CN104156830B - Kart driving training amount Forecasting Methodology based on S curve - Google Patents
Kart driving training amount Forecasting Methodology based on S curve Download PDFInfo
- Publication number
- CN104156830B CN104156830B CN201410410559.1A CN201410410559A CN104156830B CN 104156830 B CN104156830 B CN 104156830B CN 201410410559 A CN201410410559 A CN 201410410559A CN 104156830 B CN104156830 B CN 104156830B
- Authority
- CN
- China
- Prior art keywords
- training
- driving
- amount
- predicted
- students
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of kart driving training amount Forecasting Methodology based on S curve, comprises the following steps：(1) the history training amount data in city to be predicted are gathered；(2) the city driving amount forecast model to be predicted is built, the model includes student enrollment's driving amount forecast model, local of the right age treats start-up's driving amount forecast model and nonlocal source of students driving amount forecast model；(3) predicted value for obtaining the following car driving training total amount in the city to be predicted is calculated.The present invention is with preferable real-time and reliability is high, repeatable, parameter is easily changed key character, the Characteristics of Development in present stage of china driving training market is complied fully with, the overall development and planning for town government department and driving training are significant.
Description
Technical field
The present invention relates to computer realm, and in particular to a kind of kart driving training amount prediction side based on S curve
Method.
Background technology
In recent years, due to the fast development of social economy, China's resident's per capita disposable income increase, the purchase of kart
Buy power and also obtained raising in respective degrees, the recoverable amount of motor vehicle particularly automobile lures increasing in the trend that increases substantially
Motor vehicle driving training amount also there is blowout phenomenon.To China's traffic safety bring huge in the rapid growth of fresh driver
Under the real background challenged greatly, China in 2012 has carried out the major transformation of driving training examination system, the change of Driving Test form
So that driving training amount is also fluctuated therewith, various regions start Plan idea driving school system, and the important evidence of planning is then future
The prediction of driving training amount.
The content of the invention
Goal of the invention：It is an object of the invention to solve deficiency of the prior art there is provided a kind of based on the small-sized of S curve
Car steering training amount Forecasting Methodology.
Technical scheme：The present invention a kind of kart driving training amount Forecasting Methodology based on S curve, specifically include with
Lower step：
(1) training source of students is divided into student enrollment, local of the right age treats that start-up and the nonlocal part of source of students three collection are treated
The history training amount data of predicted city；
(2) based on the data obtained by step (1), the driver training amount to three part sources of students in (1) is predicted respectively,
Student enrollment's driving amount is predicted using quantitative approach, locality is of the right age treats that start-up's driving amount is predicted using S sigmoid growth curves
Predicted, comprised the following steps that using regression analysis with nonlocal source of students driving amount：
(2.1) according to local student enrollment's history quantity and purpose car ratio x, trend extrapolation model estimation is built pre-
Survey phase student enrollment's quantity a, then obtains student enrollment's driving training total demand a*x；University student is in each academic year middle school
The possibility for practising driving license is evenly distributed, and total demand is temporally allocated, and is produced newly-increased of annual student enrollment and is driven
Demand P1；
(2.2) it is locally of the right age to treat that training personnel add up training amount and obey S sigmoid growth curves；If t local of the right age society
It is y to treat that training personnel add up training amountt, variable ytS curve model profile is obeyed, by S curve modelTake the logarithm change
Shape obtains ln yt=ln K+btLn a, wherein, it is local of the right age to treat that training personnel refer to that age bracket belongs to 18-70 one full year of life crowd and not
Including student enrollment；
(2.3) parameter K is determined：Parameter K is ytSaturation value, according to local statistics and population ages feature, obtain
Age bracket ((18-t)~(70-t)) size of population m, with reference to similar urban development experience, determines that kart driving license holds rate
Saturation value is c, then K=c*m；
(2.4) according to the formula fitting historical data after deformation, the value of parameter a, b is obtained, you can obtain time span of forecast sheet
The of the right age society in ground treats training personnel's driving training demand P2；
(2.5) kart is accounted for according to nonlocal source of students in recent years and drives training amount percentage Ψ0, build regression mathematical model fitting and estimate
Calculate following nonlocal source of students kart and drive training demand P3；
(3) the driver training amounts of the three part sources of students in step (2) predicts the outcome, and calculating obtains city to be predicted
The predicted value of city's future car driving training total amount.
Further, the step of data acquisition of step (1) is：
(1.1) city student enrollment quantity size to be predicted and purpose car ratio are obtained by way of field investigation
Example；
(1.2) city nonlocal source of students population over the years is gathered by way of sample investigation and data collection.
Further, the process that implements of step (3) is to predict the training amount of three models in step (2) and tie
Fruit is added, and the result of addition is the following car driving training Prediction of Total value in city to be measured, i.e.,：P=(P1+P2+P3)。
Beneficial effect：Compared with prior art, the present invention can solve the problem that existing Forecasting Methodology and Characteristics of Development at this stage
Mismatch, the present invention is with preferable real-time and reliability is high, repeatable, parameter is easily changed key character, completely symbol
Close the Characteristics of Development in present stage of china driving training market, overall development for town government department and driving training and
Planning is significant.
Brief description of the drawings
Fig. 1 is flow chart of the invention；
Fig. 2 is S type curve matching schematic diagrames under normality in the present invention.
Embodiment
Technical solution of the present invention is described in detail in conjunction with the accompanying drawings and embodiments below.
As depicted in figs. 1 and 2, a kind of kart driving training amount Forecasting Methodology based on S curve, is specifically included following
Step：
(1) the history training amount data in city to be predicted are gathered：
(1.1) city student enrollment quantity size to be predicted and purpose car ratio are obtained by way of field investigation
Example；
(1.2) urban history is gathered by way of sample investigation and data collection and adds up accredited number xt, it is local suitable
Age, society treated that the accredited number of start-up accounts for the ratio of accredited total number of personsAnd nonlocal source of students population over the years.
(2) based on the data obtained by step (1), the city driving amount forecast model to be predicted is built, the model is included in
School university student's driving amount forecast model, local of the right age treat start-up's driving amount forecast model and nonlocal source of students driving amount prediction mould
Type, is comprised the following steps that：
(2.1) according to local student enrollment's history quantity and purpose car ratio x, trend extrapolation model estimation is built pre-
Survey phase student enrollment's quantity a, then obtains student enrollment's driving training total demand a*x；University student is in each academic year middle school
The possibility for practising driving license is evenly distributed, and total demand is temporally allocated, and is produced newly-increased of annual student enrollment and is driven
Demand P1；
(2.2) it is locally of the right age to treat that training personnel add up training amount and obey S sigmoid growth curves；Make t local of the right age society
It is y to treat that training personnel add up training amountt, variable ytS curve model profile is obeyed, by S curve modelTake the logarithm change
Shape obtains ln yt=ln K+btLn a, wherein, it is local of the right age to treat that training personnel refer to that age bracket belongs to 18-70 one full year of life crowd and not
Including student enrollment；
(2.3) parameter K is determined：Parameter K is ytSaturation value, according to local statistics and population ages feature, obtain
Age bracket ((18-t)~(70-t)) size of population m, with reference to similar urban development experience, determines that kart driving license holds rate
Saturation value is c, then K=c*m；
(2.4) according to the formula fitting historical data after deformation, the value of parameter a, b is obtained, you can obtain time span of forecast sheet
The of the right age society in ground treats training personnel's driving training demand P2；
(2.5) kart is accounted for according to nonlocal source of students in recent years and drives training amount percentage Ψ0, build regression mathematical model fitting and estimate
Calculate following nonlocal source of students kart and drive training demand P3。
(3) three models in step (2) predict the outcome, and calculating obtains the following car in city to be predicted and driven
The predicted value of total amount is trained, i.e.,：The training amount of three models in step (2) is predicted the outcome addition, the result of addition is
The following car driving training Prediction of Total value in city to be measured, i.e.,：P=(P1+P2+P3)。
Embodiment：Below by taking the prediction process of Huai'an kart driving training demand as an example to the present invention specifically
It is bright.
1st, Huai'an urban history statistics and population characteristic's investigation：The side being combined by forum and field investigation
Formula gathers urban history statistics, analyzes its population ages feature；
2nd, according to Huai'an student enrollment's quantity size and purpose car ratio, prediction student enrollment driving training is needed
Ask, comprise the following steps that：
According to the statistics of Huai'an Bureau of Education, Huai'an 2008-2012 student enrollment's quantity is as shown in table 1.
8.7 ten thousand Hes will be respectively reached by building regression model calculating prediction 2015 and the year two thousand twenty Huai'an student enrollment's quantity
9.0 ten thousand people.
University student is evenly distributed in the possibility of each academic year learning driving license, 2013-2015 wish car ratio
It is 40%~50% for 30%~40%, 2016-2020 wish car ratio.Then plan that year student enrollment drives training amount pre-
Measured value is as shown in table 2.
The nearly 5 years student enrollment quantity (unit in the Huai'an of table 1：Ten thousand people)
Table 2 project period of Huai'an student enrollment drives training amount predicted value (unit：Ten thousand person-times)
3rd, the ratio and variation tendency of driving training total amount, the nonlocal source of students driving training of prediction are accounted for according to nonlocal source of students over the years
Demand, detailed process is as follows：
According to investigation statisticses, the nonlocal source of students that Huai'an kart human pilot training is participated in recent years accounts for small size gasoline
Car drives the 10% of training amount, and the nonlocal source of students number fluctuating range driven to Huaian in each year is smaller.Predict 2015 and the year two thousand twenty
Training amount is driven in the part will respectively reach 5500 person-times and 6000 person-times.
4th, treat that training personnel learn the feature that demand of driving tends to market saturation according to locally of the right age, be fitted with S curve, in advance
Survey is locally of the right age to treat training personnel's driving training demand, comprises the following steps that：
According to《Huai'an overall city planning (2009-the year two thousand thirty)》, 2015, the year two thousand twenty Huai'an administrative region of a city total population will
5,300,000,5,600,000 people are respectively reached, meanwhile, according to the national census data of Huai'an 2010 year the 6th time, Huaian in 2010
City permanent resident population about 4,800,000, account for the 85% of household registration population, it is about 70% that 15 to 64 years old population, which accounts for permanent resident population's ratio, compare accordingly
Example measuring and calculating, then 18-70 one full year of life populations in the year two thousand twenty Huai'an are up to 3,600,000 or so.With reference to the Experience in Development in similar city, 2020
The saturation value that year Huai'an colony's kart driving license holds rate is 30%, then the of the right age society in Huaian treats that training personnel are accumulative
The saturation value of training amount is 1,100,000.
According to Huai'an public security organ statistics, each year, local of the right age society treated that training personnel add up to drive the training amount such as institute of table 3
Show.
Training personnel, which drive training demand and are predicted, to be treated to the local of the right age society of project period according to S curve model, S curve model
Formula is as follows：
In above formula：
T refers to predict time t；
ytRefer to that the local of the right age societies of t treat that training personnel are accumulative and drive training amount；
K, a, b-parameter to be estimated.
It is fitted according to above-mentioned formula and historical data with mathematica softwares and tries to achieve each parameter value：K=
110；A=0.145；B=0.82.S curve basically reached saturated level after 2025.
Accordingly, y in the planning time limit can be drawntPredicted value, so as to show that local of the right age society of each year treats that training personnel drive training
Predicted value is measured, as shown in table 4.
The local of the right age society over the years of table 3 treats that training personnel are accumulative and drives training amount (unit：Ten thousand people)
The local of the right age society of table 4 treats that training personnel drive training amount predicted value (unit：Ten thousand person-times)
Three fractional prediction results are collected and produce the following each year kart driving training amount predicted value in Huai'an, such as table 5
It is shown：
The local of the right age society of table 5 treats that training personnel drive training amount predicted value (unit：Ten thousand person-times)
Above-described embodiment shows that the present invention quick and convenient can accurately predict small in the following regular period of a certain city
Car steering training amount.
Claims (3)
1. a kind of kart driving training amount Forecasting Methodology based on S curve, it is characterised in that comprise the following steps：
(1) training source of students is divided into student enrollment, local of the right age treats that start-up and the nonlocal part of source of students three collection are to be predicted
The history training amount data in city；
(2) based on the data obtained by step (1), the driver training amount to three part sources of students in (1) is predicted respectively, in school
University student's driving amount is predicted using quantitative approach, locality is of the right age treats that start-up's driving amount is predicted and outer using S sigmoid growth curves
Ground source of students driving amount is predicted using regression analysis, is comprised the following steps that：
(2.1) according to local student enrollment's history quantity and purpose car ratio x, trend extrapolation model estimation time span of forecast is built
Student enrollment quantity a, then obtains student enrollment's driving training total demand a*x；University student drives in each academic year learning
According to possibility be evenly distributed, total demand is temporally allocated, produce annual student enrollment increase newly drive demand
Measure P1；
(2.2) it is locally of the right age to treat that training personnel add up training amount and obey S sigmoid growth curves；If t local of the right age society waits to train
It is y that personnel, which add up training amount,t, variable ytS curve model profile is obeyed, by S curve modelTake the logarithm and deform
To lnyt=lnK+btLna, wherein, it is locally of the right age to treat that training personnel refer to that age bracket belongs to 18-70 one full year of life crowd and is not included in school
University student；
(2.3) parameter K is determined：Parameter K is yt saturation value, according to local statistics and population ages feature, obtains the age
Section ((18-t)~(70-t)) size of population m, with reference to similar urban development experience, determines that kart driving license holds the saturation of rate
It is worth for c, then K=c*m；
(2.4) according to the formula fitting historical data after deformation, the value of parameter a, b is obtained, you can obtain time span of forecast and locally fit
Age, society treated training personnel's driving training demand P2；
(2.5) kart is accounted for according to nonlocal source of students in recent years and drives training amount percentage Ψ0, build regression mathematical model fitting estimation not
Carry out nonlocal source of students kart and drive training demand P3；
(3) the driver training amounts of the three part sources of students in step (2) predicts the outcome, and calculating obtains city to be predicted not
Carry out the predicted value of car driving training total amount.
2. the kart driving training amount Forecasting Methodology according to claim 1 based on S curve, it is characterised in that：Step
Suddenly the step of data acquisition of (1) is：
(1.1) city student enrollment quantity size to be predicted and purpose car ratio are obtained by way of field investigation；
(1.2) city nonlocal source of students population over the years is gathered by way of sample investigation and data collection.
3. the kart driving training amount Forecasting Methodology according to claim 1 based on S curve, it is characterised in that：Step
Suddenly the process that implements of (3) is, the training amount of three models in step (2) is predicted the outcome addition, and the result of addition is
For the following car driving training Prediction of Total value in city to be measured, i.e.,：P=(P1+P2+P3)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410410559.1A CN104156830B (en) | 2014-08-19 | 2014-08-19 | Kart driving training amount Forecasting Methodology based on S curve |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410410559.1A CN104156830B (en) | 2014-08-19 | 2014-08-19 | Kart driving training amount Forecasting Methodology based on S curve |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104156830A CN104156830A (en) | 2014-11-19 |
| CN104156830B true CN104156830B (en) | 2017-08-25 |
Family
ID=51882325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410410559.1A Expired - Fee Related CN104156830B (en) | 2014-08-19 | 2014-08-19 | Kart driving training amount Forecasting Methodology based on S curve |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104156830B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106096769A (en) * | 2016-05-31 | 2016-11-09 | 中国地质科学院矿产资源研究所 | Zinc demand prediction method based on S-shaped model |
| CN106126769A (en) * | 2016-05-31 | 2016-11-16 | 中国地质科学院矿产资源研究所 | Lead demand prediction method based on S-shaped model |
| CN107463581B (en) * | 2016-06-02 | 2020-09-08 | 广州泰尔智信科技有限公司 | Application download amount acquisition method and device and terminal equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008037088A1 (en) * | 2006-09-29 | 2008-04-03 | Nortel Networks Limited | A method and system for predicting the adoption of services, such as telecommunication services |
| CN101561972A (en) * | 2008-04-17 | 2009-10-21 | 北京宣爱智能模拟技术有限公司 | Method and system for interactive automobile driving case teaching |
| CN101645161A (en) * | 2009-08-27 | 2010-02-10 | 浙江维尔生物识别技术股份有限公司 | Driving training management and time- and charge-counting system based on radio function |
| CN101867955A (en) * | 2010-03-01 | 2010-10-20 | 北京邮电大学 | Gompertz function based different network user number predicting method |
-
2014
- 2014-08-19 CN CN201410410559.1A patent/CN104156830B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008037088A1 (en) * | 2006-09-29 | 2008-04-03 | Nortel Networks Limited | A method and system for predicting the adoption of services, such as telecommunication services |
| CN101561972A (en) * | 2008-04-17 | 2009-10-21 | 北京宣爱智能模拟技术有限公司 | Method and system for interactive automobile driving case teaching |
| CN101645161A (en) * | 2009-08-27 | 2010-02-10 | 浙江维尔生物识别技术股份有限公司 | Driving training management and time- and charge-counting system based on radio function |
| CN101867955A (en) * | 2010-03-01 | 2010-10-20 | 北京邮电大学 | Gompertz function based different network user number predicting method |
Non-Patent Citations (1)
| Title |
|---|
| 长沙市驾培行业需求预测分析;周南金等;《长沙大学学报》;20110915;第25卷(第05期);第77-80页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104156830A (en) | 2014-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wetzels et al. | Bayesian parameter estimation in the Expectancy Valence model of the Iowa gambling task | |
| Kong et al. | Unravelling winter diatom blooms in temperate lakes using high frequency data and ecological modeling | |
| Cheng et al. | Complementarity and substitutability: A review of state level renewable energy policy instrument interactions | |
| CN106504475A (en) | Method for detecting fatigue driving based on EEG signals | |
| Gadhavi et al. | Student final grade prediction based on linear regression | |
| CN104156830B (en) | Kart driving training amount Forecasting Methodology based on S curve | |
| Bukhari et al. | Consequence of FDI on CO2 emissions in case of Pakistan | |
| Ozcan et al. | An empirical investigation between renewable energy consumption, globalization and human capital: A dynamic auto-regressive distributive lag simulation | |
| van Middelkoop et al. | Are Dutch residents ready for a more stringent policy to enhance the energy performance of their homes? | |
| Wang et al. | Leveraging First Response Time into the Knowledge Tracing Model. | |
| CN104376388A (en) | Ultra-short period wind power prediction method based on wind speed factor control model | |
| Liu et al. | Visual observation or oral communication? The effect of social learning on solar photovoltaic adoption intention in rural China | |
| Zheng et al. | Advancing higher education and its implication towards sustainable development: Moderate role of green innovation in BRI economies | |
| Frank | Expert-based Bayesian Network modeling for environmental management | |
| Yu et al. | Towering sustainability: Unraveling the complex effects of skyscrapers on urban resilience | |
| Arifin et al. | Trend Analysis of the ARIMA Method: A Survey of Scholarly Works | |
| Besharat et al. | Perfectionism and Competitive Anxiety in Athletes | |
| Jia et al. | Modeling the Impact of Macroeconomic Factors on Multimodal Transport Demand: A Cointegration Approach | |
| Eltaief | Efficiency and productivity analysis of construction companies in Libya | |
| Zhang | Prediction of freight ability in country base on GRNN | |
| Chaabouni | Comparisons of inverse modelling approaches for predicting energy consumption, economic growth and CO2 emissions: a case study for low-income countries | |
| Shin | A study of college students who were granted special admissions for vocational high school students | |
| Ertiaei et al. | Factors influencing rural youth, participation regarding agricultural development (Case study: Kermanshah Township) | |
| Leighton et al. | Social promotion in primary school: immediate and cumulated effects on attainment | |
| CN103955742B (en) | A kind of PM2.5 forecasting procedure based on integrated study |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170825 Termination date: 20210819 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |