The Analysis Of High Profitable Passengers Seeking Recommendation Strategy Based On Taxi GPS Data

With the rapid development of economy and urbanization,people's demand for travel services is increasing.Taxi,as an important supplement of urban public transport,plays an important role in resident trip.However,because of the unbalanced spatial and temporal distribution of passengers source,high no-load rate and unhealthy competition keep releasing negative impacts on taxi business.With the development of ITS technology,more and more attempts have been made on taxi passengers seeking pattern research,but generally lacking the analysis of high profitable passengers seeking strategy based on profit evaluation indexes and the situation of taxi saturation and traffic congestion.Based on the analysis of the factors that affect the profit of taxi seeking passengers,this paper proposes a high profitable passengers recommendation strategy with full consideration of the relationship between supply and demand of the taxi and the traffic status.This method can provide decision support for the high profit of taxi.At the same time,it has great significance for improving the current situation of uneven distribution of taxi resources,and giving full play to the role of taxis in urban transportation.This paper takes the combination of car load state and adjacent no-load state as the research object and proposes a profit index model based on the combination of income per unit time in the load state and the passengers seeking time in the adjacent no-load state.On the basis of the index model and the data preprocessing and map matching of taxi GPS data,the classical mathematical statistics and geostatistical analysis methods are used to analyze the characteristics of the spatial and temporal distribution of high profitable passengers.The comparative analysis on the variation characteristics of passenger volume and the spatial distribution characteristics of the high profitable passengers in different periods in a day and in working days and non-working days is conducted to provide support to the recommendation of high profitable passengers.In the course of recommendation of high profitable passengers,based on the division of high profitable hotspot area and the full account of the taxi saturation and the current state of traffic congestion after taxi driving to the area,a recommendation index model based on the profit of taxi is put forward.The model takes the shortest travel time from the start point to the hotspot area for the foundation to calculate the taxi supply and demand ratio and Traffic Performance Index,which can get the recommendation index to assess whether the high profitable hotspot area is suitable for the empty taxi.And then form a complete set of high profitable passengers seeking recommendation strategy.In the end,we take the taxi of the early peak in April 18,2011 in Shenzhen as an example,and carry on the high profitable hotspot area recommendation based on the taxi profit index model and the recommendation index model.Furthermore,the effectiveness of the recommendation strategy is verified by comparative experimental analysis.