A passenger flow prediction method of the highest section of a bus line

Abstract

The invention discloses a passenger flow prediction method of the highest section of a bus line, comprising the following steps of extracting influence factors of the passenger flow of the section ofthe bus line in each period of time, and establishing a data space; providing the evaluation index based on the cost of prediction error; carrying out the parameter optimization on the data space to minimize the error cost; in the process of parameter optimization, using a Shepard interpolation algorithm to carry out the interpolation prediction on the passenger flow in the target section. The method of the invention utilizes interpolation algorithm to predict, has high stability of prediction, performs better in the prediction model which takes the prediction error cost as the evaluation index, can provide references for the bus line dispatching frequency setting, the capacity dispatching and the optimal load rate. Meanwhile, the evaluation index based on prediction error cost is put forward by using the idea of newsboy model, which reflects the cost of redundancy and passenger detention caused by the shortage of bus number, and provides a more direct reference for the follow-up optimization of bus departure frequency.

Description

technical field [0001] The invention relates to the field of passenger flow prediction in public transport operation management, in particular to a method for predicting passenger flow at the highest section of a bus line based on error cost and Shepard interpolation. Background technique [0002] One of the basic goals of bus service is to ensure that the passenger capacity in a given period of time is compatible with the maximum passenger flow along the bus line. According to the forecast time span, passenger flow forecasting can be divided into long-term passenger flow forecasting and short-term passenger flow forecasting. Long-term passenger flow forecasting Forecasting generally serves public transport system infrastructure construction and route planning, while short-term passenger flow forecasting generally serves public transport operation management, vehicle personnel scheduling, etc. [0003] For short-term bus passenger flow forecasting, the methods currently used...

AI Technical Summary

Problems solved by technology

However, there are defects in each of them. For example, the statistical forecasting method simply analyzes the passenger flow law from the perspective of data statistics, and its forecasting quality largely depends on the quality of statistical data. Therefore, such methods are not accurate and reliable.

Compared with the total line passenger flow, because the cross-section passenger flow involves the distribution of passenger boarding and alighting numbers along the line, its prediction will be more complex and uncertain

[0005] As mentioned above, the number of departures during the operating period depends on the predicted value of the highest cross-section passenger flow. When the forecast error of the cross-section passenger flow during an operating period does not lead to changes in the number of departures, the prediction results are used as the decision-making basis for public transport capacity deployment It is reliable, but if the cross-section passenger flow prediction error of a certain operation period does not reach or exceed the carrying capacity of the planned vehicle, using it as a public transport capacity will result in insufficient or wasted capacity, and the resulting operating loss (trips Too many passengers or stranded passengers) is the cost loss caused by the forecast error, that is, the error cost

Therefore, there are cases where the average error of the passenger volume prediction results in each operating period is small but the error cost is too large in some periods, and there are also situations where the average error is large but the error cost is small in most periods

Most of the existing technologies follow the traditional evaluation method with the goal of the lowest average error. Although this method has better prediction results, it is not applicable in the actual bus operation management with the goal of matching the transport capacity with the traffic volume.

Images