Study on the Early Warning System of Road Passeng
Study on the Early Warning System of Road Passenger Transport
Jian-you ZHAO1，Yun-jiao ZHANG2, Yi-fan PAN3, Jian HE4
1. School of Automobile, Chang’ an University ,710064 ,xi’ an,
Tel:139******** , jyzhao@https://www.wendangku.net/doc/7f4848618.html,;
2.School of Automobile, Chang’ an University ,710064 ,xi’ an,
Tel:135******** , 178560167@https://www.wendangku.net/doc/7f4848618.html,;
3.School of Automobile, Chang’ an University ,710064 ,xi’ an,
Tel:159******** , 782359104@https://www.wendangku.net/doc/7f4848618.html,;
4.FujianTransportation vocational college, Fu Zhou, Fujian,350007
ABSTRACT
Road passenger transport warning systems were built using the early warning theory. It helped to enrich management means and methods of the road passenger transport industry, an dim proved the correctness and pertinence of macroscopic decisions. It also promoted healthy and fast development of roads for the passenger transport industry. The impact factors of road passenger transport are analyzed in this paper, and an evaluation system and its limits are constructed with scientific,systematic, feasibility and effectiveness principles. The warning score is estimated based on the Entropy theory and its trends are forecasted based on GM (1,1) model. The Fujian Province is carried out via empirical research as an example. The result shows that the early warning model works, and can provide evidence for guiding and regulating the development of the road passenger transport industry.
INTRODUCTION
Road passenger transportation is one of the most important modes of passenger trips in China. Recently, a new developing requirement was raised by road passenger transportation due to the development of the national economy and the acceleration of China's urbanization. What’s more, a new challenge was put forward by road passenger transportation due to the competition of railway passenger transportation, aviation passenger transportation and others. In order to promote the healthy and rapid development of the road passenger transportation industry, it is practical to design a new scientific management method and means to improve road passenger transportation macro decision accuracy and timeliness.
Early warning theory is widely used in the management of economic, political and social problems, and can control the process of solution development by monitoring it. If we use the early warning theory in
management, we could monitor the developing state of the industry, forecast its trend of development, and take the appropriate action to help the management department find problems that may occur in the industry development process early to ensure road passenger transportation remains in a positive development state.
THE CONSTRUCTION OF EARLY WARNING THEORY INDEX SYSTEM OF ROAD PASSENGER TRANSPORTATION
According to the principle scientific, systematic, practicability, reliability and road passenger industry development analysis, the early warning theory index system of road passenger transportation has been built. It reflects the road passenger transportation industry development state totally. The indexes can not only be the representative of the whole industry state, but also represent one of the specific conditions of industry development. According to the index properties it can be divided into industry development index, industry management index, industry construction index, supply balance index and industry quality index, as shown in figure 1.
Figure 1.Early warning theory index system of road passenger transportation
EARLY WARNING SYSTEM OF ROAD PASSENGER TRANSPORTATION
Road Passenger Warning Degree Judgment Based On Entropy Value-Comprehensive Point Rating Method
The first step of warning degrees forecast is to use an evaluation index system to evaluate the past and present developing state of the highway passenger transportation industry on the basis of the existing data. Highway passenger industry has complicated factors, both of quantitative and qualitative indexes which are hard to quantitative directly are contained in the evaluation index system. A comprehensive evaluation method could be used as an evaluation model to measure each index based on the warning degree interval they are in. Overview of Entropy
Entropy is a measure of disorder, or more precisely unpredictability. Information is a measure of an orderly system. The two absolute values were equal but signs were opposite(Li 2007).The definition of information entropy is that if there is a system A with multi-event in it A=｛x x x n ,...2,1｝ ,the probability distribution of each event were p=｛
p p p n ...,2,1｝ ,the entire system has the information entropy p p ij ij k S ∑?-=ln ，（i=1,2,...n ） (1)
in the formula is information entropy of system, is Boltzmann constant, is the probability of events in the system
A .
For an unknown index weight system, the greater entropy of the index, the more disorder it will cause in the system, conversely, system is developing in an orderly direction. If an index’s weight of entr opy is smaller, the provided information will be greater, and the index’s weight should also be greater, and vice versa. Warning degree judgment based on entropy value-comprehensive point rating method
Comprehensive evaluation method is a common evaluation method to evaluate the development state through the results of the evaluation of each index’s integrated weight. In this paper, the method of entropy determines the index weight which will make the evaluation more objective. The specific procedure is as follows:
1) Build a highway passenger evaluation factor set
U ={Industry development levelU1，Industry structure U2, industry level of supply and demand U3, industry quality level U4, industry management level U5}, level 2 evaluation for subset U1 = {the passenger growth rate U11 passenger turnover growth rate, U12 volume growth rate, the coach U13}, U2 = {level 3 above qualification enterprise complete passenger bus number worth, the proportion of U21 U22}, U3 = {average daily management U31 utilization rate and the average real load rate U32}, U4 = {passenger traffic accident
U41 accident rate, U42 mortality, travel accident rate cuts U43, high, middle-grade proportion car U44}, U5 = {road transport management regulations perfect degree U51 passenger market system, perfect degree U52}.
2) Determine the evaluation set V
According to the alarming, divide the warning interval into "adversity", " adversity to be", "prosperity" three state, and according to the three state evaluation set evaluation set V into 3 degrees, V = {V 1 (poor), V 2 (worse), V 3 (general)}, the corresponding points 1, 2, 3.
3) Evaluation weight determining based on the method of evaluation
The procedure of index weight calculation by Entropy method is as follows:
(1)Data standardization
As a result of each index data dimension was different, data must be standardized. The two methods of data dimensionless may refer to the following:
a. Level quantification. For qualitative indexes, using the experts scoring.
b. Ratio method. For quantitative indexes, using formula 4.3 (Liang, et. al.2002).
u u u u u
ij
ij ij
ij ij min max min '--= （2） In the formula,u ij is the j value of the index u i , u ij for the standardization of index value.
(2) Grey relation entropy method determines the index weight (Liang, et. al. 2002).
a. relational formula
Set u i =｛u u u in i i ,...2,1｝(which i=1,2,…， m ） for the index system data.
Y=｛
y y y n ,...2,1｝ for the reference sequence. The value of the reference sequence is y i =max （u ij ）( j 1,2, ,n ) , so the i index j data has the gray relational εij ,the value could calculate by formula 4.4:
y u y u y u y u j ij j i j ij j ij j
i j ij j i ij -+--+-=max max 5.0max max 5.0min min ε （3）
b. Determine Entropy value
Definition
∑=εεij ij ij p
（4）
The index j exporting Entropy value p p s ij ij
j k ∑?-=ln ,constant ())(ln 1m k -= Easy to know ,define the degree of deviation []1,0∈s j ,define the degree of deviation dj ,
[]),1(1n j s d j i ∈-- （5）
Suppose the decision maker has no obvious preferences to the index, the weight ωj of index j is: ∑=
j j j j d d ω (6)
When the decision maker preferences )(,...2,1g g g n G =, the weight j of index j is:
∑=j j j j j j d g d g ω (7)
4) Computation of comprehensive warning degree
According to the systematic method and method combined with experience data the index data could be divided as each warning range and then confirm the evaluation of estimate ,According to the calculated index weight proceeding weighted summation, the calculation of comprehensive warning degree Vi .
∑=V ij ij V ω （8）
5) Highway passenger industry developing alert judgment
According to the calculation formula, it is known that the calculation results of highway passenger transport industry development comprehensive warning degrees are always in the 1 ~ 3 range. Based on the experience data, each warning range was divided. Compared with the calculation results, the judgment of the warning could be completed. The comprehensive warning decision criteria are shown in Table 1.
The Highway Passenger Early Warning Model Based On TheGrey GM (1, 1) Model
The purpose of the highway passenger warning system is to forecast the development of the industry of the warning state degrees through accurate prediction of crisis and damage degree that may arise during the highway passenger industry development process.
Highway passenger warning system is a complex system. The warning index information has level complexity obviously, the structure has fuzzy relations, the development and change of the indexes have the properties of randomness and data integrity. The highway passenger early warning system could be regarded as a grey system for the error brought about by the statistical technology. This paper uses a gray GM (1, 1) model for highway passenger transport development warning degrees forecast. The specific steps are as follows:
1) For the original series ｝,...2,1)(｛)
0()0(n t t x x == ,conduct once accumulated generating series: ∑==k i t i k x x 1)0()()()( （k=1,2,...n ） （9）
2) Establish GM (1, 1) model
GM (1, 1) model for the general form is the differential equation:
u t ax dt t dx =+)()()1()1(； （10）
Solving the a,u by the least squares solution:
[][]n T T T Y B B B u a a 1,?-==
The discrete response of differential equation (4.11), namely the grey prediction model x (1) is :
()a u e a u x t x at +??? ??-=+-)1(?1?)0()1( （11） The forecasting model of the x (0) could calculated by the regressive of x （1） , namely ：
)(?)1(?)1(?)1()1()0(t x t x t x
-+=+ （12） 3) Model precision inspection
Calculation residual
())(?)()0()0(t x
t x t E -= （13） Set the average of the original series x （0）and residue series }{n t t E E ,,2,1)
( ==as x ,E respectively. Set the mean square error of the original series x （0） and residue series }
{n t t E E ,,2,1)
( ==as S1,S2 respectively. Thus 2
0021))((1∑=-=n t x t n x S （14） 2
022))((1∑=-=n
t E t E n S （15） Posterior error ratio: S S C 12
=
(16)
Small error frequency: }{16745.0)(S E t E P P <-= (17)
According to C and P two indexes can test predict the precision of the model, and as shown in table 2.
CASE ANALYSIS
We use the southeast coastal province in China as the example in this paper.
The Province Early Warning Index System Structure of Road Passenger Transportation
We divided the province early warning index system of road passenger transportation into three parts which contains a target layer, a rule layer and an index layer,as shown in table 3.
The Alert Judgment of Province Road Passenger Transportation Early Warning
The Division of Road Passenger Transportation Early Warning Limited
Due to the different index attributes, we use a different index to take a different limit determination method in this paper, as shown in table 4.
T he Determination of Province Road Passenger Transportation Early Warning Index Weight
It is based on the grey relational grades of entropy values to determine the weight in this paper, and then achieve index weight, as shown in table 5.
The Alert Judgment of Province Road Passenger Transportation Early Warning from 1995 To 2006
1)The warning value calculation
Based on the limited warning,we put the index into adversity, as well as adversity and prosperity with the respective values of 1, 2, 3 points. Sum up the index weight with the entropy method calculation to get the industry development comprehensive warning degrees, as shown in the table 6.
2)The comprehensive warning degree of province road passenger transportation early warning development from 1995 to 2006
With the 40%, 60%of the score in a final three points for boundaries which was (1.2, 1.8) to be the warning limited to judge the road passenger development state yearly, the results are as follows.
The Province Road Passenger Industry Development Condition Analysis From 1995 To 2006
From 1995, the province road passenger industry of each warning index overall leaned towards a good trend of development. From single indicators, the industry degree, especially the high-grade and middle-grade car proportion, increased greatly and traffic accident frequency decreased from 2000. Compared to previous spans, we know that the province road passenger development states are good.
The Province Road Passenger Transportation Early Warning Model
Based On The Grey GM (1, 1) Model Road Passenger Transportation Early Warning
Based on the warning degree, we use the grey GM (1, 1) method to do the industry development early warning, and the models that are as follows:
[] better. is accuracy model the ,9167.0,2357.016745.004864
.01/2,1700.02,3495.014402.14)1(07488.0)1(?07488.0==*=====-?=+P S S S S C S S e X t X
The Forecast of Province Road Passenger Transportation Industry Development Situation
Based on the forecast of province road passenger transportation industry development situation by road passenger transportation early warning model from 2007 to 2010, the results are shown in table 8.
The warning conclusion is that the province highway passenger industry in the next four years will continue to keep a good development tendency as a result of the sustainable development of highway transportation, supply of quality improvement and passenger transportation enterprise and the continuous development toward the intensive and scale.
The Measures and Suggestions On The Early Warning System Of Road Passenger Transport Of One Province
1) According to planning of the state and the province, railway construction will be vigorously carried out in the next few years. High speed railway will greatly improve railway transport capacity. Speedup the integration of small and medium-sized enterprises to enlarged scale is not only good for reducing operating costs for enterprises, but also can provide resistance to common external risks. Enterprise group management is conducive to rectify the order of the passenger transport market,and the prevention of unfair competition. At the same time, it can improve the quality of service and promote the benign development of industry.
2) To improve the quality of service, form a variety of supply levels. To limit ordinary passenger car production, and encourage of the purchase of high, mid-range passenger cars in order will further increase the high, mid-range passenger proportion. In order to change from traditional development of quantity to quality, we need
the training for crew staff, accident accountability for drivers, and improvement on the quality and efficiency of service.
3) Further expand their business, and achieve diversification.To seek a new economic point of growth, and to achieve diversification can meet a variety of transportation demands and can evoke new passenger demands. Because of the characteristics of flexibility and convenience of highway passenger transportation, we can find new passenger lines, passenger transport, tourism chartered passengers and so on.
CONCLUSION
(1) Based on the study of our country’s hig hway management function and early warning theory, we use the early warning theory on the industry of highway passenger traffic management, which leads to a discussion of feasibility on highway passenger traffic warning management and puts forward the implementation method. It reaches the management means and management method of the industry of highway passenger traffic.
(2) The construction of highway passenger transport industry development early warning index system is based on the theories of systematic, scientific, and practical, which can reflect the development status of highway passenger transport industry comprehensively.
2)(3) By using an entropy method to determine the weight of index, and statistical methods combined with expert experience to divide the warning limit, we can improve the objectivity and practicability of the early warning system. Application of grey GM (1,1)theory to establish early warning model of highway passenger traffic. With trend prediction of highway passenger transport development, we put forward corresponding countermeasures aiming at the possible crisis.
(4) After the empirical study, the warning result basically tally with the actual situation, which shows that this highway passenger transport early warning model is effective and feasible, and can be as a guide and regulation of highway passenger transport industry to provide a scientific basis.
REFERENCES
Li,Jizun, (2007)“China Energy Warning and the Creative Warning Index”[J].China's Oil University Press(Natural science edition),2007.31(6)
Qin,Yongdong, O,Xiangjun, Zhen,Feng,(2008)“Based on the Method of Entropy Living Environment Quality Evaluation Research”[J].City Question,2008.10:19-24
3)Liang,Jun,Jiang,Wei, Li,Xuhong.(2002)“The Fuzzy Comprehensive Evaluation Method Im provement in Traffic Management and Planning in the Application”[J].Journalof Transportation Engineering, 2002,12,2(4)