Proposal for optimal placement platform of bikes using queueing networks.

Area of Science:

• Operations Research
• Urban Planning
• Transportation Science

Background:

• Previous bike-sharing system experiments showed unclear effectiveness and frequent discontinuation.
• Existing distribution models lacked clarity on optimal node arrangement and resource allocation.

Purpose of the Study:

• To develop an effective model for arranging bike distribution nodes.
• To optimize the placement and number of bikes in rental systems.
• To provide a foundation for efficient rental cycle business operations.

Main Methods:

• Constructed a queueing network model for bike distribution node arrangement.
• Utilized Google Maps API for realistic distance and transit time data.
• Applied a gravity model to population distribution for transition probability calculation.

Main Results:

• The model enables precise system design when node arrangement and bike counts are known.
• Demonstrated optimization using convenience stores as nodes.
• Achieved simultaneous optimization of node quantity, placement, and bike allocation per node.

Conclusions:

• The developed queueing network model effectively optimizes bike distribution systems.
• The system provides a robust framework for establishing and managing rental cycle businesses.
• Optimized node configuration and bike allocation enhance operational efficiency and user experience.