number planning and management of public bicycles was explored. We found the following. (1) There were 492 OFO bicycle gathering areas that can be divided into four types: high inefficient, normal inefficient, high efficient, and normal efficient. The high inefficient and normal inefficient areas gathered about 110,000 bicycles with low usage. (2) More types of bicycle gathering area will affect the accuracy of the classification algorithm. The random forest classification had the best performance in identifying bicycle gathering area types in five classification algorithms with an accuracy of more than 75%. (3) There were obvious differences in the characteristics of 25 impact factors in four types of bicycle gathering areas. It is feasible to use these factors to predict area type to optimize the number of available bicycles, reduce operating costs, and improve utilization efficiency. This work helps operators and government understand the characteristics of dockless PBS and contributes to promoting long-term sustainable development of the system through a machine learning approach.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".