A lease cabinet cluster scheduling and transport capacity allocation method based on time series prediction

By constructing a unified timeline operation sequence and time-series prediction model for rental lockers, imbalance results within future windows are generated. Combined with the locker group scheduling relationship matrix, the supply and demand mismatch problem in the rental locker system during peak periods is solved, achieving efficient equipment scheduling and capacity allocation, and improving the system's responsiveness and equipment scheduling efficiency.

CN122453006APending Publication Date: 2026-07-24广州浩安智能科技有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
广州浩安智能科技有限公司
Filing Date
2026-04-21
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Existing rental locker systems suffer from several problems during peak tourist seasons, popular route shifts, and after performances. These problems include continuously empty lockers at entrances, continuously full lockers at exits, prolonged periods of unavailability at certain nodes, and frequent but ineffective transfers. Furthermore, the system lacks the ability to incorporate service accessibility relationships, path costs, current receiving capacity, and current transfer capacity between lockers into the locker group relationship model, resulting in supply and demand complementarity at the prediction level but ineffective implementation at the execution level.

Method used

By acquiring data on the borrowing and returning behavior, inventory status, and regional passenger flow of rental lockers, a locker operation sequence with a unified time axis is constructed. A time-series prediction model is used to generate an operation prediction sequence within the future window. The imbalance result is calculated by combining the time decay factor and the passenger flow squeeze term. A locker group scheduling relationship matrix is ​​constructed, and a scheduling instruction set that meets physical constraints is generated to realize equipment allocation and capacity distribution.

Benefits of technology

It improved the success rate of borrowing and returning during peak hours, reduced the duration of local node imbalances and the pressure on operation and maintenance scheduling, solved the problems of continuous stockouts and full containers, and improved the system's responsiveness and equipment scheduling efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122453006A_ABST
    Figure CN122453006A_ABST
Patent Text Reader

Abstract

The application provides a lease cabinet cluster scheduling and transport capacity allocation method based on time sequence prediction, which comprises the following steps: obtaining the lease cabinet's lending and returning log, inventory state and regional passenger flow data, and constructing a cabinet body operation sequence of a unified time axis; secondly, using a time sequence prediction model to generate an operation prediction sequence in a future window, and calculating a future window imbalance result in combination with time attenuation and passenger flow extrusion factors; thirdly, generating a cabinet group scheduling relationship result in combination with a service cost and a current capacity boundary according to the imbalance result; finally, generating a scheduling instruction set meeting physical constraints in combination with actual capacities on both supply and demand sides according to a relationship allocation coefficient. Through the deduction of continuous time windows and the modeling of cabinet group relationships, the application effectively solves the problems of supply and demand mismatch and local node imbalance in peak periods, and improves the overall balancing capacity and service efficiency of the lease cabinet cluster.
Need to check novelty before this filing date? Find Prior Art