Research On Price Discounting Strategy Of Carsharing System Based On User Characteristic Analysis

Car sharing can effectively alleviate traffic congestion,reduce air pollution,and reduce dependence on energy.It is an important part of the sustainable transportation system.However,the development of carsharing is also facing new problems and challenges.A scientific and reasonable price discount strategy is a prerequisite to ensure the safe,stable and efficient development of the sharing economy.It can not only improve resource utilization and improve the imbalance of supply and demand at carsharing stations,but also improve the economic benefits of carsharing operators.Different user groups have different travel characteristics.In the process of formulating price discount strategies,it is necessary to consider the differences in travel behaviors of different user groups in order to better meet the needs of different types of users.In recent years,autonomous driving car technology has gradually been applied to the transportation field,and autonomous driving carsharing will soon become a new service mode of carsharing systems.The addition of the autonomous driving carsharing service model to the carsharing system enriches the service model of the carsharing system,allowing users to enjoy point-to-point services,and also helps to increase the user’s acceptance of autonomous driving cars and promote the carsharing system The development of intelligence.Therefore,the main research contents of this article are:First of all,based on the user’s order and station data,the k-means clustering method is used according to the first use time(the number of days between the user’s first use of the carsharing and the data deadline)and the last use time(the number of days between the user’s last use of the carsharing and the data deadline)and monthly usage frequency(the average monthly frequency of users using carsharing)three indicators group users into three types.Analyze the differences in travel characteristics of the three types of users in terms of travel frequency,travel distance,etc.,and respectively predict the short-term needs of the three types of users according to the ARIMA(Differential Autoregressive Moving Average Model)and LSTM(Long Short Term Memory Neural Network)models.The results show that the prediction accuracy of the LSTM model is high.Secondly,according to the previous clustering results,combined with the price demand elasticity coefficient and travel efficiency of different types of users,the optimization goal is to maximize the revenue of carsharing operators,and different price discount strategies are adopted for different types of users,and different considerations are established.The carsharing price discount strategy model based on the travel characteristics of the types of users is solved by genetic algorithm,and finally the optimal price discount strategy is formulated for each type of users.Among them,the first type of users has the most trip volume but the corresponding price discount is also the highest.The "killing familiarity" strategy commonly used by operators appears.In order to further analyze the advantages and disadvantages of considering different types of user travel characteristics in the price discount strategy,the price discount strategy based on user classification is compared with the price discount strategy model based on unclassified users,and it is found that the price discount strategy model based on user classification can be allow operators to benefit more and increase the number of trips in the carsharing system.Finally,the autonomous driving carsharing service mode is added to the original ordinary carsharing system to form a carsharing system that is a hybrid of the ordinary carsharing system service mode and the autonomous driving carsharing service mode.In the hybrid carsharing system,based on the travel characteristics of the three types of users analyzed in the previous article,a price discount strategy model for the hybrid carsharing system based on user classification is established,and the optimal price discount strategy is formulated for different types of users to choose different service modes.And decide the vehicle and parking space configuration of the hybrid carsharing system.The genetic algorithm is used to solve the model.Finally,the impact of the user-classified price discount strategy and the user-unclassified price discount strategy on the hybrid carsharing system is compared,and the single carsharing system(only the common carsharing service mode is configured in the system)and the trip volume of the hybrid carsharing system and the revenue of operators.The results show that,compared with the user-unclassified price discount strategy,the user-classified price discount strategy can increase the revenue of operators and increase the total number of user trips;compared with a single carsharing system,the hybrid carsharing system can increase the total trip volume and revenue of the system.In the hybrid carsharing system,users prefer the autonomous driving carsharing service mode.The first type of users has the highest trip volume under the autonomous driving carsharing service mode,but the corresponding price discount is also the highest.There is a "killing familiarity" strategy that operators often use in price discount strategies.The price discount strategy of the carsharing system is directly related to the profit of the operator and the travel cost of the user.A reasonable price discount strategy can increase the revenue of the operator,promote the balance of supply and demand,and achieve a win-win situation of maximizing the profit of the operator and minimizing the travel cost of the user.The addition of the autonomous driving carsharing service mode to the ordinary carsharing system enriches the service mode of the carsharing system.When establishing a price discount strategy model in a single carsharing system or a hybrid carsharing system,the travel characteristics of different types of users can be considered,which can provide targeted services to different types of users and better meet the needs of different types of users.The price discount strategy in this article can provide a reference for operators to formulate price discount strategies.It also provides new ideas for the future development of carsharing systems to add the autonomous driving carsharing to the ordinary autonomous driving car system,and it is of great significance of the intelligent development of carsharing.