Research On Dynamic Pricing Of Rented Vehicles Based On Customer's Choice Preference

In 2017,the number of motor vehicles in China was 310 million,of which 217 million were automobiles,and the proportion of automobiles have rapidly increased from 54.93% to 70.17% in the past five years,which led to the pressure of urban traffic that has not been effectively alleviated.Meanwhile,cities such as Beijing,Shanghai and Hangzhou have issued measures such as license plate and driving limited to relieve the traffic pressure in the city.The emergence of car rental that not only reduces the number of private cars,alleviates the traffic pressure,but also provides convenience for the residents who have car rental needs,is a supplement to public transportation,but there are still some potential problems in the development of car rental.For example,the car rental permeability was only 0.4% in China,compared with 1.5% in developed countries,which mainly caused by the high cost of car rental.Therefore,this paper will discuss and study the problem of setting price of car rental by the means of building a dynamic pricing model with the in-depth study of customer rental preference,in order to make more reasonable prices and improve the permeability of car rentals.Furthermore,it will help expand the rental market and increase the profits of leasing enterprises.First,the relevant literatures on the theory and method of car rental pricing at home and abroad as well as the pricing experience of aviation and perishable products were studied,while the deficiencies of these researches in the field of car rental pricing were summed up to determines the research direction of this paper.At the same time,the current situation of car rental industry was discussed,and the pricing of different cars at different time periods of four car rental companies in CAR,eHi,Car2 go and Pandauto were analyzed in detail to help understand the status of customer rental demand and different preferencesSecondly,in order to get the probability expression of customer renting,LN,PLN and DPLN were derived in detail,and the advantages and disadvantages of different distribution were also enumerated.The results showed that the fitting degree of DPLN was the best.Combined the characteristics of car rental probability and customer selection preference,the classical newsvendor model was derived and improved,and the dynamic pricing model of this study was obtained.Then,NLM and MLE functions in R software were used to estimate and calibrate the model parameters.Meanwhile,ZIF law was used for model checking.Finally,the initial questionnaire was designed for the preference of the customer's car rental choices.Through the reliability test and validity test of the cronbach reliability analysis,KMO test and Bartlett's test,an effective questionnaire was obtained,and then the data of the survey were analyzed,as well as the structure validity and reliability of each part of the questionnaire,which ensure the reliability of the data sources.Based on the above data base,the model parameters were calibrated to obtained the pricing rules of different preference groups,and compared the pricing of different preference groups at different time periods.The results showed that the dynamic pricing model based on the preference of customer rental preference was effective and practical,and customer preference had strong significance on pricing.The classification of customers by preference to customers and the different car rental schemes could not only meet the needs of different customer groups,but also guaranteed the significant promotion of car rental rate and further expanded the market size for the leasetener.The increase in revenue has laid a good foundation.