Pick-up point recommendation strategy based on user incentive mechanism.

Area of Science:

• Spatial Crowdsourcing
• Transportation Network Services
• Operations Research

Background:

• Online car-hailing is a key application of spatial crowdsourcing.
• Existing methods focus on task allocation but neglect optimal pick-up point selection.
• Real-world scenarios require considering user, worker, pick-up, and destination coordinates.

Purpose of the Study:

• To design an effective pick-up point recommendation strategy for online car-hailing.
• To address the limitations of existing spatial crowdsourcing models by incorporating a four-dimensional approach.
• To optimize task allocation by incentivizing users to reach suitable pick-up locations.

Main Methods:

• Developed a novel four-dimensional crowdsourcing model.
• Designed a user incentive mechanism focused on cost optimization and encouraging walking to pick-up points.
• Introduced the concept of 'forward rate' to decrease computation time.
• Utilized maximum user walking distance and task cost as key recommendation factors.

Main Results:

• The proposed strategy effectively recommends pick-up points.
• Demonstrated improvements in driver efficiency.
• Showcased a reduction in overall order trip costs.
• Validated the model's practicality in real-world online car-hailing scenarios.

Conclusions:

• The developed pick-up point recommendation strategy enhances the efficiency of spatial crowdsourcing in online car-hailing.
• The four-dimensional model and incentive mechanism contribute to significant cost reductions and improved operational performance.
• This research provides a practical solution for optimizing ride-hailing logistics.