A Bicycle Quantity Prediction Method Based on Binary Gaussian Inhomogeneous Poisson Process

A forecasting method and bicycle technology, applied in forecasting, data processing applications, complex mathematical operations, etc., can solve problems such as delays, inability to conduct comprehensive research, and low prediction accuracy of the number of bicycles at the site

Active Publication Date: 2019-12-17
CHENGDU UNIV OF INFORMATION TECH
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Since there are many factors affecting bicycle use, and the acquisition of related data is subject to various restrictions, researchers usually have limited data, which leads to the inability to comprehensively study the factors affecting bicycle use
At the same time, the weather data is not updated in real time, which will cause delays for single-vehicle scheduling tasks with high implementation requirements
As a result, the prediction accuracy of the number of bicycles at the site is low, the reference value is not high, and the assistance provided for the scheduling of bicycles and the use of bicycles by users is limited

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Bicycle Quantity Prediction Method Based on Binary Gaussian Inhomogeneous Poisson Process
  • A Bicycle Quantity Prediction Method Based on Binary Gaussian Inhomogeneous Poisson Process
  • A Bicycle Quantity Prediction Method Based on Binary Gaussian Inhomogeneous Poisson Process

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0135] The operating environment of this embodiment is: Matlab2016b, Java (JDK 1.7), Windows 10, CPU (Core i7 (7500U), 8GB RAM.

[0136] In this embodiment, for the bay area bicycle system, bicycle station A (i=10), assume that the current date is August 23, 2013, at 10 am, and the number of bicycles is 10, that is, X bike (10 o'clock) = 10, predict the number of bicycles at the bicycle station A at 12:30 in the future.

[0137] Specific steps include:

[0138] S1. According to the historical data of the bicycle station A, extract the bicycle riding time series data of the past 10-15 days (that is, from the 13th to the 22nd, excluding the 23rd) from the bay area bicycle management system. sequentially and the time series of the user arriving and returning the bike ), the process of the user arriving and returning the bicycle and the process of the user arriving and riding the bicycle are regarded as Poisson processes, and the intensity function is μ i (x) and λ i (x) is...

Embodiment 2

[0148] The operating environment of this embodiment is: Matlab2016b, Java (JDK 1.7), Windows 10, CPU (Core i7 (7500U), 8GB RAM.

[0149] In this embodiment, for the hubway bicycle system, bicycle station A (i=20), assuming that the current date is August 23, 2011, at 9 am, the number of bicycles is 8, that is, X bike (9:00)=8, predict the number of bicycles at the bicycle station A at 10:00 am in the future.

[0150] Specific steps include:

[0151]S1. According to the historical data of bicycle station A, extract the bicycle riding time series data of the past 10 days (that is, 1st to 10th, excluding 11th) from the hubway bicycle management system (the time series of users riding bicycles) and the time series of users arriving and returning bicycles ), the process of the user arriving and returning the bicycle and the process of the user arriving and riding the bicycle are regarded as Poisson processes, and the intensity function is μ i (x) and λ i (x) is modeled as a n...

Embodiment 3

[0161] Such as image 3 and 4 As shown, in the two single-vehicle systems of Embodiment 1 and 2, the time after the current moment (Embodiment 1 is at 10:00 on August 23, 2013, and Embodiment 2 is at 9:00 on August 23, 2011) The continuous prediction of the number of bicycles verifies the usability of this method and the effect it achieves. The verification index is the root-mean-square-error (RMSE), and its calculation formula is:

[0162]

[0163] Among them, K represents the prediction time set, is the actual usage, and X(t+s) is the predicted usage.

[0164] The smaller the RMSE value, the higher the accuracy rate.

[0165] The prediction effect of bicycle is as follows: image 3 and Figure 4 As shown, on the Bay Area bicycle system and the Hubway bicycle system, the comparison methods are HA, ARMA and QMP, where:

[0166] HA (Historical Average), described in Z.Yang, J.Hu, Y.Shu, P.Cheng, J.Chen, and T.Moscibroda, "Mobilitymodeling and prediction in bike-sharin...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a method for predicting the number of bicycles based on a binary Gaussian non-homogeneous Poisson process, which belongs to the technical field of data mining and prediction. According to the historical data of bicycle sites, the Poisson theory is used to establish a non-homogeneous Poisson model, and the environment is considered Influenced by parameters, the final prediction model is established, the time series is simulated and the number of bicycles is corrected. Only the historical bicycle usage data of the bicycle site can be used for prediction. Not only can the corresponding data acquisition be limited, the site The prediction of the number of bicycles can reduce the impact of weather and other non-real-time update data on the prediction results, and significantly improve the accuracy and timeliness of the number of bicycles at the site in a certain period of time in the future.

Description

technical field [0001] The invention relates to the technical field of data mining prediction, in particular to a method for predicting the number of bicycles based on a binary Gaussian non-homogeneous Poisson process. Background technique [0002] As a new type of transportation, shared bicycles have developed rapidly in recent years. In the process of use, the distribution of bicycles is often uneven. In order to facilitate the management of the bicycle system administrator, it is necessary to dynamically schedule the bicycles. To dispatch bicycles, it is necessary to predict the number of bicycles at the station in advance. Predicting the number of bicycles is also a challenging task as bicycle usage is affected by many uncertain factors. [0003] In the existing technology, the commonly used method is to mine the factors that affect the use of bicycles, and then combine some data mining methods (such as regression model, decision tree, neural network, support vector ma...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06Q10/04G06F17/18
CPCG06F17/18G06Q10/04
Inventor 乔少杰黄飞虎韩楠彭京魏军林温敏冉先进
Owner CHENGDU UNIV OF INFORMATION TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products