Research On The Availability Of Shared Bikes Based On Abnormal Trip Data Mining

The operation and management of shared bikes is of great significance for reducing carbon emissions and realizing sustainable development of urban transportation.However,with the increase of use time,shared bikes will gradually show a trend of deterioration,the level of availability decreases,resulting in a decline in user satisfaction,resulting in frequent rent-out phenomenon.If the bicycle is not repaired in time,it may have more serious consequences,such as affecting the safety of users and urban traffic problems.Currently,due to the lack of fault and maintenance information on shared bikes,there is no reliable online information indicating the availability of bikes,which increases the difficulty of quickly identifying unusable shared bikes from the perspective of system reliability.Therefore,it is of great significance to quickly identify the availability level of shared bikes.Therefore,by exploring user cancellation data from the travel dataset,the relative level of availability of shared bikes can be ranked.Based on the characteristics of users’ rental cancellation data,a user shared bike travel chain was constructed.The binary topology relationship is extracted from the user travel chain,and the relative availability level information of two shared bikes in any topology unit is transmitted.Then the topological ranking algorithm of directed cyclic graph is used to determine the topological ranking of the relative availability level of all shared bikes based on the binary topological relationship.However,this study can only understand the relative ranking of bikes,and the specific degree of failure of bikes,that is,the unusable probability of shared bikes and the specific number of unusable bikes in the site cannot be known.In addition,the research on the relative availability level of shared bikes based on activity chain is to analyze the abnormal travel data of users,namely the abnormal transfer data of users.This part of data is only for shared bikes with transfer relationship,not bikes without transfer relationship,so we cannot get the relative availability level ranking of all bikes.Based on the above,in order to obtain the availability level of all bikes,specifically understand the degree of failure of bikes and the number of unusable bikes in the site,a Bayesian model is proposed in this paper.According to the rental transaction data of users’ trips,the rent-out data is extracted,site attributes and the time information of users’ rent-out shared bikes are introduced,and a Bayesian extended model with covariables is constructed.The unavailability probability of all shared bikes and the number of unusable bikes in the site are estimated by using online transaction data.Finally,this paper further discusses how to determine the optimal sliding time window,and uses genetic algorithm to plan the maintenance path of the faulty bike.The method proposed in this paper is based on real data from bike-sharing system database to prove its effectiveness.This thesis has enriched the usability research from the perspective of the shared bicycle itself.At the same time,in the process of data mining analysis,abnormal travel data was identified and used as a basis for usability research on shared bicycles.Most of the existing studies have been conducted on normal user travel data,and this paper provides a direction for future data research.In addition,the research in this thesis provides an understanding of the level of availability of shared bicycles and provides a basis for managers to arrange maintenance tasks for faulty bicycles in a timely and orderly manner.