Truck Overloading Study in Developing Countries and ...eprints.qut.edu.au/28561/1/Ying_Chan_Thesis.pdf · Conclusion and Recommendation 7.1 ... Traffic volume of truck at bypass route

Truck Overloading Study in Developing Countries and

Strategies to minimize its Impact

Ying Chuen Chan

Master of Engineering

2008

QUT

(i)

Table of Content

Acknowledgments viii

Abstract ix

Glossary of Terms x

1. Introduction

1.1 General 1

1.2 Background and history of overloading truck traffic 1

1.3 Research aim and scope 2

2. Statement of Problem

2.1 General 4

2.2 Economical loss 4

2.3 Pavement service life 4

2.4 Maintenance and rehabilitation cost 5

2.5 Summary 5

3. Literature review of the overloaded heavy vehicle

phenomenon

3.1 Introduction 6

3.2 Overloading of trucks in developed countries 7

3.3 Overloading of trucks in Developing countries 19

3.4 Scouting of the Fourth Power Rule 26

(ii)

3.5 The use of Weight-in-Motion 29

3.6 Truck Overloading and Safety 33

3.7 Concluding Discussion 34

4. Methodology

4.1 Data collection 36

4.1.1 General 36

4.1.2 Anhui Province 36

4.1.3 Road selection 37

4.1.4 Investigation schedule 40

4.1.5 Random sample 41

4.1.6 Site selection and WIM system 42

4.1.7 Preliminary analysis of raw data 45

4.1.8 Summary of data collection 49

4.2 Data Analysis 49

5. Overloaded truck traffic analysis

5.1 Methodology of data analysis 51

5.1.1 Classification of data 51

5.1.2 ESAL of each axle group 51

5.1.3 Design ESAL 52

5.2 Data analysis result 55

5.2.1 Class 4 truck 55


(iii)




5.2.6 ESAL analysis results comparison 64

5.2.7 Distribution of ESAL 66

5.3 Traffic component analysis 67

5.3.1 Standard ESAL of Anhui Province 67

5.3.2 Standard ESAL of Queensland 68

5.3.3 Actual ESAL of Anhui Province 68

5.4 Summary 69

6. Pavement service life analysis

6.1 Methodology of pavement service life analysis 71

6.1.1 Pavement design 71

6.1.2 Effect of overloading truck traffic 72

6.2 Pavement service life analysis result 73

6.2.1 Single truck traffic analysis 75

6.2.2 Case study for Highway G206 81

6.3 Summary 84

(iv)

7. Conclusion and Recommendation

7.1 Conclusion 85

7.2 Practice in developing and developed countries 86

7.3 Overloaded truck traffic control in Anhui 86

7.4 Recommendation 87

7.5 Directions for future research 87

Appendix

Appendix A

Standard ESAL of China and Queensland (Australia) 89

Appendix B

Accumulated Standard ESAL of China and Queensland 90

Appendix C

Detail Dataset 97

Appendix D

ESAL Data analysis 106

Appendix E

Accumulated ESAL of G206 Highway 128

Appendix F

Calculation of ESAL comparison of G206 131

Appendix G

Calculation of pavement service life of G206 133

Appendix H

Calculation of actual service Life 154

(v)

Appendix I

Calculation of net present value of investment 156

List of References 161

List of Figure

Figure 3.1(a) Traffic volume of truck at bypass route H51 10

Figure 3.1(b) Traffic volume of truck at bypass route Ehlen Road 11

Figure 3.2 Overweight violation rate versus enforcement level 13

Figure 3.3(a) Axle load percentage distribution of SU2 20

Figure 3.3(b) Axle load percentage distribution of SU3 20

Figure 3.3(c) Axle load percentage distribution of 2-S2 21

Figure 3.3(d) Axle load percentage distribution of 2-F2 21

Figure 3.4 (a-b) the site layout of WIM system 29

Figure 4.1 Location of Anhui province 38

Figure 4.2 Location of investigated highways 39

Figure 4.3(a) On site investigation by human 43

Figure 4.3(b) Trucks travels on the left lane (fast lane) of highway 43

Figure 4.3(c) Inspection station shows in this figure 44

Figure 4.3(d) The fixed WIM shows in the above figure 44

Figure 4.3(e) Mobile WIM was combined with fixed WIM in this survey

44

Figure 4.3(f) Truck was weighed by WIM statically 45

Figure 4.4 Flow chart of data analysis in chapter five and six 50

Figure 5.1 Exponential distribution result of class 4 truck 56

(vi)


Figure 5.3 Pearson5 distribution result of class 8 truck 59

Figure 5.4 LogLogistic distribution result of class 9 truck 61

Figure 5.5 LogLogistic distribution result of class 12 truck 63

Figure 6.1 pavement design system 71

Figure 6.2 typical pavement life-cycle performance curve 73






List of Table

Table3.1 Enforcement model parameter estimates 9

Table 3.2 Overweight violation rate across state agency (U.S.) 12

Table 3.3 the relationship between truck operator benefit and damage

caused for various levels of overloading and distances hauled 14

Table 3.4 Four typical trucks found in Anhui’s traffic 19

Table 4.1 Investigation implement schedule 41

Table 4.2: Summary of investigated trucks 46

Table 4.3 Traffic component of investigated road sections 48

Table 5.1 ESAL factor of Queensland standard 53

Table 5.2 ESAL factor of China standard 53

Table 5.3 Accumulated ESAL of China and Queensland Standard 54

(vii)

Table 5.4 Exponential distribution result of class 4 truck 55

Table 5.5 Exponential distribution result of class 5 truck 57

Table 5.6 Pearson5 distribution result of class 8 truck 58

Table 5.7 LogLogistic distribution result of class 9 truck 60

Table 5.8 LogLogistic distribution result of class 12 truck 63

Table 5.9 Comparison between standard and actual ESAL of each class

65

Table 5.10 Total Heavy vehicle ESAL simulated by data sample 66

Table 5.11 Traffic component of G206 in Anhui province in 2003 67

Table 5.12 ESAL Comparison of G206 in Anhui province in 2003 69

Table 5.13 Summary of ESAL analysis result 69

Table 6.1 Exponential distribution of service life of class 4 truck 75





Table 6.6 Mean service life for each dataset at different design 81

Table 6.7 Traffic component of G206 in Anhui province in 2003 82

Table 6.8 Comparison of actual and design service life 82

Table 6.9 comparison of NPV between actual and design service life

83

(viii)

Acknowledgments

I would like to acknowledge the support and guidance of my supervisor

team. Dr. Jonathon Bunker, my Principal supervisor in School of Urban

and Development and the support and direction of Prof. Arun Kumar,

Professor of Infrastructure Management.

Furthermore, I would like to thank Dr. Anthony Piyatrapoomi for teaching

me to use data analysis software in this research program.

I particularly wish to thank the staff of Anhui Province Communication

Department (APCD) and Southeast University (SEU) for the friendly

support and help to undertake this research program.

Finally, I would like to thank my parent for supporting me with the

opportunity and suitable environment to undertake this research program.

This research thesis is of interest to road pavement management and

economizer, and other concerned overloaded truck traffic problem in

developing countries.

(ix)

Abstract

Overloading truck traffic is an untenable problem around the world. The

occurrence of overloaded truck traffic can be evidence of rapid

development of an economy. Most of the developing countries emphasize

the development of economy, thus supporting reform of infrastructure is

limited. This research investigates the relationship between truck

overloading and the condition of road damage. The objective of this

research is to determine the amount of economic loss due to overloaded

truck traffic is. Axle load will be used to calculate the total ESAL to

pavement.

This study intends to provide perspective on the relationship between

change in axle load due to overloading and the resultant service life of

pavement. It can then be used in the estimation of pavement damage in

other developing countries facing the problem of truck overloading.

In conclusion, economical loss was found, which include reduction of

pavement life and increase in maintenance and rehabilitation (M&R) cost.

As a result, net present value (NPV) of pavement investment with

overloading truck traffic is higher than normal truck traffic.

(x)

Glossary of Terms

Overloaded

Truck traffic

Type of Vehicle

Equivalent Single

Axle (ESA)

Annual Average

Daily Traffic (AADT)

Pavement

Design Life

Pavement Service

Life

Net Present Value

One of the components of traffic is truck traffic.

When the weight of cargo on the truck exceeds

the legal limit, it is overloaded.

Vehicle class may refer to the thirteen-class

vehicle sorting system established by the

Federal Highway Administration (FHWA) of U.S.

The number of standard single axle loads which

are equivalent in damaging effect on a

pavement to a given vehicle or axle.

The total volume of vehicle traffic of a highway

or road

for a year divided by 365 days.

The expectation service period for a new

pavement. For flexible pavement of highway, it

is 15 to 20 years. For rigid pavement, it is 20 to

40 years.

The actual pavement service period. In the ideal

case, it is same as pavement service life, but it

may shorten by many reasons.

The difference between the present value of

cash inflows and the present value of cash

outflows. NPV is used in capital budgeting to

analyse the profitability of an investment or

project.

Truck overloading study in developing countries and strategies to minimise its impacts

-1-

1. Introduction

1.1 General

The road network plays an important role in any country’s transport and

communications. Pavement condition is one factor to assess the efficiency of

road network. Design life and bearing capacity of pavement are dependant

on the construction materials and the type of highway. Usually, pavement life

and bearing capacity of expressway and national highway are higher than for

a local street.

Design life of new flexible pavement is frequently fifteen years and rigid

pavement thirty years, which includes regular maintenance and rehabilitation

within its service period. Schedule of maintenance and rehabilitation plan is

according to the proportion of vehicle types in the traffic flow. Thus, Annual

Average Daily Traffic (AADT) of each class of vehicle is a key input to the

schedule of maintenance and rehabilitation plan.

However, the occurrence of overloading truck traffic induces incorrect

estimation in total Equivalent Single Axle Loads (ESALs), therefore the

frequency of maintenance and rehabilitation within the service period are

corrupted by overloaded truck traffic. Nevertheless, maintenance and

rehabilitation which are related to the economy of the country are provided in

the short run. Meanwhile, reconstruction of a new pavement would cause a

long run economic loss. Thus, overloaded truck traffic is an important

phenomenon.

1.2 Background and history of Overloading truck traffic

Overloading truck traffic is an untenable problem around the world. This

phenomenon not only occurs in developing countries, but also developed

countries. Nowadays, developed countries such as the U.S. and Australia

cannot eliminate overloaded truck traffic entirely. In developed countries,

less than 5 percent of truck traffic in the traffic stream is overloaded.


-2-

Extremely high enforcement and inspection are applied to ensure this.

However, overloaded truck traffic in developing countries is more serious

than developed countries as enforcement and inspection are not as

effective.

The occurrence of overloaded truck traffic can be evidence of rapid

development of an economy. Most of the developing countries emphasise

the development of economy, thus support to reform of infrastructure

management is limited. Such is the case in China, development of road

network system was overlooked during the 1980s. The economy of China

has grown rapidly since the 1990s, and freight demand has increased to the

same time. However, the road network cannot bear the huge growth of

freight demand therefore pavement was damaged by excess traffic.

Meanwhile, enterprises and truck drivers have tended to overload their

trucks, because they can reduce their running cost and overhead for freight

transport. Thus, the impacts to the whole country and society have tended to

be ignored.

1.3 Research aim and scope

This research investigates the relationship between truck overloading and

the condition of road damage. Anhui Province (China) is the case study for

this research, and the traffic data of Anhui Province will be analysed. The

objective of this research is to determine the amount of economic loss due to

overloaded truck traffic. Axle load will be used to calculate the total ESAL to

pavement, as a result it will be possible to determine actual pavement life.

The actual service life of pavement under the effect of overloaded truck

traffic can be use to analyses the economic loss in terms of construction,

maintenance and rehabilitation, as well as the social cost.

This study intends to provide perspective on the relationship between

change in axle load due to overloading and the resultant service life of


-3-

pavement. It can then be used in the estimation of pavement damage in

other developing countries facing the problem of truck overloading.


-4-

2. Statement of Problem

2.1 General

Impacts of overloaded truck traffic include economic, social and

environmental losses. Many developing countries are confronted with these

problems. Overloaded truck traffic induces extreme harm to the economy of

an entire country, thus economic impact always is the major concern for

Government.

2.2 Economic loss

The major economic impact induced by overloaded truck traffic is

unexpected expenditure on pavement investment. Because pavement

design is based on normal traffic load and total ESAL, the ESAL caused by

overloaded truck traffic is not the expected traffic load in pavement design.

As a result, the bearing capacity of pavement is lower than the actual

demand. Actual pavement service life therefore cannot reach the original

design life.

Pavement service life has a direct relationship with net present value of

investment. Construction cost for a new pavement is the most direct cost,

which occurs when pavement service life is reduced. On the other hand,

increase in annual maintenance and rehabilitation costs are the most evident

economic loss induced by overloaded truck traffic.

2.3 Pavement service life

The calculation of pavement service life is based on AADT and ESAL with

overloaded truck traffic. The case study investigated in this research is

Highway G206 of Anhui province and the AADT of 2003 is adopted.

According to the ESAL calculated from the dataset, the actual ESAL of each

vehicle class was found. After that, total ESAL with overloaded truck traffic

could be found. Comparison between ESAL with and without overloaded


-5-

truck traffic is the important factor to estimate the reduction in pavement

service life, because pavement service life is directly driven by traffic load.

2.4 Maintenance and rehabilitation cost

Determination of maintenance and rehabilitation (M&R) cost is the final

objective of this research. Calculations of ESAL of overloaded truck traffic

and pavement service life were the steps to work out the total M&R cost over

the service period. According to the information provided by Anhui province,

the annual budgeting M&R cost is different from actual M&R cost. Meanwhile,

the actual service life of pavement is also reduced by overloaded truck traffic.

Therefore, the calculation of net present value (NPV) of investment to the

pavement must be based on service life and M&R cost. Thus, the difference

in NPV of pavement investment with and without overloaded truck traffic

becomes the important indicator to determine the value of economic loss.

2.5 Summary

Economic loss may include many items. However, the major concern in this

research is M&R cost, because it is the most direct cost involved in

pavement management when overloaded truck traffic occurs. Estimation of

NPV of pavement investment involves the calculation of ESAL of overloaded

truck traffic and pavement service life. Thus, distributions of ESAL and

pavement service life are the analysis targets in this research.


-6-

3. Literature review of the overloaded heavy vehicle

phenomenon

3.1 Introduction

Overloaded truck traffic is a serious problem in many developing countries

because it incurs huge costs in terms of maintenance and rehabilitation of

damaged road networks. Overloaded truck traffic not only causes economic

loss but also safety and environmental problems. Many African and Asian

countries have been attempting to address this problem in recent years.

However, it is an inevitable feature of economic development and expansion.

This literature review examines the reasons, background and history of the

occurrence of overloading in truck transport. The review is based on journal

articles and relevant traffic reports around the world.

The literature review is structured as a review of overloading of trucks in

developing countries, and review of overloading in developing countries,

scrutiny of theory, scrutiny of weight measurement practice, review of safety

impacts of truck overloading, and concluding discussion.

In this review, three main points have been identified. First, overloaded truck

transport is an inevitable outcome of economic growth. Each developed

country faces this problem before their economic system becomes well

developed. Second, the problem of overloading cannot be totally eliminated

as evidenced by the fact that overloading also exists in the traffic systems of

developed countries like the U.S. and Canada. However, the overloading

percentage in developed countries is 2 to 5 percent, while in developing

countries it can reach as high as 80% of the total number of trucks on the

road. Finally, the only way to control this problem is through monitoring,

legislation and education. Thus, it is revealed that the developed countries

invest in road safety education continuously, whereas developing countries

require both funding and technology to achieve this.


-7-

3.2 Overloading of trucks in developed countries

Truck overloading in the U.S.A.

Strathman (2001) focused on the relationship between the economical effect

of overloading and weight enforcement. Strathman studied the economic

rationale of vehicle overloading activities when they face weight enforcement.

The study shows that both the intensity of weight enforcement and the level

of penalty can deter overloading activities. Also, the patterns of weight

enforcement practices by the government combine different level of

enforcement and penalties. As a result, the response of overloading

activities depends on these two factors.

The operating revenues and costs described by strathman (2001) in this

paper is the major study aim, and Equations (3.1) to (3.3) show the function

of net operating profit per mile to overloading vehicle.

whereWWcWfPWWr excessitexcessdexcessit ),(*)(*)(* limlim +−−+=π (3.1)

r = revenue per ton-mile;

Wlimit = the legal load limit, in tonnes;

Wexcess = the load in excess of the legal limit, in tonnes;

Pd = the probability per mile of detection by weight enforcement activity;

f(Wexcess) = the pentalty associated with overloading, which is defined to be a function

of the level of overloading;

c = operating costs per ton-mile.

orcWfPrW excessdexcess ,0)(/ ' =−−=∂∂π (3.2)

wherecWfPr excessd ,)(' += (3.3)

excessexcessexcess WWfWf ∂∂= /)()('

The net operating profit per mile to the overloading carrier

(Strathman 2001)

The Equation (3.1) shows that the intensity of weight enforcement activity ( )

and the severity of the marginal fine ( ) are the variables of the


-8-

expected penalty. As a result, the legislation can focus on these two factors

to reduce overloading activities.

Four weight enforcement regimes are studied by Strathman (2001), which

includes relatively small penalties with relatively extensive enforcement,

relatively low levels of enforcement with relatively high penalties, relatively

high penalties with relatively intensive enforcement and relatively small

penalties with relatively low levels of enforcement. Under these enforcement

regimes, two regression models were developed. Model 1 studies the total

number of weighings and Model 2 determines the different weighings at fixed

location and on portable/semi-portable scales.

Strathman (2001) found that fines can deter overloading activities but they

have relatively inelastic effects in Model 1. On the other hand, vehicle miles

travelled (VMT) have a relatively elastic relationship. In Table 3.1, the data

analysis results are shown. Model 2 shows that the result of fixed weighings

is relatively inelastic when compared with portable scale weighings. The

elasticity of portable scale is six times that of fixed location. The results of the

data analysis in this paper are reliable because it is similar to most previous

studies on overloading enforcement strategies.

Strathman (2001) concludes that fixed locations stations have low elasticity

to seize overloading vehicles. Thus, overloading vehicles can evade the

inspection easily. Also “the regression results indicate that the relative

consequences of emphasizing enforcement intensity or overweight penalties

are about the same in terms of deterring overloading activity” (Strathman

2001 ).


-9-

Table3.1 Enforcement Model Parameter Estimates (Strathman 2001)

(Dependent Variable=Ln Overweight Citations)

Variable Mean**

(St. Dev.) Model 1 Model 2

Ln fine $182.1

(138.8)

-0.286

(-2.00)

-0.238

(-1.56)

Ln Weighings total 2,081,400

(2,751,500)

0.259

(3.62) - -

Ln Weighings fixed 2,047,500

(2,753,000) - -

0.041

(1.80)

Ln Weighings

portable

33,877

(57,926) - -

0.251

(2.75)

Ln VMT (millions) 3,855.1

(3,755.2)

0.885

(6.23)

1.033

(7.79)

Ln Value per Ton $603.0

(221.4)

0.382

(1.11)

-0.019

(-0.05)

Constant - - -2.747

(-1.23)

-1.018

(-0.46)

R2 - - 0.78 0.76

n 48 48 48

* Coefficients in bold type are statistically significant at the 0.05 level

** Means and standard deviation are reported in nominal values.

In summary Strathman (2001) clearly shows that the vehicle overloading

enforcement strategy is highly correlated to fines and enforcement, and

elasticity depends on different weight enforcement regimes. As a result, fines

must be associated with intensified enforcement when considered in further

strategy recommendations.


-10-

A weight enforcement and evasion study for Oregon was carried out by

Strathman and Theisen (2002). They examined the incidence of overweight

trucks and their relationship to regulatory enforcement activity. The

inspection station on I-5 corridor had been closed for four months. Traffic

flow in the study area had been recorded before, during and after this period,

which raised questions of scale operations in relation to weight violations and

the effectiveness of enforcement levels, automated pre-clearance systems

and weigh-in-motion (WIM). The reason to carry out this study is that Oregon

Department of Transportation and Federal Highway Administration wanted

to find out about the low incidence of weight violation and its factors, which

may include the deterrent effect of enforcement activity and extensive scale

evasion.

Strathman and Theisen’s (2002) results show that that the traffic volume

data did not indicate evasion behaviour on the bypass routes, nor diversion

to I-5 WIM station during closure, also the evidence of diversion was limited.

After an initial downward shift in volume at closure, the traffic volume pattern

exhibited a continuous upward trend through closure and after reopen .Detail

of traffic volume pattern shown in Figure 3.1.

Figure 3.1(a) Traffic volume of truck at bypass route H51

(Strathman, Theisen)

halla

This figure is not available online. Please consult the hardcopy thesis available from the QUT Library


-11-

Figure 3.1(b) Traffic volume of truck at bypass route Ehlen Road

(Strathman, Theisen)

Although the variation of traffic pattern was limited, the overloaded vehicle

load increased from 2.27% before closure to 3.67% during closure, and then

the overloaded vehicle decreased to 3.19% after the scale reopened. The

total change of the overloaded vehicle is a gain of 40.5%.

This study shows that the variation in the incidence of overloading observed

in this study is insignificant. The possible reason for this is that Oregon’s

weight enforcement is more aggressive than other states having more

weighing stations and stiffer fines for overweight violations. The results

indicated that aggressive weight enforcement is the most efficient way to

control overloaded trucks in the long-run. After building up the reputation of

enforcement, a temporary suspension of weighing activity could be an

incentive to change overloading practices.

In summary, the results of Strathman and Theisen (2002) show that if a

country has a well developed highway monitoring system, the overloading

behaviour of truck operators are not affected by closure of inspection

stations, which means truck operators are highly educated and are less likely

to divert their route. This result can be a good model for developing countries,

because most of the developing countries experience truck diversions when

road inspections are carried out.

Enforcement and overweight violation

halla



-12-

Taylor et. al. (2000) demonstrate the demand, cost effectiveness, and

enforcement of commercial vehicle weights and dimensions regulations.

They show that effective weight enforcement, which cooperates with a

comprehensive data collection program and forms the database for a

scientifically based road asset management system. Table 3.2 and Figure

3.2 summarize the general functional form between enforcement visibility

and overweight violation rate based on several studies performed by seven

state enforcement agencies in the U.S.:

Table 3.2 Overweight Violation Rate across State Agency (U.S.)

(Taylor et. al., 2000)

State High Enforcement Level

Violation Rate

Low Enforcement Level

Violation Rate

Virginia(2*) 0.5 to2.0% 12 to 27%

Maryland(2) 1.0% 34%

Arizona(2) 1.5% 30%

Wisconsin(3) 1.0% 20%

Idaho(4) 11.9% 32%

Florida(5) 1.4% 13%

Montana(6) 1.0% 29%

*Number represents number of weigh stations included in study.


-13-

Figure 3.2 Overweight Violation Rate versus Enforcement Level


Table 3.2 and Figure 3.2 show the relationship between overloading and

enforcement level clearly, which forms an inverse proportional relationship.

Moreover, because of the exponential geometric relationship between

vehicle weight and pavement damage, the effect indicates in a power

relationship. The Fourth Power Rule was used in the past, but a higher

power relationship may be considered in pavement damage predictions.

Taylor et. al. (2000) indicate that when the average overload on a truck was

12% in excess of the legal weight, it can cause 57% extra damage of the

original truck weight when the traditional fourth power rule is applied.

Taylor et. al. (2000) shows the overloading cost in additional road damage,

according to the federally funded study undertaken in the United States in

1990, which indicated that overloaded truck axles cost between U.S. $160

million and $670 million per year in pavement damage. The interstate system

deteriorated fifty percent faster than it could be replaced due to a number of

factors, one of which was overloaded trucks.

Table 3.3 shows the economic benefit to the truck operator, compared to the

additional pavement damage caused for various levels of overloading and

distances hauled.

halla



-14-

Table 3.3 the relationship between truck operator benefit and damage caused for various

levels of overloading and distances hauled


As a result of the Taylor et. al. (2000) study, it can be seen that the impact of

commercial vehicle overloading has on pavement damage and safety is real

and of a considerable magnitude. Taxpayers and conscientious truck

operators pay directly for overweight violations of the law. Truck operations

that are overweight are likely to be safety deficient as well.

In summary Taylor et. al. (2000) was found that the relationship between

enforcement level and vehicle overloading rate is relatively inelastic in the

initial stage but elastic at the end (Figure 3.2), which means enforcement has

higher efficiency at the initial stage. However efficiency decreases rapidly

when enforcement levels increase gradually. Thus, the balance between the

level of enforcement and efficiency of enforcement must be considered in

further studies.

halla

This table is not available online. Please consult the hardcopy thesis available from the QUT Library


-15-

Enforcement action of the U.S.

Evidently enforcement can help to eliminate overloading heavy trucks.

According to the TRB (1990) Study “Truck Weight Limits: Issues and Option”

enforcement is a critical factor to control vehicle weight. Adequate

enforcement can act as deterrent by declaring that those travelling in

disregard of laws and regulations would be apprehended and would face

effective publishment.

Battelle Team (1995) carried out an overloading study in the U.S. Different

states had their own study ways, coincidently similar results were founded.

The intensity of enforcement has an inverse proportional relationship with

number of overloading trucks; meanwhile, strict enforcement induces

diversion as well. Driver behaviours and overloading activities are driven by

intensity of enforcement, which was show clearly in the previous studies.

Based on the previous studies, the author listed the important elements

which relate to overloading travel:

1. Static scales and weigh station personnel;

2. Portable/semi-portable scales and personnel;

3. Weigh-in-motion (WIM), automatic vehicle identification (AVI), and

automatic vehicle classification (AVC) equipment;

4. Degree to which WIM readings are consistent with static scale

readings;

5. Relevant evidence laws and audit information;

6. Judicial system and culpability (driver, vehicle owner, shipper);

7. Fine, penalties, sanctions; and

8. Potential for self-certification.

In general, Battelle Team (1995) shows a clear picture that how to carry out

enforcement successfully. First of all, mixed approaches using WIM are most

cost effective at than unitary approach, because use of portable enforcement

for bypass routes was found to be very promising in enhancing the

apprehension of overloading trucks and in deterring overloading travel.

Second, AVI or AVC system must fully develop; because WIM cannot be


-16-

used to record evidence of violations thus AVI or AVC systems is taking an

important role when law executor need to prosecute violator.

Third, relevant evidence laws and audits must define clearly. If a country has

vast territory, states may have different attitude and regulation to define

overloading violation, thus compatible and mutual concept must be

developed.

Finally, effective punishment system is the major component to stop

overloading activity, which includes fines, penalties and sanctions. If fines

and penalties are sufficient, then the overloading trucker has low intensity to

prosecute.

In summary, the effective means of managing truck overloading is not unitary.

It must combine monitoring, inspection, enforcement, and punishment; as a

result, a complete system can be developed.

Road efficiency and damage

Road maintenance is a remarkable public cost in most of countries. Certainly

freight traffic is the major cause of road damage, and it is the reason that

most of the researches focus on heavy truck. Increase in weight limit of truck

is the major concern of Levins and Ockwell (2000).

Increase in weight limit is a controversial issue. Insufficient capacity is one of

the reasons for vehicles overloading. According to Organization of Economic

Cooperation and Development (OECD), many countries have significant

growth rate in freight travels between 1980 and mid 90s, which countries

include Australia (119%), Korea (288%) and Turkey (229%).

Normally, OECD countries are well developed and have complete road

network system, thus they are not persecuted by overloading problem, but

they still need to face the enhancement in freight demand. Based on the

standpoint of OECD, the best option is increase in heavy vehicle mass limits

for vehicles using advanced suspension technology without causing an


-17-

increase an increase in road costs. Increase in weight limit not only can help

to increase the efficiency of truck, but also can decline transport costs.

OECD suggested a reasonable solution to face the increase in freight

demand. The author believes that the potential economic benefits are

generally double. Introduced suspension technology can help to reduce

maintenance and rehabilitation costs of road network because of the decline

in truck weight. OECD estimates that road-friendly suspension could

increase pavement life around 15 to 60 % depending on the type of

pavement.

Also, transport efficiency would be improved without extra pavement

damage after the increase in weight limit. As a result, higher mass limits can

be introduced on the highway system for vehicles fitted with road-friendly

suspensions, thus unit cost of freight can decline to a lower level.

Road-friendly suspension systems not only benefited in government, but

also transport operators. Levins and Ockwell (2000) shows that new

technology can help to release the pressure in high demand of freight. It is a

good option for developing countries when they face high freight demand,

because high freight demand is one of the significant reasons for truck

overloading. Thus they can consider this practice to reduce the pressure of

freight demand; as a result, the overloading problem can be reduced to a

certain degree.

Truck configuration and pavement damage

Since the 1980s, U.S. pavement engineers have been concerned with the

truck-tire configuration, tire types and pressure because of the potential of

pavement damage. In 1992, C.A. Bell, S.U. Randhawa, and Z.K. Xu studied

the “Impact of High-pressures tires and single-tired axles in Oregon”

indicated that tire pressure and use of single tries has no significant change

since 1986. An Oregon based literature review of single-tired axles and

pressure was involved in this study, which data were collected from five


-18-

highways entries in different months of 1992 thus they can identify the truck

components of Oregon highways.

This study proved that type of axles group, type of tire and pressure are

major components to determinate pavement damage. There is no doubt that

traffic loads can cause pavement damage. Damage is normally is

determined by the total contact area between tire and pavement, which

means larger contact area, less damage to pavement.

The analysis result shown that damage can be reduced by the increase of

axles. “Tridem axles with wide-base tires can carry 42,000lb and have a

lower damage potential than a tandem axle loaded to 34,000lb.”(U.S. Road

Engineering Journal 1997). The study also considered trucks with seven and

eight axles using wide-base single tire and regular dual tires on single axles.

The results are similar to the previous case; trucks with more axles cause

less damage.

The study result shows that three main criteria drive the level of pavement

damage. The first is number of axles; trucks with more axles cause less

damage than fewer axles. The second is number of tires; when trucks have

the same axle groups, dual tires cause less damage than single tire. The

third is types of tire; less damage is caused when trucks use wide-base tires.

TranSafety, Inc. U.S. (1997) summarizes that their study can help to

establish and modify truck standards because it clearly indicates that

physical configurations of truck influence the level of pavement damage,

thus when the relevant government department wants to eliminate damage

caused by overloading of trucks, modification of truck mechanism is one of

the criteria that needs to be considered.


-19-

3.3 Overloading of trucks in developing countries

The extent of Highway trucks overloading in Anhui, China

Hang et.al. (2005) respect that rapid deterioration of the Anhui province

(China) highway pavement is a very common and serious problem, which

has occurred early this century. In 2004, Southeast University and Nanjing

Normal University collaborated to survey and evaluate the overloading status

quo and enforcement efficiency with the support of the World Bank. Surveys,

which were carried out in six cities, had four major concerns: traffic volume,

axle load, freight information and registration information.

This survey is concerned with the overloading characteristics of the common

truck types in Anhui province, so the invalid data was eliminated from the raw

data. Thus, the data used for data analysis only included the vehicle class 4,

5, 8, 12, because these four classes are the major components of truck

traffic in Anhui province (Hang et. al. 2005). The thirteen-class vehicle

sorting system established by the Federal Highway Administration (FHWA)

of the U.S. was used. Basic information about these four types of trucks is

shown in table 3.4:

Table 3.4 Four typical trucks found in Anhui’s traffic.

Class Description Figure Abbreviation GVW (Mg)

4 Two-axle, six tire,

single-unit vehicles

SU2 16.0

5 Three-axle,

single-unit vehicles

SU3 24.0

8 Four-axle,

single-trailer trucks

2-S2 34.0

12 Four-axle,

multi-trailer truck

2-F2 36.0


-20-

Hang et. al. (2005) found from the statistical analysis that vehicle overloading

is universal and serious in the arterial highways of Anhui province. As a

result, the traffic load greatly exceeds the standard bearing capacity of the

pavement which causes wide premature pavement damage, especially on

rigid pavements. The research shows that vehicle overloading occurs with

the four types of trucks mentioned above. Class 5 (SU3) has the most

serious overloading problem among these four types, while gross vehicle

weight (GVW), front axle and tandem axle have highest overloading

proportion.

Another important fact found in this investigation is that the mean GVW

values of all loaded truck configurations exceeded the GVW limits, which

means a large proportion of trucks are overloaded (Hang et. al. 2005). In

Figure 3.3(a)-(d), the relationship of axle weight and axle load percentage of

the studied truck types is shown.

Figure 3.3(a) Axle load Percentage Distribution of SU2 (Mg=Tonnes)


-21-

Figure 3.3(b) Axle load Percentage Distribution of SU3 (Mg= Tonnes)

Figure 3.3(c) Axle load Percentage Distribution of 2-S2 (Mg= Tonnes)

Figure 3.3(d) Axle load Percentage Distribution of 2-F2 (Mg= Tonnes)

(Hang, et, al. 2005)


-22-

These distribution curves show the loading proportion of each truck type

according to the axle type, thus the amount of illegal overloading can be

seen clearly in each curve. These curves show the same result as the data

mentioned before, in particular that class 5(SU3) has a double peak in the

statistical analysis. This is a phenomenon which requires further research.

The social environment and freight operator detail is also determined by

Hang et. al. (2005), which indicates that the truck overloading problem is not

a simple problem in China (Anhui Province). Detailed information is shown in

section 3.4 and 4 of this research, which includes operating characteristics

and estimation of enforcement.

Hang et. al. (2005) argue that the truck overloading problem embodies

human behaviours, social development and geographical problems. Most

truck overloading problems occur in developing countries because of

inefficient highway management systems. When a society is developing, a

comprehensive road network is necessary. At the same time, highway

management and monitoring systems must cooperate; otherwise truck

overloading problem will accompany economic growth.

Overloading Issues in South Africa

The overloading issue in South Africa (S.A.) has been discussed by the

Railway Association S.A. (2001). Overloading issues have been considered

since 1977. Between 1977 and 1996, the gross combination mass increased

from 38 to 56 metric tonnes, and the payloads increased from 24 to 45

tonnes. While the road freight industry operators have been granted these

significant concessions, the incidence of overloading has increased.

In 1998, the Automobile Association report revealed that the overloading of

heavy vehicles had caused South Africa about R 500-million (USD 90

million) damage a year in terms of road maintenance. Furthermore, there is

an estimated backlog of R 20-billion (USD 3.6 billion) required for road

repairs arising from the combined effects of extreme climate factors coupled

with overloading practices.


-23-

Railway Association S.A. (2001) indicated that transport operators,

consignors, consignees and drivers are seeking the cheapest transport rate,

regardless of the overloading impact causing damage on pavements. Many

operators try to make an honest living but through circumstances and

economic pressures, end up with illegal loads.

According to the Council for Scientific and Industrial Research (CSIR),

"Every time a vehicle passes over a road it causes a small amount of

damage. This damage is not visible after a single passage but after many

hundreds and thousands of passages, the road becomes uneven, cracks

appear on the surface and ruts form in the wheel-paths where vehicles

normally travel." (Railroad Association of South Africa 2001)

In discussion, South Africa’s situation is similar to China. Shippers are highly

dependent on rail and highway transportation and because of economic

development, most of the under-developed regions have been pushed into

the economic stage. Freights are overloaded to achieve higher cost-benefits

to individuals but ignore the disadvantages to the public. Therefore economic

pressure is the major reason for heavy vehicle overloading, so the

responsibility of overloading lies not only with freights companies but also the

customers and government.

Railway Association S.A. (2001) summarizes the development of vehicle

load regulation and vehicle overloading history in South Africa. The legal

loading of vehicles has increased gradually in the last two decades, but the

vehicle overloading problem in South Africa has not been solved by the

increase in load size. Road damage and maintenance costs in South Africa

increase annually, which is caused by overloaded vehicles. Thus, South

Africa is highly concerned by this problem and has decided to target the

overloaded vehicles. South Africa has a similar situation to China, with the

economy and society developing in the last decade and development will

continue during the next couple of decades. Thus there are great demands

on highway networks and infrastructure management. They are facing the


-24-

same problem, whereby economic growth is faster than infrastructure

management development. As a result, the highway network is inadequately

managed and this is the major reason road users overload their trucks.

Overloading issues in Thailand

In Thailand, the problem of overloaded trucks is a remarkable social issue.

According to the statistics in 1996, overloading usually occurs on ten-wheel

trucks (class 5, three-axle single-unit vehicles) and it is occupies around 25%

of all trucks combined, which carry 78% of the shipment by weight (Worsak

2005). Based on the statistics collected by weight-in-motion (WIM), they

found that 33% of ten-wheel trucks are overloaded. 94% of these weighed

between 21 tonnes and 30 tonnes, while 21 tonnes is the legal weight for a

ten-wheel truck (axle weight is 8.2 tonnes). Worsak (2005) indicated that

81% of total damage to the highway is caused by 33% of overloaded trucks.

Thailand’s overloaded truck problem is very common in other Asian and

African countries. Highways are common properties for taxpayer thus they

must be designed, constructed and used appropriately to guarantee safety

and cost-effectiveness. Indeed, the occurrence of overloaded trucks

accelerates the deterioration of pavement structure. As a result, the

authorities need to work out suitable solutions for this problem.

Worsak (2005) indicated that two basic issues need to be considered. The

first consideration is the conflict between engineering principles and the

common weight of trucks. The second consideration is that the designed

axles load for is 8.2 tonnes, but the statistics show that most ten wheel

trucks’ ton axle load is 12.1 tonnes. If the official axle load rose from 8.2

tonnes to 12.1 tonnes, then the structure of existing and future highways

must be overhauled and designed carefully.

Three alternative options were suggested by Worsak (2005):

� Government does nothing to maintain the status quo. This option would

be heavy burden for government; meanwhile the economic loss would


-25-

transfer to taxpayers which would be unfair for the taxpayers who use

the highways appropriately.

� Governments need to reconfirm the current highway standards but

restore strict law enforcement. This option is carried out in many

countries because it is the most efficient method to eliminate overloaded

trucks and save road repairing costs. If this option is executed, existing

measures must be adjusted to a suitable level. This is also a chance for

other transportation modes to compete with truck transport, because the

freight industry in Thailand is highly dependant on truck transport as

89% of freight depends on truck transport. Thus, this is an opportunity

for the government to encourage the shift from truck transport to other

transportation modes.

� Raise the highway standard and upgrade existing roads and bridges.

The suggestion would involve a long term project for the government

because this would involve many significant factors, which include

technical feasibility, overhauling costs of roads and bridges, transport

economy, road safety, social justice, fair trade, etc. This paper also

indicated that problems may still exist after the load-bearing capacity of

the road is upgraded. Indeed, it may be unfair to normal taxpayers

because they are not overloading their vehicles.

So, raising the highway standard is the most suitable option for Thailand in

the ideal situation. However governments must consider that if the unit cost

is reduced by the saving of unlawful expenses more than the additional cost

for highway standard enhancement, then the third option is a win-win

situation for all parties. However, if the contrary situation occurs, the second

option will become the best for all parties and promotion for other

transportation mode should be carried out.

This article shows the specific status quo of Thailand, but the suggestions by

Prof. Worsak can be considered by other countries which have an

overloading truck problem.


-26-

Normally, if a country has this overloading problem, highways are damaged

prematurely. Thus, repair or reconstruction is the unavoidable outcome for

them. Authorities can combine the second and third suggestions to work out

the most suitable solution. Because of the high demand on transportation,

adequate highways are very important for economic growth. However,

highways cannot be upgraded, because this will encourage illegal

overloading behaviour if authorities concede easily. Thus the balance

between legal enforcement and raising the highway standard is the most

suitable solution to reduce the overloading problem and provide the best

conditions for the road user.

3.4 Scrutiny of the Theory

Scrutiny of the Fourth Power Rule

The Fourth Power Rule has an important role in road damage and cost

estimation, which includes toll charge, annual budget for maintenance and

user cost design. Johnsson (2004) investigated the Fourth Power Rule in a

computable general equilibrium model of Sweden, whose study scope was

the effect on road wear and deformation of alternatives to the Fourth Power

Rule. In this investigation, first to fifth powers were considered in order to

compare the results of how lower and higher powers related to the fourth.

The result of this investigation shows a significant increase in the activity of

the least damaging truck category when a larger power (i.e. fifth power) was

used. This result is reasonable; because the damage effect is according to

the power relationship which means that a larger exponent may cause the

obvious damage effect.

Johnsson (2004) also discusses the relationship between government

budgets, tax revenue and the power rule, which indicates that a higher

exponent can cause increases in tax revenue and help to decrease road

wear, and this outcome also complies with the theory. This is because a

larger damage effect may result from higher exponents in the power rule,


-27-

since when the tax and charge are calculated by the larger damage effect, it

is reasonable to obtain higher tax revenue. On the other hand, the tolerance

of damage activities caused by overweight truck would decline. Thus, the

road wear may simultaneously decrease.

In conclusion, choosing the wrong power resulted in a deviation from the

annual road wear cost.

In discussion, Johnsson (2004) investigated the relationship between the

wrong power in the fourth power rule and road wear cost. It is one of the

concerns of study in Overloaded vehicle study in Anhui province (China),

because the maintenance and rehabilitation cost of Anhui province does not

comply with the theoretical calculation. This means that an inaccurate

estimation of road damage may be one reason for this result. Also, the toll

charge collected from road users is always insufficient for road maintenance,

because of many reasons, an underestimation of road usage or traffic load

being the major reason. Thus the Fourth Power Rule has an important role in

the situation of Anhui and it was adopted to calculate the ESALs of truck in

chapter 5, because it is the most reasonable value in general.


-28-

3.5 Scrutiny of Weight Measurement Practice

Prevention of highway infrastructure damage through commercial

vehicle weight enforcement in India

Kishore and Klashinsky (2000) This paper discusses the imperative reason

for weight enforcement in India. Benefits arising from enforcement and the

new technologies that ensure effective enforcement without causing an effect

on the regular traffic flow are also discussed. Commercial vehicles transport

goods valuing billions of rupees across India and this value is growing

annually. Thus the Indian Government has introduced the Intelligent

Transportation Systems (ITS) to try to modify the existing infrastructure

management system.

Weigh in Motion (WIM) is the major technology which is used in traffic

monitoring. The reason for using WIM is that it can help to record the

condition of Infrastructure, increase safety, save taxpayers’ money, intimidate

overloaded vehicles and weigh all trucks to collect valuable data. This data

can be used to prepare for increases in traffic volume, minimize delays for the

trucking industry, use in pavement design and road management, and protect

the environment as well.

Kishore and Klashinsky (2000) show a brief plan of integrated WIM at toll

collection sites. The Indian government has combined the fixed and mobile

facilities to provide the best overall weight enforcement program. Past

observations indicate that a lower probability of being caught for overweight

infractions significantly discourages overloading. It also shows the operation

of WIM systems on the mainline or within the area of the weigh station. The

detail of the WIM site layout is shown in Figures 3.4.


-29-

Figure 3.4(a) Close up of individual Axle Sensor

(Kishore, Klashinsky, 2000)

Figure 3.4(b) Overview of inspection station

(Kishore, Klashinsky, 2000)

The in-road inductive loops have been used to catch suspect violators and an

additional console alarm will be triggered for vehicles failing to follow the

automated control signals. On secondary roads and remote areas, mobile

crews can utilize portable wheel load weighing equipment and mobile

communications systems to provide enforcement.

Kishore and Klashinsky (2000) show that an integral part of highway

management activities is the weighing of vehicles, which can help to protect

the infrastructure from premature wear and deterioration. It is also a major

component for stopping overweight trucks from damaging road structure. The

India Government’s aim in using WiM is to introduce cost-effective strategies

to minimize unnecessary expenditure.

Also, the facts show that fixed inspection stations have less elasticity to catch

illegal overloading vehicles, which may refer to other previous studies in

halla


halla



-30-

overloading enforcement. Thus, combining this with portable weighing is

necessary for a complete monitoring system.

According to Kishore and Klashinsky (2000), WIM technology weighs trucks

dynamically and reduces the overloading of highways by targeting

overweight vehicles. This helps to protect the public’s investment in

infrastructure, reduces stress on highway budgets and makes vehicles safer

by reducing rollovers, unbalanced loads, and various hazards associated

with heavy loads.

In summary, Kishore and Klashinsky (2000) identified that India is facing the

same problem as many developing countries, which is that its economic

growth is faster than the road management system. Thus, India uses the ITS

to try to make better use of existing infrastructure. There is no doubt that

WIM can help to collect useful information for further road design and

management. The authors can reasonably assume that when two types of

WIM are used together, each can redeem the weakness of the other.

However, the lack of adequate management means there will always be

weakness in the monitoring system, thus WIM works inefficiently in such

areas.

The use of Weigh-in-Motion

The Weigh-in-Motion (WIM) method is taking an important role in pavement

design. Where road assessment is concerned, WIM is the major tool to

collect pavement and traffic statistics. In 2001, Scott Wilson Pavement

Engineer Ltd studied the importance of WIM data in pavement design

(Hakim and Thom 2001). The study concerns the difference between the

actual vehicle wear factors (VWFs) and the pavement design VWFs. The

calculation of VWFs is based on the power rules and the researchers found

that the variability is quite high between actual and design VWFs. Once

again, the Fourth Power Rule is typically used to calculate an equivalent

number of standard axles.


-31-

In general, VWFs can be calculated by the following equation:

∑=

=

aN

i std

i

P

PVWF

1

4

(B. AI Hakim, A.C. Collop, N.H. Thom)

Where Pi is the force on axle i in kN,

Pstd is 80kN, and

Na is the number of axles on the vehicle.

In the UK, pavement design procedures have for a range of materials, thus

the pavement design may have up to 50 mm variation when different

exponents are used. The analysis of the calibration data from the WIM sites

has enabled estimates to be induced using the following WIM error:

1. Calibrations drift errors (5-59 percentile range of 30%, standard

deviation of between 10 and 11%)

2. Sensor error plus dynamic effect (average value of 11%)

The WIM errors presented in the traffic prediction are likely to over-predict

the traffic by 15-20% generally, and the maximum case is 40-50%. The WIM

error is the major variable of traffic prediction, and is driven by the level of

power in direct proportion. The effect of traffic prediction errors on pavement

design thickness was investigated as well. For the case study in this

research, Hakim and Thom (2001) found that 20% over-prediction in traffic

results in between a 5 and 15mm over-design of pavement thickness, while

a 40% error results in only 10-20 mm additional thickness.

It is impossible to gauge how much the calibrations error varies between the

investigation period because of insufficient information. However the results

show that the variation between the VWFs determined from the WIM

installations under investigation is thought to be largely due to calibration

error.


-32-

In summary, Hakim and Thom (2001) clearly shows that WIM has an

important role in pavement design and traffic prediction. Over-prediction

existed in this case study, thus over-design is the result of this error. As a

result, the expenditure on pavement construction may increase. However, it

is not wasted on pavement management, because a superior design can

reduce rehabilitation costs in the future, thus the pavement life can be

extended. However this becomes another issue in pavement management.

For this reason, precise traffic predictions can help pavement management

to be implemented simply and efficiently.

WIM in Urban Environment

Increases in structural depreciation of infrastructure assets are caused by

increases in commercial vehicle operations and truck loadings. The most

efficient way to maximize the life of road and bridge structures is by

monitoring and quantifying commercial truck loadings and reducing the

incentive of overloading.

Bushman et. al. (2003) shows the monitoring system in Saskatoon (Canada)

which includes weigh-in-Motion (WIM) and video surveillance system, data

collection of commercial vehicle was mentioned in this paper as well. WIM

system with video capture function is an efficient method of measuring the

capacity of commercial vehicle traffic types and volumes. The best option is

by monitoring the road 24 hours per day and through the week. The

percentage of truck overloads and severity of overloading are the major

factors to collect.

The analysis showed that the overloading is usually concentrated in

particular types of truck, which means further studies or inspections can

focus on particular trucks but not any kind of truck. This will make it more

efficient and the results can be refined as well.

The Video WIM Technology was installed at the study site, a load cell scale

in the right lane, where the majority of the traffic is expected to travel, and a

quartz axle sensor array in the left lane. This installation is very common in


-33-

most developed countries. WIM systems still play an important role in

developed countries because they can help to improve the effectiveness of

weight enforcement in an urban environment. WIM not only can be used to

enhance enforcement, but also can be a part of planning and evaluating

enforcement strategies.

The analysis showed that the greatest contributor of excessive loading were

the two and three axle vehicles, which are typically local service vehicles.

Also, the largest amount of overloading occurred during the workday hours

which are the time when local trucks would be expected to be most active.

Thus, the result of this analysis suggests that more attention should be paid

to two or three axle trucks.

In summary, the WIM and video capture system used together can help data

collection and understanding in the problem of overloading and therefore the

result can also act as part of the solution. Appropriate application of

equipment can increase the efficiency of enforcement activities, which

includes developing a strategy of when, where and who to enforce,

screening and identification of most likely violators in real time, identification

of repeat offenders, and evaluation of the effectiveness of enforcement

efforts.

3.6 Safety Impacts of Truck Overloading

Squires (2004) reported an authentic case study of the relationship between

truck inspection and fatal crashes. This study continued for three years from

1997 to 1999. During the three years period, fatal crashes involving large

trucks increased from 27 in 1997 to 41 in 1999. The result indicates that

when inspection activity declined, fatal crashes involving large trucks

increased at the same time.

This relationship indicates a significant fact that overloading trucks are taking

an important role in road scrutinize safety. Increase in overloading trucks is

caused by truck inspection decline in this case. Squires (2004) cited that

reduction in braking ability and stability is occurred in overloading truck. Thus


-34-

the relationship between truck inspection decline and road safety can be

cited that trucker trends to overload, and then the braking ability and stability

of truck is reduced. As a result, fatal accident increases.

This study showed that adequate inspection on truck is an effective way to

discourage overloading activity. The U.S. established an anti-overloading

system over the past two decades, and the system is being modified

continuously. But small group of people still have the incentive to

overloading.

In summary, two main conclusions have been found from Squires (2004).

The first is truck inspection is the effective way to eliminate overloading. The

second is that overloading trucks is inevitable, no matter whether in

developed or developing countries. When the amount of fatal crashes which

heavy trucks are involved in is correlated to the amount of overloading, we

find that fatal crashes involving large truck increased by 52%. This implies

that the number of overloading trucks also increased.

3.7 Concluding Discussion

The accumulating body of literature on overloading of trucks was reviewed in

this chapter, focusing on the overloading status in developed and developing

countries. The overall aims were to review the extent of overloading traffic in

Anhui Province (China); the relationship between enforcement intensity and

tendency of overloading truck activity from international experience; and to

identify the importance of road management in overloading truck elimination.

Section 3.2 and 3.3 show the different in attitude of developing and

developed countries. Developing countries more concern on monitoring

overloading truck traffic rather than eliminate this phenomenon. In the

meanwhile, developed countries have complete legal system and

management system, thus they are more concerned about the automaticity

of the road user.


-35-

WIM and Fourth Power Rule are taking the important roles in axle load

analysis. The summary of section 3.4 and 3.5 show the function of WIM in

monitoring and the importance of determine the factors of Fourth Power Rule.

Finally, safety is another major concern in overloading study. Unexpected

defect of road and damage of vehicle will cause fatal crashes definitely.

In summary, overloading truck traffic is concerned around the world in

different manner and level, and it is not an unsolvable problem. Most

developed countries have invested many resources to control this problem

and modifying the road management system to eliminate it continuously.

Thus developing countries must identify their unique situation and problems

in the overloading truck traffic and the most appropriate strategy may be

established.


-36-

4. Methodology

This section describes the methodology used in this research, it mainly

divided into two parts, the first part is data collection and the second part is

data analysis.

4.1 Data collection

4.1.1 General

This research is a study of the overloading problem of Anhui province. Traffic

data collection was conducted by Southeast University (SEU) and Anhui

Province Communication Department (APCD) on behalf of this study.

In 2004, APCD undertook an overloaded truck traffic survey in Anhui

Province supported by World Bank Group. Professor Arun Kumar was the

consultant of this project, and the data which was collected by SEU was

adopted in this research. In our cooperation, Australian standards and

knowledge were offered to SEU, and at the same time, SEU provided data

and information to us. In the procedure of data collection, they combined the

use of fixed and mobile weigh-in-motion (WIM). Inspection stations were

setup in front of toll stations of each investigation highway, thus investigation

activities would not affect the traffic flow seriously.

4.1.2 Anhui Province:

Six sites were investigated in this study, these sites distributed throughout

Anhui province. Anhui province is located at the middle east of China and

closed to Yangtze River, the economy mainly being primary and secondary

industries, also the major products construction material and mineral

products. In the longitudinal direction, Anhui is about 570km and 450 km in

the transverse direction, the total area of Anhui province is 139,600km2.

Because of the geographical features, Anhui province is the major transit

depot in longitudinal and transverse direction of wide range of China,


-37-

therefore it takes an important role in the west development stratagem. The

location of Anhui is shown in Figure 4-1.

4.1.3 Road selection:

In the overloading traffic survey, six roads corridors were selected for

investigation; five of them run from south to north and one runs from east to

west. Three of the selected highways are first class, two of them are second

class highway and the last one is a third class highway. Even though first

class highways are major concern of this study, for comparison purposes,

second and third class highways were also studied.

The details of investigated highways are shown in Figure 4.1. The shadowed

part is the range of Anhui province. Investigated road sections are showed in

bold lines and the major cities are showed as spots. The selected highways

include:

1. Fuyang-Yingshang (G105)

2. Huainan-Hefei (G206)

3. Liuan-Hefei (G312)

4. Hefei-Anqing

5. Wuhu-Xuancheng

6. Xiuning-Jingdezhen


-38-

Figure 4.1 Location of Anhui province

Anhui

Beijing


-39-

Figure 4.2 Location of investigated highways

The location of each investigation road shown in Figure 4.2 the map of Anhui

province.

(1) Fuyang-Yingshang (G105) and (2) Huainan-Hefei (G206) are national

highways (second class highway), which are located at the western and

middle part of Anhui province respectively. These highways are the major

passageways of gravel and coal.

(3) Liuan-Hefei (G312) is a national highway (first class highway) which

connects the important point of industry and commerce within Anhui, it is

also the only investigated road section which runs from west to east.

(4) Hefei-Anqing and (5) Wuhu-Xuancheng are expressways (first class

highway). Hefei-Anqing is located at the heartland of Anhui which is an

1 2

3 4

5

6

Fuyyang

Huainan

Hefei

(Capital city)

Liuan

Xuanzhou

Huangshan


-40-

arterial road inside Anhui. The other is located at the border of Anhui, which

is one of the major connections with Jiangsu province.

(6) Xiuning-Jingdezhen is a common road (third class highway), and is the

only common road among the selected highways, located at the southeast

part of Anhui province.

The road selection is based on the location of commercial and industrial

cities. Hefei is the capital city of Anhui, which is the economic centre of Anhui.

On the other hand, Hefei connects to the major industrial cities, Huainan and

Liuan. Thus, roads between these cities take an important role in the freight

industry. Fuyang, Xuancheng, Jingdezhen are the adjacent cities to Henan,

Jiangxi and Jiangsu province representatively, so that they play an important

role in provincial business trade. Thus, the investigation results represent the

typical traffic status in Anhui.

Developments of road network and economy have an inter-relationship.

Volume of commercial vehicle and its growth rate represent the development

status of an area. The traffic data from these areas are reliable and typical.

4.1.4 Investigation Schedule

The investigation was carried out from April 19 to April 25, 2004, preliminary

investigation between 19th to 21st April and the formal investigation from 21st

to 25th April 2004. 21st to 23rd of April 2004 were normal working days of

Anhui and 24th and 25th on the weekend. Thus this investigation combined all

typical traffic condition; as a result the investigation is reliable. The detailed

investigation schedule is shown in Table 4.1.


-41-

Table 4.1 Investigation implementation schedule

Item Preliminary

investigation investigation Remark

Traffic volume

survey 08:10-08:20

08:30-11:30 and

13:30-16:30

Investigator gathered at

16:30-16:40

Vehicle speed

survey 07:40-07:50

08:00-10:00 and

16:00-18:00 -

Axle load weight

survey -

08:30-11:30 and

13:30-16:30 -

Integrative enquiring

investigation -

08:30-11:30 and

13:30-16:30

Interview with truck driver

during axle load survey

Road surface

investigation -

On the way to

investigation site -

The aims of data collection are shown as below:

� Investigation of traffic flow based on axle types

� Investigation of truck speed based on axle types

� On site axle load investigation

� Interview with overloaded truck driver to investigate overloading reason

� Interview with investigated truck driver to assess economic effect of

overloaded truck traffic

� Investigation on the status and service life of overloaded truck

4.1.5 Random Sample

This survey just concerned one direction traffic flow due to insufficient human

resource. When the investigation was carried out, investigators were

standing at the road side and stop heavy truck by hand signal randomly.

Since not every truck driver was willing to cooperate with investigator, thus

data sample of truck type proportion were not conformed to the actual traffic

flow. According to the interview with SEU’s staff, although the traffic flow of

all investigated highway sections contain passenger car and coach;

surveyors did not stop them for axle load survey. All kinds of vehicles were

taken into account when traffic volume survey was carried out, because


-42-

traffic volume survey needed to consider the percentage of truck in total

traffic flow thus a complete and reliable representation must be obtained.

4.1.6 Site selection and WIM system

Mobile Weigh in Motion (WIM) system took an important role during the

investigation, because information of axle loads and vehicle load were

depended on WIM, another major reason to choice these locations.

For the reason of accuracy, fixed and mobile WIM were applied together,

thus mobile and fixed WIM must be present at investigation sites. However,

Anhui province has a lack of resources to install fixed WIM and setup

inspection stations on all highways, thus the sites were limited. Six sites

were chosen after the consideration of traffic flow, connection of major cities

and economic development.

It must mentioned that the traffic flow direction of China is same as the U.S.,

driving on the right hand side of the road, thus the left lane is the fast lane

and right lane is the slow lane. When inspection was carried out,

investigators were standing at the side of both lanes and chose trucks

randomly. Every investigated truck was asked to weigh by WIM.

In Figure4.3 (a), investigators were standing at the far side of road to select

trucks randomly and the inspection station is at the side of slow lane. The

speed of vehicle is quite slow at this road section thus it is easier to stop,

because vehicle is ready to stop for toll. Usually, trucks travels on the left

lane (fast lane) of highway, thus SEU arranged staff at the fast lane to stop

vehicles. Speed of the trucks show in this picture just around 40-60km/h, so

staffs were able to stop them for investigation.

The inspection station shows in Figure 4.3(d), permanent WIM system was

installed in the station but it does not operate all the time because of

insufficient human resources. It is operated when APCD or institution for

scientific research is required. During the investigation period, mobile WIM

was installed at the station as well, which is shown in Figure 4.3(e).


-43-

When the weighing was carried out, each truck was stopped in front of a

weigh scale. Then the truck moved slowly at a speed lower than 10km/h.

When the wheel of first axle located at the middle part of weight scale, the

truck was stopped and the static weight was taken. This procedure was

repeated until all axles of truck were weighed. Then the axle load and total

load of that truck were obtained.

The details of weighing procedure are shown in figure 4.3(a) to (f)

Figure 4.3 (a) On site investigation by human.

Figure 4.3(b) Trucks travels on the left lane (fast lane) of highway.


-44-

Figure 4.3(c) Inspection station shows in this figure.

Figure 4.3(d) The fixed WIM shows in the above figure.

Figure 4.3(e) Mobile WIM was combined with fixed WIM in this survey.


-45-

Figure 4.3(f) Truck was weighed by WIM statically.

4.1.7 Preliminary of Raw Data

After the survey, 441 samples were collected from six sites and valid

samples are 418. 23 samples were eliminated from 441 samples; because

human errors occurred during same survey, some information of these 23

samples were incomplete. As a result, they could not be used for data

analysis.

The brief of these samples are shown in table 4.2.


-46-


-47-

Class 6 and 7 is seldom used in freight industry, thus a zero sample was

collected in this survey. Meanwhile, class 10 and 11 would not be considered

in data analysis as well because of the insufficient sample size.

According to the truck types, class 4 can be classified as medium truck and

class 5 is heavy truck, classes 8-12 are classified as trailer truck. The

statistics data in Table 4.2 are partially different from actual traffic component

of Anhui province. This survey shows that the data of medium truck was the

largest proportion of this survey, trailer truck was the second large sample

group and the heavy truck was the smallest proportion. Moreover, a large

difference between sample size of class 4, 5 and 8 was present.

However, the actual traffic of Anhui province shows another picture.

According to the data in 2004 provided by APCD, medium truck had the

largest proportion of Anhui’s truck traffic. The second largest proportion was

heavy truck and trailer truck was the third. The different in proportion of each

truck groups are very small. The traffic component detail of Anhui in 2004 is

shown in Table 4.3.

After the interview with staff of SEU, the major reason cause the different in

traffic component was found. During investigation, drivers would not

cooperate with surveyor when their truck was overloaded. Thus surveyors

found it very hard to stop overloaded trucks.


-48-


-49-

4.1.8 Summary of data collection

The data were collected through the wide range of Anhui Province, which

combined the highways in longitudinal and transverse direction. Anhui

covered a wide area thus data collection could not be carried out intensively,

but the highway sections chosen in this investigation are typical of Anhui

thus the data collection result is highly reliable.

4.2 Data Analysis

Data analysis is the most important process to carry out research result.

Economical loss which caused by overloaded truck traffic is the target of this

study, thus the following chapters express the process of data analysis and

how to determine the economical losses from overloaded truck traffic.

The process of data collection and preliminary work were shown in Chapter

4.1, therefore data analysis was started at that point. ESAL of trucks were

analysed according to their datasets, thus the mean of each dataset could be

found. According to the mean ESAL and the actual on-road AADT

percentage of each truck group, the total ESAL which include overloaded

traffic can be simulated.

According to the information of Anhui province in 2003, the comparison

between with and without overloaded traffic can be found. Because of the

effect of overloaded truck traffic has a direct proportion relationship, Highway

G206 is the only one case study taken into account.

Design life Analysis is the advance step in the analysis process. Design life

and actual service life of pavement were calculated. In the mean while, net

present value of investment (NPV) for pavement is the most important index

in economical losses estimation. The flow chart of data analysis is shown in

Figure 4.4


-50-

Figure 4.4 Flow Chart of Data Analysis in Chapter Five and Six

Net present values(NPV) of investment for pavement with and without

overloaded truck traffic were found.

Comparison between NPV and service life were found as the targets of this

research.

The different between with and without overloaded truck traffic could be found.

Using the ESAL found in previous step, service life of pavement for each vehicle

group could be found.

Considered the whole traffic stream, using the proportion of every vehicle

group to calculate the actual service life.

Dataset input to @RISK program according to truck group.

The bestfit distribution of each truck group was developed, and the mean

ESAL was found.

Using the percentage of actual traffic stream to calculate the AADT of

Highway G206 when overloaded truck traffic was considered.


-51-

5. Overloaded truck traffic analysis

5.1 Methodology of Data Analysis

5.1.1 Classification of Data

The data used in this research was collected from six investigation points

which dispersed in Anhui Province. Classification of data was based on the

standard of FHWA (U.S.), as shown in table 4-2. According to FHWA vehicle

classification, truck can be classified into nine groups. However, only five

trucks types commonly used in Anhui province, thus data analyses are

concerned in these five groups which are type 4, 5, 8, 9 and 12.

Analysis of loading effect was based on the axle group. Each truck class was

divided in to axle group and the @RISK software was used to build up the

distribution.

5.1.2 ESAL of each axle group

ESAL (Equivalent Standard Axles Load) plays an important role in traffic

data analysis. The axle loads can be converted using standard factors to

determine the damaging power of different types of vehicles. The damaging

power is normally expressed as the number of “equivalent standard axles”

(ESA). The design lifetime of pavements are expressed in terms of ESAs

that they are designed to carry.

To calculate the Standard Axle Repetitions (SAR) of damage, a procedure is

required to calculate the damage associated with each axle group of each

axle type in the traffic load distribution relative to the damage caused by a

Standard Axle. According to Pavement Design (Austroads, 2004), the

relationship between the truck’s equivalence factor and its axle loading is

expressed in Equation (5-1).


-52-

ESAL= (Actual axle load / Standard axle load)m (5-1)

Equivalent Standard Axles Load (ESAL) indicates the damage the amount of

damage as a single passage of axle group type.

Actual axle load is obtained from on road trucks.

Standard axle load is the authority axle load for particular axle group.

m is an exponent which is specific to the damage type.

Calculation with a damage exponent m of 4 is commonly referred to as

Equivalent Standard Axles (ESA). This result is derived from field studies of

pavement performance, literature review in section 3.2.6 shown that 4 is the

most reasonable value for ESAL calculation.

5.1.3 Design ESAL

The following step of data analysis is the estimation of standard ESAL for

design lives. China standard is the major index in this study, because the

location of case study is in Anhui Province. Meanwhile, the traffic design of

Queensland (Australia) should also be considered, because a developed

country must adopt for comparison, thus Queensland is considered as a

developed country sample. The calculation of design heavy vehicle ESAL of

China is shown in Equation (5-2):

ESAL=365 x AADT of one direction x % of Heavy vehicle x ESAL factor (5-2)

Using Equation (5-2), the design heavy vehicle ESAL of China and

Queensland were calculated. ESAL factor is the major difference between

China and Queensland. In general, the ESAL factor of Queensland is smaller

than China, thus the allowable ESAL of Queensland is smaller than that of

China.

ESAL of each type of heavy vehicle were calculated separately, because

each vehicle type has a different ESAL factor and a different proportion of

traffic stream. The ESAL factor of Queensland and China is shown in table

5.1 and 5.2 respectively.


-53-

Table 5.1 ESAL factor of Queensland standard

ESAL factor of different highway type

1st type HW 2nd type HW 3rd type HW

type 4= Medium truck 0.7 0.7 0.6

type 5= Heavy truck 1.1 1.1 1.1

type 8= Single-trailer truck 1.4 1.4 1.3

type 9= Multi-trailer truck 1.3 1.3 1.1

type 12= Multi-trailer truck 1.3 1.3 1

Table 5.2 ESAL factor of China standard

ESAL factor of different highway type

1st type HW 2nd type HW 3rd type HW

type 4= Medium truck 1.0 1.0 1.0

type 5= Heavy truck 3.0 3.0 3.0

type 8= Single-trailer truck 5.0 5.0 5.0



Design ESAL based on China and Queensland standards are shown in

Table5.1 and 5.2. For detailed calculation refer to Appendix A.


-54-

Table 5.3 Accumulated ESAL of China and Queensland Standard.

Qld standard

1st type Highway 2nd type Highway 3rd type Highway

Sum ESAL of HW for 15 yrs 10,780,199 11,102,573 1,132,018



China index





In general, the standard ESAL of China is larger than that of Queensland.

However, the simulated ESAL still are much larger than the standard ESAL

of China. Simulated Heavy vehicle ESAL is 15 to 20 times that of Standard

ESAL of China and it is also 33 to 56 times that of Queensland standard. The

calculation detail of accumulated ESAL of China and Queensland are shown

in Appendix B.

It is no doubt that the damage which arises from overloaded truck traffic may

shorten the pavement service life. The effect of overloaded truck traffic to

pavement service life will be discussed in Chapter Six.

5.2 Data Analysis Result

Data analyses were based on the ESAL. ESAL of each data sample were

calculated by axle groups of individual truck. Thus the same types of truck

were grouped together when data analysis was carried out. Therefore, five

groups of data were developed and data analysis was carried out in each

data group individually.

Mean and 95th percentiles are the major indices in the analysis result,

because the 95th percentiles can represent the typical situation of data

distribution. Moreover, the mean value is the figure which is adopted in the

calculation of actual ESAL over the design period. In the process of data


-55-

analysis, @RISK can help to build up distributions of each data group, and

the best fit distribution was adopted in the subsequent calculation. @RISK is

used to analysed and uncertainty in a wide variety of industries (PALISADE,

2008). Because the sample size was limited, the distribution curve is not

smooth. Thus the best fit function which was adjusted by @RISK and shows

a reasonable distribution. Full details of data are shown in appendix C.

5.2.1 Class 4 truck

The distribution of class 4 trucks shows that the ESAL of the 90 percent of

trucks inspected are between 3.3 and 69.9. According to Queensland

standard, the ESAL of class 4 trucks is between 0.6 and 0.7.Thus the result

shows an extreme difference when the Queensland standard applied. On the

other hand, the analysis result also cannot fit for China standard, because

the design ESAL of class 4 in China is 1.0. The overloading rate which

occurs in class 4 trucks is between 116.5 and 99.9 times that of the standard

ESAL of Queensland and China respectively. Thus analysis result of class 4

truck exceeds the safety limit of highway design.

Exponential Distribution was adopted in the class 4 truck data analysis

because it can provide the most best fit distribution curve for this dataset.

The details of distribution for class 4 truck are shown in Table 5.4 and Figure

5.1

Table 5.4 Exponential distribution result of class 4 truck

Best Fit Input data

Left X 3.3 3.3

Left P 5.00% 3.28%

Right X 69.9 69.9

Right P 95.00% 95.08%

Minimum 2.1349 2.3204

Maximum +Infinity 106.83

Mean 24.764 24.949


-56-

Figure 5.1 Exponential distribution result of class 4 truck

5.2.2 Class 5 truck

Distribution of class 5 truck shows that the ESAL of 90 percentages of trucks

inspected are between 5.3 and 84.4. According to Queensland and China

standards, however, the standard ESAL of class 5 truck is 1.1 and 3.0

respectively. The ESAL of class 5 truck in Anhui may be up to 76.7 times the

standard when Queensland standard is applied, and 28.1 times the standard

under the China standard.

In the class 5 truck analysis, Exponential Distribution was adopted. The

details of distribution for class 5 truck show in Table 5.5 and Figure 5.2

Percentage of truck

ESAL at different percentage level


-57-

Table 5.5 Exponential distribution result of class 5 truck

Best Fit Input data

Left X 5.3 5.3

Left P 5.00% 4.76%

Right X 84.4 84.4

Right P 95.00% 92.86%

Minimum 3.8981 4.5379


Mean 30.767 31.407

Figure 5.2 Exponential distribution result of class 5 truck

Percentage of truck



-58-

5.2.3 Class 8 truck

The distribution of class 8 trucks shows that 90 percent ESAL of trucks

inspected were between 15.3 and 345.3. The result of class 8 shows the

most extreme phenomenon amongst all the sample groups. The standard

ESAL of class 8 in Queensland is 1.4 and 5.0 in China. Thus the maximum

ESAL in analysis sample is 246.6 times the Queensland standard and 69

times the China standards. Although class 8 is not the largest proportion of

truck traffic, the extreme overloading phenomenon also causes considerable

damage to highway infrastructure.

In the class 8 truck analysis, Pearson5 Distribution was adopted. The details

of distribution for class 8 are shown in Table 5.6 and Figure 5.3

Table 5.6 Pearson5 distribution result of class 8 truck

Best Fit Input data

Left X 15.3 15.3

Left P 5.00% 6.90%

Right X 345.3 345.3

Right P 95.00% 96.55%

Minimum -3.7979 11.748


Mean 116.994 94.366


-59-

Figure 5.3 Pearson5 distribution result of class 8 truck

5.2.4 Class 9 truck

The distribution of class 9 has been considered in two cases. The first case

of the distribution is developed by whole samples, and the second case is

developed by eliminated sample. The reason for developing two distributions

is that atypical samples occurred in the sample group. Fifteen samples of

class 9 were collected in the site survey, and the result shows that two

sample results were extremely different from the rest of the sample group.

Therefore, class 9 truck was analysed two times to obtain a reasonable and

smooth distribution curve. Detail of data samples can be seen in Appendix D.

In the first case, all samples were adopted in distribution. Thus the range of

ESAL is between 24.7 and 401.5. This result shows the widest range of

ESAL amongst all case analysed. Also the greatest ESAL also found in this

sample group. Although class 9 is not the major stream in truck traffic, it is

Percentage of truck



-60-

the most popular trailer truck used in Anhui, and thus class 9 truck certainly

have potential to cause serious damage to pavement. The standard ESAL of

class 9 in Queensland and China is 1.3 and 5.0 respectively. Thus the

maximum ESAL in analysis sample is 308.8 times that of Queensland

standard and 80.3 times that of China standards.

In the first case of class 9 truck analysis, a LogLogistic Distribution was

adopted. Detail of data samples can be seen in Appendix D

In the second case, the distribution was formed by thirteen samples from the

dataset of 9 class truck, because two extreme data samples were eliminated.

The ESAL range of case two is between 24.7 and 210.8; it shows a different

picture compared to the first case. Although some samples were eliminated,

the distribution curve is smoother than the first case, and in addition, the gap

between the 5th percentile and 95th percentile is smaller. As a result, the

second case can provide a more realistic best fit curve than first case. In the

second case, ESAL of class 9 truck is 162.2 times of Queensland standard

and 42.2 times that of China standards.

In the second case of class 9 truck analysis, a LogLogistic Distribution was

adopted. The details of distribution for first case of class 9 truck are shown in

Table 5.7 and Figure 5.4.

Table 5.7 LogLogistic distribution result of second case of class 9 truck

Best Fit Input data

Left X 24.6 24.6

Left P 5.00% 7.69% Right X

210.8 210.8 Right P

95.00% 100.00% Minimum -3.8774 16.532 Maximum

+Infinity 184.81 Mean

91.741 87.431


-61-

Figure 5.4 LogLogistic distribution result of second case of class 9 truck

5.2.5 Class 12 truck

The distribution of class 12 trucks has same problem as the class 9 truck,

and therefore two distributions were considered in class 12 truck as well. The

first case of distribution was using the entire dataset, and the second case

was developed after eliminating the atypical data. The reason to develop two

distributions is because atypical samples were recorded in the class 12 truck

sample group. Although nine samples of class 12 truck were collected in site

survey, and the result shows that one sample result was extremely different

from the other samples. Therefore, the class 12 truck dataset was analysed

two times in order to obtain a reasonable and smooth distribution curve.

Details of the data samples can be seen in Appendix D.

Percentage of truck



-62-

In the first case, all samples were adopted in the distribution. Thus the range

of ESAL is between 24.7 and 255.5. In the first distribution of class 12 truck,

there were only nine samples, thus distribution curve is not very smooth due

to the small sample size. The standard ESAL of class 12 in Queensland and

China is 1.3 and 5.0 respectively. Thus the maximum ESAL in analysis

sample is 196.5 times that of Queensland standard and 51.1 times that of

China standards.

In the first case of class 12 truck analysis, The LogLogistic Distribution was

adopted. The details of distribution for first case of class 12 truck are shown

in appendix D. The second time analysis of class 12 truck was similar to the

class 9 truck, one extreme sample was eliminated from dataset of class 12,

thus eight samples were used in the analysis.

The distribution shows a smooth and reasonable shape when the atypical

data is eliminated. The ESAL range of 5th percentile to 95th percentile

reduced to 16.3 and 175.4. Thus the ESAL of class 12 in the second case is

134.9 times that of Queensland standard and 35.1 times that of China

standard. Although some samples were eliminated, the distribution curve is

smoother than the first sub-analysis. Thus second sub-analysis provides a

more reasonable result than the first sub-analysis. In the second case of

class 12 truck analysis, a Logistic Distribution was adopted. The details of

distribution for first case of class 12 truck are shown in Table 5.8 and Figure

5.5


-63-

Table 5.8 LogLogistic distribution result of second case of class 12 truck

Best Fit Input data

Left X 16.3 16.3

Left P 5.00% 0.00%

Right X 175.4 175.4

Right P 95.00% 100.00%

Minimum -Infinity 22.564


Mean 95.873 95.039

Figure 5.5 LogLogistic distribution result of second case of class 12 truck

Percentage of truck



-64-

5.2.6 ESAL analysis result Comparison

The traffic data analysis in section 5.2.1 to 5.2.5 shows clearly that the

problem of truck overloading occurs in all truck classes. When Queensland

and China standard were applied, the largest overloaded rate occurs in

various truck classes because of the differences in the standard ESAL in

each country.

The largest overloading rate occurs in class 4 truck when the China standard

is applied. The reason for this result may be due to the physical design of

class 4 truck. Class 4 has the least axles amongst all truck type, thus the

bearing capacity of class 4 truck is the smallest amongst all classes.

However, because the difference in cargo volume of class 4,5 and 8 is not

determined. The cargo volume of class 4,5, and 8 is very close but the actual

bearing capacity of them are different. As a result, the problem of

overloading in class 4 truck is most serious.

On the other hand, the analysis shows that most serious overloading rate

occurred in class 8 trucks when the Queensland standard is applied. The

difference in standard ESAL of various truck classes is the reason for this

phenomenon. The Queensland Standard ESAL of class 4 for first class

highway is 0.7, and 1.4 for class 8 trucks. On the other hand, the China

Standard ESAL of class 4 and class 8 trucks for first class highway is 1.0 and

5.0. The obvious difference between Queensland and China causes a huge

variation in the analysis result. Owing to the fact that the case study was

carried out in China, the China standard was the main standard used in the

analysis.

Class 9 and 12 cannot be compared with class 4, 5 and 8, because class 9

and 12 trucks are tailored, and their physical designs have distinct variations

from class 4, 5, and 8. Furthermore, class 9 and 12 trucks were analysed in

two sub-analyses with significantly different results. Considering that the

second case sub-analysis provided a smoother and more reasonable

distribution curve, the second case sub-analyses of class 9 and 12 were

adopted in the following analysis.


-65-

Because of the small sample size, the analysis only can show a rough trend.

The design of a trailer is different from a normal truck, in which the container

can be combined freely. As a result, the cargo volume of a trailer is higher

than that of a normal truck.

Therefore, the overloading tendency of trailers is present in a reverse

relationship compared to the overloading tendency of the single cargo trucks.

According to the result discussed above, it is clear that the overloading rate

is influenced by the volume of cargo and the number of axles. The summary

of data analysis result is shown in table 5.9.

Table 5.9 Comparison between standard and actual ESAL of each class.

Vehicle class Actual mean

ESAL

China standard

ESAL

Queensland

standard ESAL

4 24.76 1.0 0.7

5 30.77 3.0 1.1

8 116.99 5.0 1.4

9 91.74 5.0 1.3

12 95.87 5.0 1.3

Owing to the different class of highway, the allowable of ESAL of truck is

various. Bearing capacity of first class highway is highest and third class

highway is lowest. Thus the comparison in table 5.2.6 is based on the

standard of first class. Although the highest standard was applied, the actual

ESAL is extremely over the limit of the standard ESAL.

In the next section, truck traffic analysis is carried out based on the

percentage of each truck class in the traffic stream.


-66-

5.2.7 Distribution of ESAL

According to the ESAL calculation result, distribution of each sample group

was built up by @RISK which a statistical software in cooperated into

Microsoft® Excel™.

Different distribution styles were adopted in various truck classes, such that

a best fit curve could be obtained. Accordingly, Exponential distribution was

adopted for class 4 and 5, Pearson5 for class 8, LogLogistic and Logistic for

class 9 and 12.

In the distribution, the mean of each sample group was found, such that the

ESAL of a particular sample group for one year could be calculated.

The result that arises from Equation (5-2) is the index for a particular truck

class. However it is only the result for one year, and therefore the

accumulated ESAL for design lives should be calculated for comparison.

The design lives for flexible and rigid pavements in the Anhui Province are

15 years and 30 years respectively. Furthermore, a 40 years design life was

considered in this comparison. Therefore, accumulated ESAL for 15, 30 and

40 years were calculated and considered. The simulated ESAL result is

shown in Table 5.10.

Table 5.10 Total Heavy vehicle ESAL simulated by data sample

Result from Data sample 1st type Highway 2nd type Highway 3rd type Highway

Sum ESAL of HW 15 yrs 605,760,173 484,392,731 37,484,775

Sum ESAL of HW 30 yrs 1,421,610,579 1,136,782,942 87,970,051

Sum ESAL of HW 40 yrs 2,114,908,350 1,691,174,624 130,871,702

Note: Calculation may refer to Appendix E

Section 5.2 has shown clearly that each class of truck has a different level of

overloading. However traffic stream is not only of the truck, but also of the

passenger vehicle and coach. Thus, the effects of the percentage of each

vehicle class under different standards and situations have to be considered

separately.


-67-

According to the Overloading Study Interim Report II (APCD, 2004),

around 35-48% of the traffic stream of Anhui province consisted of

passenger vehicles. The rest of traffic consisted of heavy vehicles, which

includes trucks, trailer and coaches. The second class national highway

Hefei-Huainan (G206) is adopted as an example in this analysis. Highway

G206’s traffic component shows the most common phenomenon amongst

six investigated highways. The aim of traffic component analysis is to find out

the damaging effect of overloaded truck traffic under different standards and

situations. Therefore, this highway can allow a more representative analysis.

The traffic component of Highway G206 is shown in Table 5.11.

Table 5.11., Traffic component of G206 in Anhui province in 2003.

Passenger Class 3 Class 4 Class 5 Class8-12 Coach

% of AADT

No./day

41.55% 10.93% 12.20% 12.25% 3.23% 19.84%

AADT

No./day

4045 1064 1188 1193 314 1932

Table 5.11 shows that class 8-12 are grouped together. This is because

class 8-12 is classified as heavy truck, and thus the statistics of Anhui

province shows this dataset as a combined group. Furthermore, class 10

and 11 had obtained zero data in the investigation, and thus they have been

ignored in the analysis.

5.3 Traffic Component Analysis

5.3.1 Standard ESAL of Anhui Province

Highway G206 is classified as a second class highway, and thus the ESAL

calculation without overloading is based on the standard of second class

highway, which may be referred to in Table 5.3.

Calculations included the percentage of passenger vehicle, coach, class

3-12 trucks. According to the information provided by APCD, class 8-12 were

group as heavy truck, and therefore they were considered as a combined


-68-

number in this analysis. The allowable AADT in one direction of Highway

G206 under standard of China is calculated as follows:

Total ESAL=∑ × ESALAADT (5.3)

Note: AADT for one direction in traffic flow

5.3.2 Standard ESAL of Queensland

Queensland Standard is the most conservative amongst the three situations.

In other words, it provides the safest standard for highway design. The

standard of Queensland for second class highway of Queensland may be

seen in Table 5.2.

The calculation of Queensland standard is the same as for the calculations

of the China standard, except that the ESAL index of class 8-12 must be

considered. Because the ESAL index of class 8 is 1.4, whilst that of class 9

and 12 is 1.3. Owing to insufficient information, the percentage of class 8, 9

and 12 cannot be determined, and thus the average value of 1.3 and 1.4

were adopted.

5.3.3 Actual ESAL of Anhui Province

The actual ESAL of Anhui province was estimated from the data sample,

which is representative of the actual conditions. The calculation has same

problem for the Standard of Queensland, because ESAL of class 8, 9 and 12

have different values, and thus the weighted average value of heavy truck

(class7-12) was calculated. The comparison of actual and standard ESAL is

shown in table 5.12.


-69-

Table 5.12, ESAL Comparison of G206 in Anhui province in 2003.

Total ESAL of heavy truck for G206

(∑ × ESALAADT )

China Standard 6337

Queensland Standard 2563

Actual 98007

Note: calculation detail could be referred to Appendix F

5.4 Summary

The calculation from section 5.3.1 to 5.3.3 shows a clear and extreme result.

The phenomenon highlighted in section 5.2 is explained by the differences in

standard ESAL indices, due to the fact that the allowable ESAL in China is

double that of Queensland. However, the most extreme situation is in the

actual ESAL which was estimated from data sample. The summary of traffic

analysis is shown in Table 5.13.

Table 5.13, Summary of ESAL analysis result

Vehicle class

ESAL Four Five Eight Nine Twelve

Standard of China 1.0 3.0 5.0 5.0 5.0

Standard of Queensland 0.7 1.1 1.4 1.3 1.3

5 percentile 3.3 5.3 15.3 24.6 16.3

95 percentile 69.9 84.4 345.3 210.8 175.4

Mean 24.764 30.767 116.994 91.741 95.873

Note: for the detail of Table 5.13, please refer to Appendix D

Although the data sample cannot represent a full picture of the traffic

situation in Anhui, it still has a certain level of reliability. The calculation result

does show that the actual ESAL on example Highway G206 is 12.5 times the


-70-

allowable ESAL of Standard of China, and 23.2 times the Standard of

Queensland.

In conclusion, overloading truck traffic has created an extra ESAL to

pavement. As a result, pavement structure could be damaged in unexpected

situations. The reduction in pavement life within the service period and the

associated economical cost will be estimated and discussed in Chapter 6.


-71-

6. Pavement service life analysis

6.1 Methodology of pavement service life analysis

6.1.1 Pavement design

Pavement service life is the most important factor in pavement management.

Usually, Pavement service life is affected by environment, traffic condition,

construction material and method, maintenance and rehabilitation. Therefore,

estimation of actual pavement service life is a complex issue in pavement

management.

Pavement life is the initial consideration in pavement design system. Traffic

design and structural design is the second stage of pavement, thus

construction material and traffic load need to accord design service life.

Pavement design process flow chart is shown in Figure 6.1.

Figure 6.1 pavement design system

Source: pavement design (AUSTROADS, 2004)

halla



-72-

In general, pavement design for a Chinese national or first class highway is

15 to 20 years. Certainly, an appropriate maintenance and rehabilitation plan

for pavement are necessary. Highway pavement design is different from

local-street, because traffic flow of highway surmounts local traffic without

doubt. Therefore, frequent construction and rehabilitation should facilitate

the handling of traffic.

6.1.2 Effect of overloading truck traffic

Heavy truck traffic takes an important role in pavement management.

Because pavement service life is the major performance criterion in

pavement design, construction and maintenance consideration play a

supporting role in pavement management system. Therefore, adjustment of

maintenance and rehabilitation plan must be done according to the

deterioration of pavement within the service period.

Maintenance and rehabilitation activities are dependent on pavement type.

In conducting cost comparisons based on present worth analyses, an

assessment must be made of future annual routine maintenance

requirements, periodic maintenance treatments such as resurfacing, and

rehabilitation such as structural overlays or strengthening.

The typical pavement life-cycle performance curve for an original pavement

structural section with an applied number of major rehabilitation cycles is

shown in Figure 6.2. Tm+1 is the ideal service life, and Tj∆ is the service time

between major maintenance activities. When the pavement management

system includes adequate maintenance and rehabilitation to maintain

performance condition index at the acceptable level (Pf), traffic disorder

induced by poor travel condition can be minimised.


-73-

Figure 6.2 typical pavement life-cycle performance curve

Source: Optimum Flexible Pavement Life-Cycle Analysis Model (Khaled A. Abaza, P.E.

2002)

However, overloading truck traffic brings unexpected deterioration for

pavement within the service period. Although maintenance and rehabilitation

activities are carried out regularly, pavement structure and overlay are

damaged unexpectedly.

The traffic analysis in Chapter Five showed that all the common truck types

in Anhui are overloaded, thus the reduction in service life which is caused by

unanticipated truck traffic is discussed in this chapter.

6.2 Pavement Service Life Analysis Result

Analysis of pavement service life was similar to traffic data analysis.

Pavement service lives were calculated from ESAL of each truck, thus

service life for pavement when the traffic is entire by the same type of vehicle.

Nevertheless, traffic flow included various vehicles, thus actual service life

for particular pavement must be estimated according to the combination of

vehicle types. Therefore, a distribution of every single dataset was built up,

which presented an extreme situation when the entire traffic is one vehicle

halla



-74-

type. The purpose of this analysis was to find out the simplest effect of each

truck type to the pavement. After that, the mean of each truck group was

adopted to calculate the actual service life for combined traffic. The equation

of actual service life is shown in Equation (6-1), and the pavement service

life calculation is shown in Appendix G.

Actual Service Life= Design Life / Actual ESAL

(6-1)

Actual Service Life is the service period of pavement under actual ESAL.

Design Life is the expected service life in the design stage.

Actual ESAL is the ESAL of on road vehicle.

Traffic is combined by different type of vehicle, standard ESAL of each

vehicle type is different. Therefore actual ESAL of each vehicle type was

calculated by Equation (6-1), and then a distribution of each vehicle group

was found.

Pavement service life analysis continues from the result of traffic data

analysis in Chapter Five, thus Highway G206 is adopted as an example in

this chapter. Mean and 95th percentiles are the major indices in the analysis

result, because the 95th percentiles can represent the typical situation of

data distribution. Moreover, the mean value is the figure which is adopted in

the calculation of actual service life under the overloaded truck traffic. In the

process of data analysis, @RISK was used to build up distributions of each

data group, and the best fit distribution was adopted in the subsequent

calculation. Because the sample size was limited, the distribution curve is not

smooth. Thus the best fit function which was adjusted by @RISK and shows

a reasonable distribution.

Each of Fifteen and thirty year design lives were adopted, which accords

with Queensland’s Pavement Design Standard (pavement design,


-75-

AUSTROADS, 2004). Because a direct proportion relationship was found

between fifteen and thirty year design, the fifteen year design is adopted in

the following analysis. The pavement service life distribution is shown in

Appendix G.

6.2.1 Single truck traffic analysis

When entire traffic of class 4 truck, the distribution shows that 90 percent

pavement service life are between 0.188 and 3.469. This result is referred to

fifteen years design. An exponential distribution was adopted in the class 4

truck data analysis because it can provide the most best fit distribution curve

for this dataset. The details of distribution for class 4 truck are shown in

Table 6.1 and Figure 6.3.

Table 6.1 Exponential distribution of service life of class 4 truck

Best Fit (years) Input data (years)

Left X 0.188 0.188 Left P 5.00% 4.10% Right X 3.469 3.469 Right P 95.00% 93.44% Minimum 0.13128 0.14042 Maximum +Infinity 6.4644 Mean 1.2454 1.2545


-76-

Service life at different percentage level (years) Figure 6.3 Exponential distribution result of class 4 truck design for 15 years

When entire traffic of class 5 truck, the distribution shows that 90 percent

pavement service life are between 0.135 and 2.77. And the result is referred

to fifteen years design. An exponential distribution was adopted in the class

5 truck data analysis because it can provide the best fit distribution curve for

this dataset. The details of distribution for class 5 truck are shown in Table

6.2 and Figure 6.4.


Best Fit(years) Input data(years)

Left X 0.135 0.135

Left P 5.00% 7.14%

Right X 2.77 2.77 Right P

95.00% 95.24% Minimum

0.08936 0.11067 Maximum

+infinity 3.3055 Mean

0.98425 1.0056

Percentage of truck


-77-

When the entire traffic is class 8 truck, the distribution shows that 90 percent

pavement service lives are between 0.051 and 0.988. This result also

referred to fifteen years design, and the distribution shows an obvious

tendency that class 8 truck has highest potential to cause overloading

among single container trucks.

The exponential distribution was adopted in the class 8 truck data analysis



6.5.

Percentage of truck

Service life at different percentage level (years) Figure 6.4 Exponential distribution result of class 5 design for 15 years truck


-78-



Left X 0.051 0.051

Left P 5.00% 6.90% Right X

0.988 0.988 Right P

95.00% 93.10% Minimum 0.03479 0.04028 Maximum

+infinity 1.2769 Mean

0.35299 0.35848

Figure 6.5 Exponential distribution result of class 8 truck design for 15 years truck

When the entire traffic is class 9 truck, the distribution shows that 90 percent

pavement service life are between 0.077 and 0.597, and also this result also

referred to fifteen years design as well.

Percentage of truck

Service life at different percentage level


-79-

The exponential distribution was adopted in the class 9 truck data analysis



6.6.



Left X 0.077 0.077 Left P

5.00% 0.00% Right X

0.597 0.597 Right P 95.00% 92.31% Minimum

0.06758 0.08117 Maximum

+infinty 0.90734 Mean 0.24422 0.2578


Percentage of truck



-80-

The last dataset is class 12 truck. When entire traffic of class 12 truck, the

distribution shows that 90 percent pavement service life are between 0.083

and 0.495 years, and also this result also referred to fifteen years design as

well.

Exponential Distribution was adopted in the class 12 truck data analysis



6.7.



Left X 0.083 0.083 Left P

5.00% 0.00% Right X

0.495 0.495 Right P 95.00% 87.50% Minimum

0.0753 0.0928 Maximum

+infinty 0.66479 Mean 0.21533 0.23284


-81-


6.2.2 Case study for Highway G206

In China, pavement design usually adopts 20 or 40 years, thus the relevant

analysis figures were worked out by the direct proportion relationship for

further analysis. When pavement under the effect of overloaded truck traffic,

it definitely cannot achieve the design life. Means of each dataset are

summarized in Table 6.6.

Table 6.6 Mean service life for each dataset at different design

Mean service life years at different design periods Vehicle Type 15 years 20 years 30years 40years

4 1.25 1.67 2.51 3.35 5 1.01 1.34 2.01 2.68 8 0.36 0.48 0.72 0.96 9 0.26 0.34 0.52 0.69

12 0.23 0.31 0.47 0.62

Percentage of truck



-82-

According to the Overloading Study Interim Report II (APCD, 2004), the

second class national highway Hefei-Huainan (G206) had 38.61% is heavy

truck traffic. The traffic component of Highway G206 is shown in table 6.7.

Table 6.7 Traffic component of G206 in Anhui province in 2003.


% of AADT No./day 41.55% 10.93% 12.20% 12.25% 3.23% 19.84%

DDAT No./day 4045 1064 1188 1193 314 1932

In the calculation, passenger vehicle, coach and class 3 truck were assumed

to present no overloading problem. Thus the service life of pavement under

these three kinds of traffic can facilitate design. The actual service life for

Highway G206 was calculated according to the proportion of traffic

combination. Actual service life for each design standard is shown in table

6.8.

Table 6.8 Comparison of actual and design service life

Years

Design service life 15 20 30 40 actual service life 11.1 14.8 22.3 29.7

Percent Reduction (%) 26% Note: detail of life reduction calculation is shown in appendix H

Table 6.8 shown that, pavement service life of Highway G206 is reduced by

26% under the influence of overloading truck traffic. This reduction induced

an economic loss which caused by overloaded truck traffic. When a

pavement designs for 15 years, it only performs for 11 years. The total worth

value for fifteen years reduced to eleven years. The net present worth value

of pavement was calculated for comparison, and the comparison result is

shown in Table 6.9.


-83-

Table 6.9 comparison of NPV between actual and design service life.

Design service life 15 years 20 years 30 years 40 years

Net present worth value (NPV)

of investment

4.57 5.43 6.64 7.40

actual service life 11 years 15 years 22 years 29 years

Net present worth value (NPV)

of investment

8.96 11.14 14.08 16.20

Different in NPV of investment 4.39 5.71 7.44 8.80

Remark: currency unit in this table is US$M. Detail of calculation may referred to Appendix I

Calculation in table 6.9 was adopted from the maintenance and rehabilitation

information of Anhui province and ShangDong Province. The information

provided by Anhui province shown that annual budgeting maintenance and

rehabilitation cost is US$1,797 and US$2,643 per kilometre respectively.

However, the actual annual maintenance and rehabilitation cost for Highway

G206 was US$1,541 and US$9,872 per kilometre respectively.

Comparison of each design shows an obvious phenomenon that actual NPV

is double the design NPV. NPV not only shows the total worth value of an

asset, but also the total investment to that asset. Highway G206 proved that

overloaded truck traffic reduced its service life for 26% and the total

investment increases by 105%.


-84-

6.3 Summary

Data analysis continued from Chapter Five ESAL analysis to determine

economical losses in Chapter Six. Chapter Five and Six show a general

structure for overloaded truck traffic analysis and how to estimate

economical loss which is induced by overloaded truck traffic.

Case study of Highway G206 was adopted in Section 6.2.2. It shown a 26%

reduction in pavement life can cause a 105% increase in actual maintenance

and rehabilitation expenditure.


-85-

7. Conclusion and Recommendation

7.1 Conclusion

Overloading of heavy goods road vehicles is a critical problem in developing

countries. It has an immediate impact in terms of increased road damage,

which causes a dramatic increase in road maintenance costs. Overloading of

these vehicles is a very common phenomenon in China, not only in Anhui

province but also across the whole country. The analysis result in chapter 5

and 6 show that around fifty to seventy percent of heavy vehicles are

overloaded and thereby cause accelerated deterioration of the road network

and pavement. Meanwhile, the analysis result shows that the overloading

rate may up to one hundred percent in Section 6.2.2.

The literature review in 3.2.2 and 3.2.3 show the behaviour model of

overloaded truck driver and risks involved. Truck overloading has certain

benefits for individual drivers, such as reduction in running cost and

overhead. Therefore truck drivers have an incentive to take the risk and get

the benefit. However, the study in 3.2.3 shows that the benefit of overloading

by the operator cannot cover that pavement damage loss to the community.

The pavement life analysis and net present value estimation presented the

result that overloaded truck traffic induced a huge economical loss.

Pavement service life reduced by 26% and net present value of total

pavement investment increased by 105%. Pavement cannot achieved its

design life when overloaded truck traffic is applied, thus construction cost for

new pavement is involved.


-86-

7.2 Practice in developing and developed countries

Developing countries such as China, Zambia and South Africa have paid

attention to the overloading problem, because overloaded truck can be

involved in fatal traffic accidents. However, developing countries have

insufficient resources and experience in legislation and the technicalities of

pavement management, therefore overloading is a common phenomenon.

Most of the developed countries have a similar system to control overloaded

truck traffic. Complete legislation and twenty four hour all weather monitoring

system is the most important means of overloaded truck traffic control. The

U.S.A. has adequate experience and resource to treat overloaded trucks, but

they still exist. In general, 0.5-2.0% overloaded truck was found in traffic

stream. According to the literature review in 3.2.3, the level of enforcement

and percentage of overloading has an inverse relationship. Nearly 95%

overloaded truck can be eliminated, however a small proportion of

overloaded vehicle cannot be inspected. If government want to eliminate that

0.5-2.0% of overloaded vehicles, they must input huge resources for

inspection, and this behaviour does not fit to the standpoint of the economy.

As a result, less than 5% of overloaded truck traffic is tolerated in the

pavement management system.

7.3 Overloaded truck traffic control in Anhui

Anhui Province has paid attention to overloaded truck traffic since 2000. The

overloaded truck traffic control strategy of Anhui province is based on three

phases; monitoring, legislation and education.

Monitoring and legislation are the compulsory conditions to control

overloading, 24hour all weather monitoring system is the common

equipment in The U.S.A. and Australia, which can reduce overloaded truck

traffic efficiently.

Clear and complete legislation supports the monitoring system. Insufficient

inspection and monitoring are serious problems in China for overloaded


-87-

truck traffic control, however, the obscure legislation system also provides a

loophole for illegal operators and enterprises.

Education is a noticeable strategy in overloaded truck traffic control.

Although education cannot show its effect immediately, it provides dividends

in the long term.

7.4 Recommendation

A complete overloaded truck traffic strategy can be divided to two stages.

The first stage is monitoring and legislation. Twenty four hour all weather

monitoring system is necessary in overloading control. This system can

inspect most vehicles on the road and prevent overloaded drivers detouring.

Meanwhile, an adequate penalty to restrain the driver and enterprise is very

important. It can have the potential to stop overloading activity immediately.

Education is the second stage in the overloading control strategy. It may take

fifteen to thirty years to carry out. After people learn to understand the harm

of overloading, the incentive of overloading may decrease.

7.5 Direction for Future Research

This research has provided a step for overloading study. Overloaded truck

traffic impacts not only affect the economy, but also to society and

environment. However, economic impact is the major concern in developing

countries; and this thesis along with the relevant literature emphasizes

economic impact. Example of areas that would be beneficial to research

includes:

� Environment and social health,

� Economic loss in term of toll and taxation loss,

� Comparison between different Infrastructure management systems. These subjects are valuable to study, because these issues are related to

human life directly. Developed countries concern on the social health and

environment impact, because they are emphasized quality of life, thus health


-88-

is the significant subject to concern. In developing countries, they are

concern on the economic growth, thus economical efficiency is the major

concern of them.


-89-

Appendix A

Standard ESAL of China and Queensland (Australia)

Standard ESALs of each vehicle type on highway is different. The standard

ESAL of China for each vehicle type is shown in table A1.

Table A1 the standard ESALs of China of different vehicle type on highway

Index from

China Car

Medium

bus

Large

Bus

Light

truck

Medium

truck

Heavy

truck

Trailer

Truck

Vehicle Type -- -- -- 3 4 5 8-12

ESAL factor 0.02 0.4 0.7 0.1 1 3 5

Source: Koji T, Riaz-ul-Islam, Guan C

Standard ESALs of Queensland is different from China, vehicle type not the

only reason to affect the ESALs, but also the type of highway. The standard

ESAL of Queensland for each vehicle on various highways is shown in table

A2.

Table A2 the standard ESALs of Queensland of different vehicle type on highway

Daily ESAL of each vehicle type 1st type

Highway

2nd type

Highway

3rd type

Highway

med truck (type 4) 0.7 0.7 0.6

heavy truck (type 5) 1.1 1.1 1.1

Single-trailer truck(type 8) 1.4 1.4 1.3

Multi-trailer truck (type 9) 1.3 1.3 1.1

Multi-trailer truck(Type 12) 1.3 1.3 1

Source: pavement design (AUSTROADS, 2004)


-90-

Appendix B

Accumulated Standard ESAL of China and Queensland

The accumulated Standard ESAL of China and Queensland were calculated

in the same method. First of all, total ESALs for one year was found. And

then the cumulative Growth Factor (CGF) was adopted from pavement

design (AUSTROADS, 2004) to calculate the total EASLs of pavement

throughout the design period.

The following step shows the detail of calculation in Queensland Standard of

G206 Highway:

Step 1:

Accumulated ESAL for Heavy vehicle=365xAADTfor both direction x 0.5 x

percentage of heavy vehicle x standard ESAL for heavy vehicle

Thus, for first class highway,

Accumulated ESAL for type 4 vehicle =365x5396x0.5x0.2217x0.7

=152,826.64

Accumulated ESAL for type 5 vehicle = 365x5396x0.5x0.1497x1.1

=162,162.08


=249,127.11


=30,980.86


=28,036.40


-91-

For second class highway,


=171,681.40


=294,815.33


=77,835.47


=3,202.09


=94,233.00

For third class highway,


=25,297.72


=40,136.89

Accumulated ESAL for type 8 vehicle : Data not available




-92-

Table B1, the summary of calculation for 1 year ESAL for Queensland Standard

ESAL for 1 year

Vehicle Type 1st type highway 2nd type highway 3rd type highway

Type 4 152,862.46 171,681.40 25,297.72

Type 5 162,162.08 294,815.33 40,136.89

Type 8 249,127.11 77,835.47 Not available



To calculate the accumulated ESAL for design period, CGF were adopted

from pavement design (AUSTROADS, 2004, Table 7.4).

Total ESAL for design period= 1 year accumulated ESAL of heavy vehicle x

CGF

15, 30 and 40 design period are considered in this calculation, thus three

cases are shown in Table B2 to B4.

Table B2, the accumulated ESAL for each heavy vehicle type in 15 design period

ESAL for 15 year


Type 4 2,643,897.69 2,970,088.19 437,650.54

Type 5 2,805,403.91 5,100,305.16 694,368.21

Type 8 4,309,899.08 1,346,553.61 Not available


Type 12 485,029.75 1,630,230.82 Not available

Total ESAL 10,780,199.39 11,102,573.97 1,132,018.76

Note: the CGF factor for 15 years design is 17.3 when annual growth rate assumed as 2%


-93-


ESAL for 30 year


Type 4 6,204,754.13 6,970,264.77 1,027,087.40

Type 5 6,583,780.28 11,969,502.28 1,629,557.78


Type 9 1,257,823.09 130,004.94 Not available


Total ESAL 25,299,196.26 26,055,751.63 2,656,645.18



ESAL for 30 year


Type 4 9,230,717.96 10,369,556.45 1,527,982.25

Type 5 9,794,589.38 17,806,845.76 2,424,268.22




Total ESAL 37,637,227.94 38,762,743.81 3,952,250.46


In the second step, the accumulated ESAL for China Standard in G206 were

calculated. The calculation of China Standard is similar to Queensland. And

the major different is China Standard does not grade the highway.

Meanwhile class 8 to 12 trucks are combined in a group.

The following step shows the detail of calculation in China Standard of G206

Highway:

Step 1:


percentage of heavy vehicle x standard ESAL for heavy vehicle



-94-


=152,826.64


=152,826.64

Accumulated ESAL for type 8-12 vehicle =365x5396x0.5x0.2268x5.0

=1,116,729.18



=245,259.14


=804,041.80

Accumulated ESAL for type 8-12 vehicle =365x6427x0.5x0.0474x5.0

=625,734.15



=42,162.87


=109,464.25

Accumulated ESAL for type 8-12 vehicle : Data not available


-95-

Table B5, the summary of calculation for 1 year ESAL for China Standard

ESAL for 1 year


Type 4 218,323.51 245,259.14 42162.8655

Type 5 442,260.21 804,041.80 109464.2483

Type 8-12 1,116,729.18 652,734.15 Not available



Total ESAL for design period= 1 year accumulated ESAL of heavy vehicle x

CGF


cases are shown in Table B6 to B8.


ESAL for 15 year


Type 4 3,776,996.71 4,242,983.13 729,417.57

Type 5 7,651,101.58 13,909,923.16 1,893,731.49


Total ESAL 30,747,513.10 29,445,207.15 2,623,149.07



ESAL for 30 year


Type 4 8,863,934.47 9,957,521.09 1,711,812.34

Type 5 17,955,764.40 32,644,097.13 4,444,248.48

Type 8-12 45,339,204.71 26,501,006.64 Not available

Total ESAL 72,158,903.58 69,102,624.87 6,156,060.82




-96-

ESAL for 30 year


Type 4 13,186,739.94 14,813,652.07 2,546,637.08

Type 5 26,712,516.50 48,564,124.80 6,611,640.59

Type 8-12 67,450,442.47 39,425,142.89 Not available

Total ESAL 107,349,698.92 102,802,919.75 9,158,277.67


The summary of Accumulated ESAL of China and Queensland Standard is

shown in table B9.

Table B9, Accumulated ESAL of China and Queensland Standard.

Qld standard





China index






-97-

Appendix C

Detail Dataset

In this research, six investigation points were setup for data collection. As a

result, 418 valid data were collected. Details of data are show in Table C1 to

C5.

Table C1, Detail of data for type 4 vehicle

Vehicle laden mass axle 1 axle 2

No Rated(t) Actual(t) Rated(t) Actual(t) Rated(t) Actual(t)

1 13.60 18.51 5.40 5.17 8.20 13.34

2 13.60 22.15 5.40 7.25 8.20 14.90

3 13.60 21.21 5.40 6.09 8.20 15.12

4 13.60 15.23 5.40 1.14 8.20 14.09

5 13.60 21.34 5.40 7.25 8.20 14.09

6 13.60 25.14 5.40 5.38 8.20 19.76

7 13.60 22.96 5.40 4.56 8.20 18.40

8 13.60 25.67 5.40 5.06 8.20 20.61

9 13.60 25.36 5.40 5.08 8.20 20.28

10 13.60 26.84 5.40 7.45 8.20 19.39

11 13.60 14.34 5.40 3.72 8.20 10.62

12 13.60 28.64 5.40 7.45 8.20 21.19

13 13.60 23.56 5.40 4.82 8.20 18.74

14 13.60 20.31 5.40 5.98 8.20 14.33

15 13.60 24.54 5.40 5.05 8.20 19.49

16 13.60 27.34 5.40 3.88 8.20 23.46

17 13.60 21.68 5.40 4.41 8.20 17.27

18 13.60 28.67 5.40 7.20 8.20 21.47

19 13.60 22.64 5.40 5.10 8.20 17.54

20 13.60 20.80 5.40 4.49 8.20 16.31

21 13.60 26.22 5.40 5.77 8.20 20.45

22 13.60 19.81 5.40 5.82 8.20 13.99

23 13.60 21.13 5.40 5.84 8.20 15.29

24 13.60 18.05 5.40 4.37 8.20 13.68

25 13.60 21.52 5.40 6.43 8.20 15.09

26 13.60 17.79 5.40 3.52 8.20 14.27


-98-

27 13.60 24.04 5.40 4.05 8.20 19.99

28 13.60 21.42 5.40 4.95 8.20 16.47

29 13.60 16.20 5.40 3.70 8.20 12.50

30 13.60 17.95 5.40 4.88 8.20 13.07

31 13.60 13.24 5.40 2.80 8.20 10.44

32 13.60 24.94 5.40 5.35 8.20 19.59

33 13.60 21.50 5.40 4.02 8.20 17.48

34 13.60 15.80 5.40 4.38 8.20 11.42

35 13.60 20.32 5.40 4.44 8.20 15.88

36 13.60 26.40 5.40 7.03 8.20 19.37

37 13.60 19.15 5.40 1.96 8.20 17.19

38 13.60 21.21 5.40 4.46 8.20 16.75

39 13.60 17.31 5.40 5.60 8.20 11.71

40 13.60 21.36 5.40 4.20 8.20 17.16

41 13.60 23.66 5.40 4.54 8.20 19.12

42 13.60 26.04 5.40 6.11 8.20 19.93

43 13.60 24.19 5.40 4.86 8.20 19.33

44 13.60 18.31 5.40 4.37 8.20 13.94

45 13.60 24.43 5.40 6.96 8.20 17.47

46 13.60 21.62 5.40 6.40 8.20 15.22

47 13.60 17.81 5.40 3.71 8.20 14.10

48 13.60 18.24 5.40 4.12 8.20 14.12

49 13.60 17.49 5.40 3.20 8.20 14.29

50 13.60 23.52 5.40 5.77 8.20 17.75

51 13.60 17.58 5.40 4.97 8.20 12.61

52 13.60 25.30 5.40 5.91 8.20 19.39

53 13.60 22.80 5.40 4.63 8.20 18.17

54 13.60 20.51 5.40 6.41 8.20 14.10

55 13.60 20.16 5.40 3.39 8.20 16.77

56 13.60 20.50 5.40 5.70 8.20 14.80

57 13.60 19.46 5.40 4.74 8.20 14.72

58 13.60 28.20 5.40 6.04 8.20 22.16

59 13.60 29.10 5.40 7.44 8.20 21.66

60 13.60 16.90 5.40 5.46 8.20 11.44

61 13.60 22.34 5.40 5.38 8.20 16.96


-99-

62 13.60 17.00 5.40 3.44 8.20 13.56

63 13.60 23.38 5.40 5.20 8.20 18.18

64 13.60 23.32 5.40 3.52 8.20 19.80

65 13.60 15.16 5.40 3.68 8.20 11.48

66 13.60 19.64 5.40 6.00 8.20 13.64

67 13.60 16.08 5.40 5.28 8.20 10.80

68 13.60 20.08 5.40 7.28 8.20 12.80

69 13.60 22.28 5.40 5.66 8.20 16.62

70 13.60 29.74 5.40 4.92 8.20 24.82

71 13.60 19.74 5.40 4.84 8.20 14.90

72 13.60 16.90 5.40 4.78 8.20 12.12

73 13.60 26.28 5.40 4.82 8.20 21.46

74 13.60 19.80 5.40 5.75 8.20 14.05

75 13.60 18.05 5.40 6.55 8.20 11.50

76 13.60 23.15 5.40 8.75 8.20 14.40

77 13.60 24.00 5.40 7.05 8.20 16.95

78 13.60 28.70 5.40 6.45 8.20 22.25

79 13.60 32.80 5.40 8.65 8.20 24.15

80 13.60 20.50 5.40 4.55 8.20 15.95

81 13.60 20.65 5.40 4.35 8.20 16.30

82 13.60 24.55 5.40 8.05 8.20 16.50

83 13.60 29.15 5.40 7.30 8.20 21.85

84 13.60 27.80 5.40 8.65 8.20 19.15

85 13.60 20.55 5.40 7.00 8.20 13.55

86 13.60 27.30 5.40 5.40 8.20 21.90

87 13.60 18.30 5.40 5.70 8.20 12.60

88 13.60 20.45 5.40 5.70 8.20 14.75

89 13.60 13.10 5.40 3.10 8.20 10.00

90 13.60 20.10 5.40 5.65 8.20 14.45

91 13.60 15.40 5.40 4.35 8.20 11.05

92 13.60 22.90 5.40 7.90 8.20 15.00

93 13.60 27.15 5.40 5.80 8.20 21.35

94 13.60 21.50 5.40 4.80 8.20 16.70

95 13.60 26.70 5.40 7.55 8.20 19.15

96 13.60 21.30 5.40 5.60 8.20 15.70


-100-

97 13.60 21.80 5.40 5.65 8.20 16.15

98 13.60 23.05 5.40 7.50 8.20 15.55

99 13.60 14.55 5.40 4.10 8.20 10.45

100 13.60 23.85 5.40 5.40 8.20 18.45

101 13.60 25.50 5.40 7.10 8.20 18.40

102 13.60 30.55 5.40 9.15 8.20 21.40

103 13.60 34.95 5.40 9.10 8.20 25.85

104 13.60 26.95 5.40 5.20 8.20 21.75

105 13.60 17.15 5.40 4.75 8.20 12.40

106 13.60 32.25 5.40 7.80 8.20 24.45

107 13.60 29.30 5.40 7.50 8.20 21.80

108 13.60 31.45 5.40 7.15 8.20 24.30

109 13.60 22.70 5.40 4.25 8.20 18.45

110 13.60 29.50 5.40 6.45 8.20 23.05

111 13.60 16.50 5.40 4.55 8.20 11.95

112 13.60 22.20 5.40 6.25 8.20 15.95

113 13.60 17.05 5.40 5.75 8.20 11.30

114 13.60 17.65 5.40 5.05 8.20 12.60

115 13.60 30.65 5.40 8.35 8.20 22.30

116 13.60 20.40 5.40 2.50 8.20 17.90

117 13.60 15.85 5.40 4.30 8.20 11.55

118 13.60 19.95 5.40 6.50 8.20 13.45

119 13.60 21.15 5.40 5.85 8.20 15.30

120 13.60 28.60 5.40 6.00 8.20 22.60

121 13.60 24.80 5.40 6.85 8.20 17.95

122 13.60 31.35 5.40 7.20 8.20 24.15


-101-


axle 1 axle 2

No Rated(t) Actual(t) Actual(t) Actual(t)

FRONT REAR

1 5.40 6.05 21.55 15.16

2 5.40 5.41 15.83 12.91

3 5.40 8.33 15.37 16.08

4 5.40 6.17 22.62 22.25

5 5.40 4.07 12.99 11.78

6 5.40 5.75 11.37 13.71

7 5.40 5.23 16.59 18.43

8 5.40 7.36 17.54 15.66

9 5.40 8.80 18.21 16.01

10 5.40 7.89 19.54 20.20

11 5.40 5.73 18.26 17.23

12 5.40 5.19 12.31 13.58

13 5.40 5.37 15.59 13.57

14 5.40 9.45 46.25

15 5.40 6.23 45.17

16 5.40 7.24 36.14

17 5.40 7.62 31.24

18 5.40 6.70 14.28 15.04

19 5.40 4.74 19.60

20 5.40 4.48 20.88

21 5.40 5.50 31.60

22 5.40 7.58 25.52

23 5.40 5.35 10.60 10.60

24 5.40 6.30 13.50 10.35

25 5.40 5.00 16.05 8.45

26 5.40 6.95 13.45 10.25

27 5.40 7.20 17.80 13.30

28 5.40 5.85 11.10 7.30

29 5.40 6.30 18.50 17.05

30 5.40 4.40 9.65 11.35

31 5.40 5.95 14.85 9.35


-102-

32 5.40 5.90 17.50 20.70

33 5.40 7.55 12.70 13.50

34 5.40 8.45 11.60 21.00

35 5.40 7.00 11.85 12.30

36 5.40 5.50 15.00 16.30

37 5.40 4.65 8.50 13.50

38 5.40 7.50 14.05 10.65

39 5.40 6.00 11.95 11.70

40 5.40 8.05 12.65 11.00

41 5.40 9.80 16.70 17.40

42 5.40 5.35 12.20 11.65


axle 1 axle 2 axle 3


1 5.40 3.33 8.20 16.46 13.80 32.72

2 5.40 5.40 8.20 19.31 13.80 35.80

3 5.40 3.84 8.20 11.91 13.80 22.48

4 5.40 5.10 8.20 19.22 13.80 33.02

5 5.40 4.38 8.20 15.76 13.80 22.40

6 5.40 2.84 8.20 21.14 13.80 35.46

7 5.40 7.04 8.20 15.76 13.80 31.86

8 5.40 4.26 8.20 12.80 13.80 25.92

9 5.40 3.02 8.20 12.76 13.80 25.70

10 5.40 5.85 8.20 17.58 13.80 37.12

11 5.40 2.68 8.20 18.16 13.80 34.38

12 5.40 4.54 8.20 17.08 13.80 32.26

13 5.40 5.04 8.20 21.92 13.80 43.34

14 5.40 3.86 8.20 21.20 13.80 39.68

15 5.40 3.98 8.20 10.38 13.80 24.58

16 5.40 3.02 8.20 16.98 13.80 29.76

17 5.40 2.38 8.20 17.86 13.80 28.18

18 5.40 6.48 8.20 11.60 13.80 22.62

19 5.40 6.90 8.20 21.45 13.80 25.90


-103-

20 5.40 6.80 8.20 21.80 13.80 25.75

21 5.40 7.40 8.20 21.80 13.80 25.20

22 5.40 3.50 8.20 20.45 13.80 17.95

23 5.40 3.60 8.20 17.10 13.80 17.60

24 5.40 3.50 8.20 15.00 13.80 16.50

25 5.40 4.45 8.20 22.50 13.80 20.05

26 5.40 7.00 8.20 18.30 13.80 16.75

27 5.40 4.00 8.20 10.40 13.80 19.55

28 5.40 6.80 8.20 15.15 13.80 11.20

29 5.40 1.90 8.20 14.20 13.80 12.60

30 5.40 6.95 8.20 23.10 13.80 25.75

31 5.40 4.55 8.20 18.00 13.80 24.25

32 5.40 4.35 8.20 14.70 13.80 19.30

33 5.40 5.70 8.20 14.55 13.80 15.55

34 5.40 5.30 8.20 18.95 13.80 15.40

35 5.40 3.80 8.20 34.10 13.80 20.90

36 5.40 5.00 8.20 21.05 13.80 19.20

37 5.40 8.50 8.20 23.15 13.80 27.25

38 5.40 4.10 8.20 19.90 13.80 20.75

39 5.40 3.30 8.20 13.75 13.80 14.85

40 5.40 8.45 8.20 24.60 13.80 27.10

41 5.40 5.25 8.20 25.75 13.80 25.30

42 5.40 4.30 8.20 13.80 13.80 15.20

43 5.40 2.75 8.20 14.65 13.80 13.70

44 5.40 3.60 8.20 13.20 13.80 14.80

45 5.40 6.05 8.20 16.25 13.80 20.90

46 5.40 6.20 8.20 15.75 13.80 19.35

47 5.40 3.40 8.20 15.95 13.80 14.50

48 5.40 4.25 8.20 13.95 13.80 18.75

49 5.40 3.70 8.20 15.40 13.80 13.80

50 5.40 3.55 8.20 10.85 13.80 17.30

51 5.40 3.55 8.20 16.15 13.80 17.20

52 5.40 5.35 8.20 21.95 13.80 21.75

53 5.40 2.85 8.20 23.15 13.80 21.75

54 5.40 5.70 8.20 18.50 13.80 16.75


-104-


axle 1 axle 2 axle 3


1 5.40 8.80 8.20 27.21 18.50 74.26

2 5.40 5.74 8.20 14.86 18.50 26.90

3 5.40 7.35 8.20 21.65 18.50 41.70

4 5.40 8.90 8.20 26.75 18.50 75.30

5 5.40 4.15 8.20 26.05 18.50 40.60

6 5.40 8.60 8.20 22.70 18.50 44.35

7 5.40 8.85 8.20 26.15 18.50 46.75

8 5.40 6.25 8.20 15.60 18.50 43.40

9 5.40 7.40 8.20 24.75 18.50 58.25

10 5.40 7.55 8.20 19.25 18.50 40.20

11 5.40 5.20 8.20 18.45 18.50 44.60

12 5.40 7.05 8.20 18.25 18.50 39.25

13 5.40 6.95 8.20 18.30 18.50 40.05

14 5.40 9.45 8.20 24.75 18.50 53.30

15 5.40 5.05 8.20 21.05 18.50 38.00

55 5.40 4.75 8.20 14.85 13.80 15.20

56 5.40 6.60 8.20 19.05 13.80 18.05

57 5.40 4.65 8.20 13.55 13.80 11.50

58 5.40 6.35 8.20 20.95 13.80 28.20


-105-


axle 1 axle 2 axle 3 axle 4

No Rated(t) Actual(t) Rated(t) Actual(t) Rated(t) Actual(t) Rated(t) Actual(t)

1 5.40 3.69 8.20 3.99 5.40 9.34 5.40 10.32

2 5.40 3.56 8.20 16.68 5.40 14.70 5.40 15.98

3 5.40 3.76 8.20 17.56 8.20 13.20 8.20 15.44

4 5.40 3.52 8.20 21.60 8.20 12.94 8.20 17.82

5 5.40 3.44 8.20 20.74 8.20 17.44 8.20 20.12

6 5.40 3.78 8.20 27.62 8.20 14.22 8.20 18.08

7 5.40 3.64 8.20 22.88 8.20 17.88 8.20 19.68

8 5.40 3.32 8.20 21.54 5.40 14.90 5.40 19.02

9 5.40 3.14 8.20 15.84 5.40 12.78 5.40 14.36


-106-

Appendix D

ESAL Data analysis

ESAL data analysis is the core part of this thesis, this process involves many

graph and figure. Thus, details of distribution information are shows in this

section.

Type 4 truck,

Table D1, detail of ESALs for type 4 truck dataset

ESAL of Ax1 ESAL of Ax2 Total ESAL

No for each truck

1 0.84 7.00 7.84

2 3.25 10.90 14.15

3 1.62 11.56 13.18

4 0.00 8.72 8.72

5 3.25 8.72 11.97

6 0.99 33.72 34.71

7 0.51 25.35 25.86

8 0.77 39.91 40.68

9 0.78 37.41 38.20

10 3.62 31.26 34.89

11 0.23 2.81 3.04

12 3.62 44.59 48.22

13 0.63 27.28 27.91

14 1.50 9.33 10.83

15 0.76 31.91 32.68

16 0.27 67.00 67.26

17 0.44 19.67 20.12

18 3.16 47.00 50.16

19 0.80 20.93 21.73

20 0.48 15.65 16.13

21 1.30 38.68 39.99

22 1.35 8.47 9.82


-107-

23 1.37 12.09 13.46

24 0.43 7.75 8.18

25 2.01 11.47 13.48

26 0.18 9.17 9.35

27 0.32 35.32 35.63

28 0.71 16.27 16.98

29 0.22 5.40 5.62

30 0.67 6.45 7.12

31 0.07 2.63 2.70

32 0.96 32.57 33.54

33 0.31 20.65 20.96

34 0.43 3.76 4.19

35 0.46 14.07 14.52

36 2.87 31.14 34.01

37 0.02 19.31 19.33

38 0.47 17.41 17.88

39 1.16 4.16 5.32

40 0.37 19.18 19.54

41 0.50 29.56 30.06

42 1.64 34.90 36.53

43 0.66 30.88 31.54

44 0.43 8.35 8.78

45 2.76 20.60 23.36

46 1.97 11.87 13.84

47 0.22 8.74 8.97

48 0.34 8.79 9.13

49 0.12 9.22 9.35

50 1.30 21.96 23.26

51 0.72 5.59 6.31

52 1.43 31.26 32.70

53 0.54 24.11 24.65

54 1.99 8.74 10.73


-108-

55 0.16 17.49 17.65

56 1.24 10.61 11.85

57 0.59 10.38 10.98

58 1.57 53.34 54.90

59 3.60 48.68 52.29

60 1.05 3.79 4.83

61 0.99 18.30 19.29

62 0.16 7.48 7.64

63 0.86 24.16 25.02

64 0.18 33.99 34.17

65 0.22 3.84 4.06

66 1.52 7.66 9.18

67 0.91 3.01 3.92

68 3.30 5.94 9.24

69 1.21 16.88 18.08

70 0.69 83.94 84.63

71 0.65 10.90 11.55

72 0.61 4.77 5.39

73 0.63 46.91 47.54

74 1.29 8.62 9.90

75 2.16 3.87 6.03

76 6.89 9.51 16.40

77 2.91 18.26 21.16

78 2.04 54.21 56.24

79 6.58 75.23 81.82

80 0.50 14.31 14.82

81 0.42 15.61 16.03

82 4.94 16.39 21.33

83 3.34 50.41 53.75

84 6.58 29.75 36.33

85 2.82 7.46 10.28

86 1.00 50.88 51.88


-109-

87 1.24 5.57 6.82

88 1.24 10.47 11.71

89 0.11 2.21 2.32

90 1.20 9.64 10.84

91 0.42 3.30 3.72

92 4.58 11.20 15.78

93 1.33 45.96 47.29

94 0.62 17.20 17.83

95 3.82 29.75 33.57

96 1.16 13.44 14.59

97 1.20 15.05 16.24

98 3.72 12.93 16.65

99 0.33 2.64 2.97

100 1.00 25.63 26.63

101 2.99 25.35 28.34

102 8.24 46.39 54.63

103 8.06 98.76 106.83

104 0.86 49.50 50.36

105 0.60 5.23 5.83

106 4.35 79.04 83.40

107 3.72 49.95 53.68

108 3.07 77.12 80.19

109 0.38 25.63 26.01

110 2.04 62.44 64.47

111 0.50 4.51 5.01

112 1.79 14.31 16.11

113 1.29 3.61 4.89

114 0.76 5.57 6.34

115 5.72 54.70 60.41

116 0.05 22.71 22.75

117 0.40 3.94 4.34

118 2.10 7.24 9.34


-110-

119 1.38 12.12 13.50

120 1.52 57.70 59.22

121 2.59 22.96 25.55

122 3.16 75.23 78.39

Table D2, Distribution of total ESAL of type 4 truck dataset

Fit Input

Function N/A

Shift 2.134919845 N/A

β 22.62907263 N/A

Left X 3.3 3.3

Left P 5.00% 3.28%

Right X 69.9 69.9

Right P 95.00% 95.08%

Diff. X 66.6299 66.6299

Diff. P 90.00% 91.80%

Minimum 2.1349 2.3204


Mean 24.764 24.949

Mode 2.1349 9.3453 [est]

Median 17.82 17.315

Std. Deviation 22.629 21.321

Variance 512.075 450.87

Skewness 2 1.4322

Kurtosis 9 4.7455

Chi-Sq A-D K-S

Test Value 8.23 0.3742 0.05574

P Value 0.6926 > 0.25 > 0.25

Rank 5 3 3


-111-

Figure D1 Exponential distribution result of class 4 truck

Percentage of truck



-112-

Type 5 Truck,


No ESAL of Ax1 ESAL of Ax2 Total ESAL

for each truck

1 1.58 50.08 51.65

2 1.01 18.81 19.82

3 5.66 26.98 32.64

4 1.70 111.77 113.47

5 0.32 10.38 10.70

6 1.29 10.91 12.19

7 0.88 41.47 42.35

8 3.45 33.50 36.95

9 7.05 37.81 44.86

10 4.56 68.77 73.33

11 1.27 43.74 45.01

12 0.85 12.39 13.24

13 0.98 19.94 20.91

14 9.38 126.16 135.54

15 1.77 114.78 116.56

16 3.23 47.04 50.27

17 3.97 26.26 30.23

18 2.37 20.38 22.75

19 0.59 4.07 4.66

20 0.47 5.24 5.71

21 1.08 27.49 28.57

22 3.88 11.70 15.58

23 0.96 5.57 6.53

24 1.85 8.92 10.77

25 0.74 9.93 10.67

26 2.74 8.70 11.44

27 3.16 25.79 28.95

28 1.38 3.16 4.54

29 1.85 44.04 45.89

30 0.44 5.36 5.80

31 1.47 9.46 10.93


-113-

32 1.43 58.71 60.14

33 3.82 12.99 16.81

34 6.00 31.14 37.14

35 2.82 9.38 12.20

36 1.08 26.46 27.54

37 0.55 6.46 7.01

38 3.72 10.26 13.98

39 1.52 8.63 10.15

40 4.94 8.63 13.56

41 10.85 37.28 48.13

42 0.96 8.92 9.88

Table D4, Distribution of total ESAL of type 5 dataset

N/A Fit Input

Function N/A

Shift 3.898117109 N/A

β 26.86906985 N/A

Left X 5.3 5.3

Left P 5.00% 4.76%

Right X 84.4 84.4

Right P 95.00% 92.86%

Diff. X 79.1143 79.1143

Diff. P 90.00% 88.10%

Minimum 3.8981 4.5379


Mean 30.767 31.407

Mode 3.8981 10.802 [est]

Median 22.522 20.367


Variance 721.947 921.54

Skewness 2 1.8889

Kurtosis 9 6.2925


-114-

N/A Chi-Sq A-D K-S

Test Value 3.333 0.4594 0.1156

P Value 0.8526 > 0.25 > 0.25

Rank 1 3 5

Figure D2 Exponential distribution result of class 5 truck

Percentage of truck



-115-

Type 8 Truck,


ESAL of Ax1 ESAL of Ax2 ESAL of Ax3 Total ESAL

No for each truck

1 0.14 16.24 31.60 47.98

2 1.00 30.75 45.29 77.04

3 0.26 4.45 7.04 11.75

4 0.80 30.18 32.78 63.76

5 0.43 13.64 6.94 21.02

6 0.08 44.17 43.60 87.85

7 2.89 13.64 28.41 44.94

8 0.39 5.94 12.45 18.77

9 0.10 5.86 12.03 17.99

10 1.38 21.13 52.35 74.85

11 0.06 24.06 38.52 62.64

12 0.50 18.82 29.86 49.19

13 0.76 51.06 97.28 149.11

14 0.26 44.68 68.36 113.29

15 0.30 2.57 10.06 12.93

16 0.10 18.39 21.63 40.11

17 0.04 22.50 17.39 39.93

18 2.07 4.00 7.22 13.30

19 2.67 46.82 161.56 211.05

20 2.51 49.95 223.37 275.84

21 3.53 49.95 193.21 246.69

22 0.18 38.68 44.79 83.65

23 0.20 18.91 32.30 51.41

24 0.18 11.20 23.70 35.08

25 0.46 56.69 106.67 163.82

26 2.82 24.81 24.49 52.12

27 0.30 2.59 55.70 58.59

28 2.51 11.65 37.94 52.11

29 0.02 8.99 24.81 33.82


-116-

30 2.74 62.98 197.75 263.48

31 0.50 23.22 112.56 136.29

32 0.42 10.33 42.81 53.56

33 1.24 9.91 17.81 28.97

34 0.93 28.52 22.48 51.93

35 0.25 299.06 73.09 372.40

36 0.74 43.43 55.70 99.86

37 6.14 63.53 232.72 302.39

38 0.33 34.69 68.84 103.86

39 0.14 7.91 13.60 21.64

40 6.00 81.00 265.40 352.40

41 0.89 97.24 143.94 242.08

42 0.40 8.02 13.29 21.72

43 0.07 10.19 21.60 31.85

44 0.20 6.71 13.19 20.10

45 1.58 15.42 61.53 78.53

46 1.74 13.61 73.09 88.44

47 0.16 14.31 12.70 27.17

48 0.38 8.38 56.29 65.05

49 0.22 12.44 14.55 27.21

50 0.19 3.07 18.71 21.96

51 0.19 15.05 35.35 50.59

52 0.96 51.34 139.19 191.49

53 0.08 63.53 79.83 143.44

54 1.24 25.91 22.63 49.78

55 0.60 10.76 19.50 30.86

56 2.23 29.13 47.88 79.24

57 0.55 7.46 4.24 12.24

58 1.91 42.61 249.57 294.09


-117-

Table D6, Distribution of total ESAL of type 8 dataset

N/A Fit Input

Function N/A

Shift -3.797897552 N/A

α 1.66375291 N/A

β 80.17603592 N/A

Left X 15.3 15.3

Left P 5.00% 6.90%

Right X 345.3 345.3

Right P 95.00% 96.55%

Diff. X 329.9882 329.9882

Diff. P 90.00% 89.66%

Minimum -3.7979 11.748


Mean 116.994 94.366

Mode 26.301 12.306 [est]

Median 55.818 52.842

Std. Deviation N/A 93.031

Variance N/A 8505.56

Skewness N/A 1.5

Kurtosis N/A 4.2111

N/A Chi-Sq A-D K-S

Test Value 4.069 0.3677 0.07011

P Value 0.8508 N/A N/A

Rank 2 2 1


-118-

Figure D3 Pearson5 distribution result of class 8 truck

P

ercentage of truck



-119-

Type 9 Truck,


ESAL of Ax1 ESAL of Ax2 ESAL of Ax3 Total ESAL

No for each truck

1 7.05 121.24 259.62 387.91

2 1.28 10.78 4.47 16.53

3 3.43 48.59 25.81 77.84

4 7.38 113.25 274.47 395.10

5 0.35 101.85 23.20 125.40

6 6.43 58.73 33.03 98.19

7 7.21 103.43 40.78 151.42

8 1.79 13.10 30.29 45.18

9 3.53 82.99 98.29 184.81

10 3.82 30.37 22.30 56.49

11 0.86 25.63 33.78 60.27

12 2.91 24.54 20.26 47.70

13 2.74 24.81 21.96 49.51

14 9.38 82.99 68.90 161.27

15 0.76 43.43 17.80 61.99


-120-

Table D8-1, First case distribution of total ESAL of type 9 dataset

N/A Fit Input

Function N/A

γ 9.323892348 N/A

β 77.76628271 N/A

α 1.819930141 N/A

Left X 24.7 24.7

Left P 5.00% 6.67%

Right X 401.5 401.5

Right P 95.00% 100.00%

Diff. X 376.7077 376.7077

Diff. P 90.00% 93.33%

Minimum 9.3239 16.532


Mean 145.203 127.97

Mode 48.772 55.424 [est]

Median 87.09 77.84

Std. Deviation N/A 117.46

Variance N/A 12876.04

Skewness N/A 1.4962

Kurtosis N/A 4.0302

N/A Chi-Sq A-D K-S

Test Value 2.8 0.3002 0.1369


Rank 8 2 1


-121-

L ogL ogistic(9.3239, 77.766, 1.8199)

0.0

0.2

0.4

0.6

0.8

1.0

0

50

100

150

200

250

300

350

400

>90.0%

24.7 401.5

@RISK Trial V ersionFor Evaluation Purposes Only

Figure D4-2 LogLogistic distribution result of second case of class 9 truck


Percentage of truck


-122-

Table D8-2, First case distribution of total ESAL of type 9 dataset

N/A Fit Input

Function N/A

Shift N/A N/A

γ -3.877424996 N/A

β 78.13132833 N/A

α 2.913032666 N/A

Left X 24.6 24.6

Left P 5.00% 7.69%

Right X 210.8 210.8

Right P 95.00% 100.00%

Diff. X 186.2504 186.2504

Diff. P 90.00% 92.31%

Minimum -3.8774 16.532


Mean 91.741 87.431

Mode 57.236 55.424 [est]

Median 74.254 61.992


Variance 6662.145 2526.83

Skewness N/A 0.6225

Kurtosis N/A 2.0891

N/A Chi-Sq A-D K-S

Test Value 0.6154 0.3424 0.1603


Rank 4 1 1


-123-


Percentage of truck



-124-

Type 12 Truck,


No ESAL of

Axle1

ESAL of

Axle2

ESAL of

Axle3

ESAL of

Axle4

Total ESAL

for each truck

1 0.22 0.06 8.95 13.34 22.56

2 0.19 17.12 54.92 76.69 148.91

3 0.24 21.03 6.71 12.57 40.55

4 0.18 48.15 6.20 22.30 76.83

5 0.16 40.92 20.46 36.25 97.80

6 0.24 128.72 9.04 23.63 161.64

7 0.21 60.61 22.61 33.18 116.60

8 0.14 47.61 57.97 153.91 259.63

9 0.11 13.92 31.37 50.01 95.42

Table D10-1, Distribution of total ESAL of type 12 dataset

Fit Input

Function N/A

γ -53.32378552 N/A

β 155.2310652 N/A

α 4.280429974 N/A

Left X 24.7 24.7

Left P 5.00% 11.11%

Right X 255.5 255.5

Right P 95.00% 88.89%

Diff. X 230.8073 230.8073

Diff. P 90.00% 77.78%

Minimum -53.324 22.564


Mean 116.77 113.33

Mode 85.569 99.611 [est]

Median 101.91 97.796


-125-


Variance 6626.67 4493.81

Skewness 3.5178 0.7739

Kurtosis 115.2844 3.0981

N/A Chi-Sq A-D K-S

Test Value 0.1111 0.1665 0.1211


Rank 4 1 2

L ogL ogistic(-53.324, 155.23, 4.2804)

0.0

0.2

0.4

0.6

0.8

1.0

0

50

100

150

200

250

300

< >5.0%90.0%

24.7 255.5



Percentage of truck



-126-

Table D10-2, Distribution of total ESAL of type 12 dataset

N/A Fit Input

Function N/A

a 95.87336894 N/A

b 27.01514247 N/A

Left X 16.3 16.3

Left P 5.00% 0.00%

Right X 175.4 175.4

Right P 95.00% 100.00%

Diff. X 159.0889 159.0889

Diff. P 90.00% 100.00%

Minimum -Infinity 22.564


Mean 95.873 95.039

Mode 95.873 95.576 [est]

Median 95.873 96.607

Std. Deviation 49 48.349

Variance 2401.005 2045.46

Skewness 0 -0.1172

Kurtosis 4.2 1.9368

N/A Chi-Sq A-D K-S

Test Value 0 0.1984 0.1357

P Value 1 > 0.25 > 0.1

Rank 1 2 2


-127-


Percentage of truck



-128-

Appendix E

Accumulated ESAL of G206 Highway

The accumulated ESAL of G206 was calculated in the same method

appendix B mentioned. First of all, total ESALs for one year was found. And

then the cumulative Growth Factor (CGF) was adopted from pavement

design (AUSTROADS, 2004) to calculate the total EASLs of pavement

throughout the design period.

The following step shows the ESAL calculation for dataset of G206 Highway:

Step 1:


percentage of heavy vehicle x mean ESAL for heavy vehicle



=5,406,563.38


=4,535,673.26


=20,818,841.18


=2,186,319.59


=2,067,641.51


-129-



=6,073,597.35


=8,245,984.70


=6,504,487.75


=225,971.64


=6,949,538.43



=1,044,121.20


=1,122,628.84





-130-

Table E1, the summary of calculation in 1 year ESAL for G206

ESAL for 1 year


Type 4 5,406,563.38 6,073,597.35 1,044,121.20

Type 5 4,535,673.26 8,245,984.70 1,122,628.84




Total 35,015,038.92 27,999,579.87 2,166,750.04



Total ESAL for design period= 1 year accumulated ESAL of heavy vehicle x CGF


cases are shown in Table E2.

Table E2, the accumulated ESAL for each heavy vehicle type in 15, 30, 40 design period

Total ESAL 1st type Highway 2nd type Highway 3rd type Highway

15 years design 605,760,173.12 484,392,731.73 37,484,775.75

30 years design 1,421,610,579.70 1,136,782,942.67 87,970,051.75

40 years design 2,114,908,350.10 1,691,174,624.07 130,871,702.61

Note: the CGF factor for 15 years design is 17.3, 30 years design is 40.6 and 40 years

design is 60.4, when annual growth rate assumed as 2%


-131-

Appendix F

Calculation of ESAL Comparison of G206

Result of Standard and Actual ESAL comparison for G206 are shown in

Chapter 5, thus the calculation are discuss at here.

Actual ESAL arose from dataset, standard ESAL of China and Queensland

standard are shown below.

Total ESAL=∑ × ESALAADT (F1)

Table F1, AADT of G206 in 2003


% of AADT

No./day 41.55% 10.93% 12.20% 12.25% 3.23% 19.84%

AADT

No./day 4045 1064 1188 1193 314 1932

Table F2, Summary of ESAL analysis result on G206

Vehicle class

ESAL Four Five Eight Nine Twelve

Standard of China 1.0 3.0 5.0 5.0 5.0

Standard of Queensland 0.7 1.1 1.4 1.3 1.3

Mean 24.764 30.767 116.994 91.741 95.873


-132-

Thus, calculation can be carried out by Equation (F1)

For the Standard of China,

Total Heavy vehicle ESAL = 1188x1.0+1193x3.0+314x5.0

= 6337

For the Standard of Queensland,

Total Heavy vehicle ESAL = 1188x0.7+1193x3.0+314x5.0

= 2563

For Dataset,

Total Heavy vehicle ESAL

= 1188x24.764+1193x30.767+314x(116.994+91.741+95.873)/3

= 98007

In conclusion, the calculation result is,

Total ESAL of heavy truck for G206

(∑ × ESALAADT )

China Standard 6337

Queensland Standard 2563

Actual 98007


-133-

Appendix G

Calculation of Pavement service life of G206

Pavement service life analysis is another core part of this thesis, this process

involves many graph and figure. Thus, details of distribution information are

shows in this section.

Type 4 truck,

Table G1, actual service life of pavement when type 4 truck domain entire traffic Resultant Service life

No 15 Yr Design 30 Yr Design

1 1.91 3.82

2 1.06 2.12

3 1.14 2.28

4 1.72 3.44

5 1.25 2.51

6 0.43 0.86

7 0.58 1.16

8 0.37 0.74

9 0.39 0.79

10 0.43 0.86

11 4.94 9.87

12 0.31 0.62

13 0.54 1.07

14 1.38 2.77

15 0.46 0.92

16 0.22 0.45

17 0.75 1.49

18 0.30 0.60

19 0.69 1.38

20 0.93 1.86

21 0.38 0.75

22 1.53 3.05


-134-

23 1.11 2.23

24 1.83 3.67

25 1.11 2.23

26 1.60 3.21

27 0.42 0.84

28 0.88 1.77

29 2.67 5.34

30 2.11 4.21

31 5.56 11.11

32 0.45 0.89

33 0.72 1.43

34 3.58 7.15

35 1.03 2.07

36 0.44 0.88

37 0.78 1.55

38 0.84 1.68

39 2.82 5.64

40 0.77 1.53

41 0.50 1.00

42 0.41 0.82

43 0.48 0.95

44 1.71 3.42

45 0.64 1.28

46 1.08 2.17

47 1.67 3.35

48 1.64 3.29

49 1.60 3.21

50 0.64 1.29

51 2.38 4.75

52 0.46 0.92

53 0.61 1.22

54 1.40 2.80


-135-

55 0.85 1.70

56 1.27 2.53

57 1.37 2.73

58 0.27 0.55

59 0.29 0.57

60 3.10 6.21

61 0.78 1.56

62 1.96 3.93

63 0.60 1.20

64 0.44 0.88

65 3.70 7.39

66 1.63 3.27

67 3.82 7.65

68 1.62 3.25

69 0.83 1.66

70 0.18 0.35

71 1.30 2.60

72 2.78 5.57

73 0.32 0.63

74 1.51 3.03

75 2.49 4.97

76 0.91 1.83

77 0.71 1.42

78 0.27 0.53

79 0.18 0.37

80 1.01 2.02

81 0.94 1.87

82 0.70 1.41

83 0.28 0.56

84 0.41 0.83

85 1.46 2.92

86 0.29 0.58


-136-

87 2.20 4.40

88 1.28 2.56

89 6.46 12.93

90 1.38 2.77

91 4.03 8.07

92 0.95 1.90

93 0.32 0.63

94 0.84 1.68

95 0.45 0.89

96 1.03 2.06

97 0.92 1.85

98 0.90 1.80

99 5.05 10.10

100 0.56 1.13

101 0.53 1.06

102 0.27 0.55

103 0.14 0.28

104 0.30 0.60

105 2.57 5.15

106 0.18 0.36

107 0.28 0.56

108 0.19 0.37

109 0.58 1.15

110 0.23 0.47

111 2.99 5.98

112 0.93 1.86

113 3.07 6.13

114 2.37 4.73

115 0.25 0.50

116 0.66 1.32

117 3.46 6.92

118 1.61 3.21


-137-

119 1.11 2.22

120 0.25 0.51

121 0.59 1.17

122 0.19 0.38

Table G2, Distribution of dataset in G1

N/A Fit Input

Function #NAME? N/A

Shift 0.13128 N/A

b 1.11412 N/A

N/A N/A N/A

N/A N/A N/A

N/A N/A N/A

N/A N/A N/A

Left X 0.188 0.188

Left P 5.00% 4.10%

Right X 3.469 3.469

Right P 95.00% 93.44%

Diff. X 3.2805 3.2805

Diff. P 90.00% 89.34%

Minimum 0.13128 0.14042

Maximum #NAME? 6.4644

Mean 1.2454 1.2545

Mode 0.13128 0.27769 [est]

Median 0.90353 0.86663


Variance 1.2413 1.4173

Skewness 2 1.9578

Kurtosis 9 7.1512


-138-

Service life at different percentage level (years) Figure G1 Exponential distribution result of class 4 truck design for 15 years

Percentage of truck


-139-

For type 5 truck,



1 0.29 0.58

2 0.76 1.51

3 0.46 0.92

4 0.13 0.26

5 1.40 2.80

6 1.23 2.46

7 0.35 0.71

8 0.41 0.81

9 0.33 0.67

10 0.20 0.41

11 0.33 0.67

12 1.13 2.27

13 0.72 1.43

14 0.11 0.22

15 0.13 0.26

16 0.30 0.60

17 0.50 0.99

18 0.66 1.32

19 3.22 6.43

20 2.62 5.25

21 0.53 1.05

22 0.96 1.93

23 2.30 4.59

24 1.39 2.78

25 1.41 2.81

26 1.31 2.62

27 0.52 1.04

28 3.31 6.61

29 0.33 0.65


-140-

30 2.58 5.17

31 1.37 2.74

32 0.25 0.50

33 0.89 1.78

34 0.40 0.81

35 1.23 2.46

36 0.54 1.09

37 2.14 4.28

38 1.07 2.15

39 1.48 2.96

40 1.11 2.21

41 0.31 0.62

42 1.52 3.03


N/A Fit Input

Function N/A

Shift 8.94E-02 N/A

b 0.89489 N/A

N/A N/A N/A

N/A N/A N/A

N/A N/A N/A

N/A N/A N/A

Left X 0.135 0.135

Left P 5.00% 7.14%

Right X 2.77 2.77

Right P 95.00% 95.24%

Diff. X 2.6349 2.6349

Diff. P 90.00% 88.10%

Minimum 0.08936 0.11067


Mean 0.98425 1.0056


-141-

Mode 0.08936 0.34060 [est]

Median 0.70965 0.73704


Variance 0.80083 0.68154

Skewness 2 1.2235

Kurtosis 9 3.8104

Percentage of truck

Service life at different percentage level (years) Figure G2 Exponential distribution result of class 5 design for 15 years truck


-142-

For type 8 truck,



1 0.31 0.63

2 0.19 0.39

3 1.28 2.55

4 0.24 0.47

5 0.71 1.43

6 0.17 0.34

7 0.33 0.67

8 0.80 1.60

9 0.83 1.67

10 0.20 0.40

11 0.24 0.48

12 0.30 0.61

13 0.10 0.20

14 0.13 0.26

15 1.16 2.32

16 0.37 0.75

17 0.38 0.75

18 1.13 2.26

19 0.07 0.14

20 0.05 0.11

21 0.06 0.12

22 0.18 0.36

23 0.29 0.58

24 0.43 0.86

25 0.09 0.18

26 0.29 0.58

27 0.26 0.51

28 0.29 0.58

29 0.44 0.89


-143-

30 0.06 0.11

31 0.11 0.22

32 0.28 0.56

33 0.52 1.04

34 0.29 0.58

35 0.04 0.08

36 0.15 0.30

37 0.05 0.10

38 0.14 0.29

39 0.69 1.39

40 0.04 0.09

41 0.06 0.12

42 0.69 1.38

43 0.47 0.94

44 0.75 1.49

45 0.19 0.38

46 0.17 0.34

47 0.55 1.10

48 0.23 0.46

49 0.55 1.10

50 0.68 1.37

51 0.30 0.59

52 0.08 0.16

53 0.10 0.21

54 0.30 0.60

55 0.49 0.97

56 0.19 0.38

57 1.23 2.45

58 0.05 0.10


-144-


N/A Fit Input

Function N/A

Shift 3.48E-02 N/A

b 0.3182 N/A

N/A N/A N/A

N/A N/A N/A

N/A N/A N/A

N/A N/A N/A

Left X 0.051 0.051

Left P 5.00% 6.90%

Right X 0.988 0.988

Right P 95.00% 93.10%

Diff. X 0.9369 0.9369

Diff. P 90.00% 86.21%

Minimum 0.03479 0.04028


Mean 0.35299 0.35848

Mode 0.03479 0.28948 [est]

Median 0.25535 0.28392


Variance 0.10125 0.09685

Skewness 2 1.3749

Kurtosis 9 4.2674


-145-

Figure G3 Exponential distribution result of class 8 truck design for 15 years truck

Percentage of truck



-146-

For type 9 truck,



1 0.04 0.08

2 0.91 1.81

3 0.19 0.39

4 0.04 0.08

5 0.12 0.24

6 0.15 0.31

7 0.10 0.20

8 0.33 0.66

9 0.08 0.16

10 0.27 0.53

11 0.25 0.50

12 0.31 0.63

13 0.30 0.61

14 0.09 0.19

15 0.24 0.48

Table G8-1, Distribution of dataset in G7

N/A Fit Input

Function N/A

Shift 2.53E-02 N/A

b 0.19057 N/A

N/A N/A N/A

N/A N/A N/A

Left X 0.035 0.035

Left P 5.00% 0.00%

Right X 0.596 0.596

Right P 95.00% 93.33%

Diff. X 0.5611 0.5611

Diff. P 90.00% 93.33%


-147-

Minimum 0.02526 0.03797


Mean 0.21583 0.22854

Mode 0.02526 0.10390 [est]

Median 0.15736 0.1927


Variance 0.03632 0.0423

Skewness 2 2.2346

Kurtosis 9 8.0636

Expon(0.19057) Shift=+0.025260

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

>5.0%90.0%

0.035 0.596


Figure G4-1 Exponential distribution result of class 9 truck design for 15 years truck

Percentage of truck



-148-


(Extreme data eliminated)

N/A Fit Input

Function N/A

Shift 6.76E-02 N/A

b 0.17664 N/A

N/A N/A N/A

N/A N/A N/A

N/A N/A N/A

N/A N/A N/A

Left X 0.077 0.077

Left P 5.00% 0.00%

Right X 0.597 0.597

Right P 95.00% 92.31%

Diff. X 0.5201 0.5201

Diff. P 90.00% 92.31%

Minimum 0.06758 0.08117


Mean 0.24422 0.2578

Mode 0.06758 0.091079 [est]

Median 0.19001 0.24197

Std.

Deviation 0.17664 0.21427

Variance 0.0312 0.04238

Skewness 2 2.2628

Kurtosis 9 7.7468


-149-


Percentage of truck



-150-

For type 12 truck,



1 0.66 1.33

2 0.10 0.20

3 0.37 0.74

4 0.20 0.39

5 0.15 0.31

6 0.09 0.19

7 0.13 0.26

8 0.06 0.12

9 0.16 0.31


N/A Fit Input

Shift 4.05E-02 N/A

b 0.15561 N/A

Left X 0.0485 0.0485

Left P 5.00% 0.00%

Right X 0.5067 0.5067

Right P 95.00% 88.89%

Diff. X 0.4582 0.4582

Diff. P 90.00% 88.89%

Minimum 0.04048 0.05777


Mean 0.19609 0.21338

Mode 0.04048 0.10330 [est]

Median 0.14835 0.15338


Variance 0.02422 0.03267

Skewness 2 1.6536


-151-

Expon(0.15561) Shift=+0.040484

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

>5.0%90.0%

0.0485 0.5067


Figure G5-1 Exponential distribution result of class 8 truck design for 15 years truck Service life at different percentage level

Percentage of truck


-152-


(Extreme data eliminated)

N/A Fit Input

Function N/A

Shift 7.53E-02 N/A

b 0.14004 N/A

Left X 0.0825 0.0825

Left P 5.00% 0.00%

Right X 0.4948 0.4948

Right P 95.00% 87.50%

Diff. X 0.4123 0.4123

Diff. P 90.00% 87.50%

Minimum 0.0753 0.0928


Mean 0.21533 0.23284

Mode 0.0753 0.16430 [est]

Median 0.17236 0.15529


Variance 0.01961 0.03334

Skewness 2 1.5604

Kurtosis 9 4.0532


-153-


Percentage of truck



-154-

Appendix H

Calculation of Actual Service Life

Actual service life calculation is combined all traffic components in the traffic

stream, thus the percentage of different vehicle type contributes in traffic is a

factor of pavement service life. In this section, the actual service life design

for 15, 20, 30, and 40 years were calculated.

Table H1, the percentage of each vehicle contributed in traffic of G206 Passenger Class 3 Class 4 Class 5 Class8-12 Coach

% of AADT

No./day

41.55% 10.93% 12.20% 12.25% 3.23% 19.84%

DDAT 4045 1064 1188 1193 314 1932

No./day

Table H2, mean of each design period when traffic dominated by same vehicle type

Mean of Design Period

Vehicle Type 15 years 20 years 30years 40years

4 1.2545 1.67267 2.5091 3.34533

5 1.0056 1.3408 2.0111 2.6816

8 0.35848 0.47797 0.71685 0.95595

9 0.2578 0.34373 0.51561 0.68747

12 0.23284 0.31045 0.46567 0.62091

Calculation may follow Equation (H1),

Pavement service life=∑ Mean of pavement life x percentage of vehicle (H1)

Due to passenger, type 3 truck and coach does not affect by overloading

problem, thus Mean of pavement life is adopted the original design life.


-155-

For 15 years design,

Pavement service life= 41.55%x15 + 10.39%x15 + 12.20%x1.2545 +

12.25x1.0056 + 3.23x(0.35848+0.2578+0.23284)/3 + 19.84x15

= 11.1334 years



12.25x1.3408 + 3.23x(0.47797+0.34373+0.31045)/3 + 19.84x15

= 14.8445 years



12.25x2.0111 + 3.23x(0.71685+0.68747+0.46567)/3 + 19.84x15

= 22.2668 years



12.25x2.6816 + 3.23x(0.95595+0.51561+0.62091)/3 + 19.84x15

= 29.689 years

The summary of calculation is shown in H3

Table H3, Comparison of actual and design service life

Years

Design service life 15 20 30 40

actual service life 11.1 14.8 22.3 29.7

Percent Reduction (%) 26%


-156-

Appendix I

Calculation of Net Present Value of Investment

Calculations of Net Present Value (NPV) of investment are shown in the

following steps.

Calculation of present and future worth cost for design:

Step 1,

Equation to calculate the NPV of investment

∑=

+ ×+×+=m

jjjmCcLC TrFPfRTrAPfMCP

11 ),,/(),,/( (I1)

[ ]1

1

)1(

1)1(1),,/(

+

+

+

−++ =

mT

mT

rr

rmTrAPf jT

rjTrFPf)1(

1),,/(+

=

When PLC= pavement life-cycle present worth cost for a give M&R plan

($/m2)

Cc= initial construction cost of original pavement structure ($/m2)

Mc= annual routine maintenance and added user cost ($/m2)

Rj= future rehabilitation cost of the jth cycle (j= 1, 2…m)

Tm+1= length of life-cycle analysis period in years

r= annual interest rate

m= number of deployed major rehabilitation cycles in an analysis period

Tj= scheduled rehabilitation time of the jth cycle in years

ƒ (P/A, r, Tm+1)= factor converting a uniform annual cost to a present one

ƒ(P/F, r, Tj)= factor converting a future cost to a present one

Source: Optimum Flexible Pavement Life-Cycle Analysis Model, Khaled A. Abaza, P.E.,

Journal of Transportation Engineering / November / December 2002


-157-

In this calculation, annual interest rate is assumed as 4.8%, and the

construction, maintenance and rehabilitation costs are provide by APCD, the

information are shown blow:

Total Construction Maintenance Total Rehabilitation

$ 386,281.00 $ 2643.7 $ 1797.7

Remark: currency unit for maintenance and rehabilitation cost in this table is US$ per

kilometer

ƒ( P/A ,r ,Tm+1)= 15.7292203 when T=30

10.5214136 when T=15

17.6395352 when T=40

12.6762836 when T=20

ƒ(P/F ,r ,T j)= 15.7292203 when T=30

10.52141362 when T=15

17.63953524 when T=40

12.67628361 when T=20

Net present value of pavement without overloading problem:

Net Present Value for 15 years:

Construction cost = 386,281.00

Total Maintenance cost = 10.521 x 2,643

= 2,490,040.09

Total Rehabilitation cost = 10.521 x 1,797

= 1,693,212.19

PLC for 15 year design= 4,569,533.27


-158-

Net Present Value for 30 years



= 3,722,540.57


= 2,531,305.06

Thus,





= 4,174,643.39


= 2,838,732.24

Thus,





= 3,000,020.29


= 2,039,995.64



-159-

Net present value of pavement with overloading problem:

Refer to Appendix H, 26% reduction has found, thus the NPV of investment

were follow this reduction to estimate. Also the actual maintenance and

rehabilitation cost were used in the calculation.

Total Construction Maintenance Total Rehabilitation

$ 386,281.00 $ 1541 $ 9872

Remark: currency unit for maintenance and rehabilitation cost in this table is US$ per

kilometer

ƒ(P/A ,r ,T m+1)= 8.394356657 when T=11

10.52141362 when T=15

13.40638258 when T=22

15.48422288 when T=29

ƒ(P/F ,r ,T j)= 8.394356657 when T=11

10.52141362 when T=15

13.40638258 when T=22

15.48422288 when T=29

Net present value of pavement with overloading problem:

Present value for 11 years (Original 15 years):



= 1,158,004.19


= 7,418,440.84

Thus,



-160-

Present value for 15 years(Original 20 years):



= 1,451,432.38


= 9,298,209.23





= 1,849,414.77


= 11,847,775.84





= 2,136,053.50


= 13,684,049.42



-161-

List of References

1. Abaza, Khaled A. 2002. Optimum Flexible Pavement Life-Cycle Analysis Model. Journal of transportation Engineering / November / December 2002:9.

2. Abaza, Khaled A., A. M. Suara. 1995. The structural wear of road pavements: an assessment of the fourth power law on the A1(M) motorway in County Durham. Engrs Transp.,:9.

3. Abaza, Khaled A., J.Middleton. 1994. The effect of dynamic loading on road pavement wear: a study of the relationship between road profiles and pavement wear on an instrumented test road. Engrs Transp.,:12.

4. Addis, R. R. 1993. International collaboration in accelerated pavement testing. Engrs Transp.,:12.

5. Aman Kishore, Rod Klashinsky. 2000. Prevention of Highway infrastructure damage through commercial vehicle weight enforcement. In Annual Indian Road Congress (IRC) Session. Calcutta.

6. Arnim H. Meyburg, Jean-Daniel M. Saphores, Richard E. Schuler. 1997. The economic impacts of a divisible-load permit system for heavy vehicles. Transpn Res.-A 32:13.

7. Austroads. 2004. Pavement design : a guide to the structural design of road pavements, Pavement technology series.

8. B. Al Hakim, A. C. Collop, N. H. Thom. 2001. The use of weigh-in-motion data in pavement design. Transport 147:10.

9. Batelaan, Justin. 2003. A thin-walled, dual-curved, steel band wheel with inherent spring suspension for use on heavy vehicles. Thin-Walled structure 41:12.

10. Brian Taylor, Art Bergan, Norm Lindgren, Curtis Berthelot. 2000. The importance of commercial vehicle weight enforcement in safety and road asset management. In Traffic Technology International 2000, Annual Review. Canada.


-162-

11. Ceallach Levins, Anthony Ockwell. 2000. Truck: the road to ruin or

increased efficiency. http://www.oecdobserver.org/news/fullstory.php/aid/236/Trucks:_the_road_to_ruin_or_increased_efficiency_.html.

12. Cheng Ling-gang, Zhou Hai-tao. 2001. A Review of Theories and Methods of Calculation of National Economic Benefits from Highway Construction Projects. Journal of Highway and Transportation Research and Development 18:4.

13. Comprehensive Truck Size and Weight (TS&W) Study. 1995. edited by B. Team. Ohio: Federal Highway Administration U.S. Department of Transportation.

14. Handbook of Highway Maintenance and Rehabilitation. 2002. Beijing: China Communications Press.

15. Huang, Wen, Li Xu-hong, Ju Peng, He Jie. 2005. Site Survey and Analysis of Highway Trucks Overloading Status Quo in Anhui. Journal of Eastern Asia Society for Transportation Studies 6:14.

16. James G. Strathman, Greg Theisen. 2002. Weight Enforcement and Evasion: Oregon case study. Oregon: Center for Urban Studies College of Urban and Public Affairs Portland State University.

17. Johnsson, Richard. 2004. The cost of relying on the wrong power-road wear and the importance of the fourth power rule (TP446). Transport Policy 11:9.

18. Kanok-Nukulchai, Worsak. 1997. Heavy weight champion of the road. http://www.thaiengineering.com/viewnew.php?id=231.

19. Karan Kapoor, Jorge A. Prozzi. 2005. Methodology for Quantifying Pavement Damage Caused by Different Axle and Load Configurations. Texas: Center for Transportation Research University of Texas at Austin.

20. Koji Tsunokawa, Riaz-ul-Islam. 2003. Optimal Pavement Design and Maintenance Strategy for Developing Countries: An Analysis using HDM-4. The International Journal of Pavement Engineering 4:16.

21. Koji Tsunokawa, Riaz-ul-Islam, Guan Changyu. 2002. Optimal Strategies for Highway Pavement Management in Developing Countries. Computer-Aided Civil and Infrastructure Engineering 17:9.

22. Noppadol Piyatrapoomi, Arun Kumar. 2004. Review of the Literature:


-163-

Maintenance and Rehabilitation Costs for Roads (Risk-based Analysis). Melbourne: RMIT University.

23. Overloading Issues. 2004. Railroad Association of South Africa 2002 [cited Jul 12 2004]. Available from http://www.rra.co.za/overloading.shtml.

24. Overloading Study for the Proposed Anhui Highway Project II. 2004. In Interim Report II, edited by T. C. o. S. University. Anhui: Anhui Province Communications Department.

25. Overloading Study for the Proposed Anhui Highway Project II. 2004. In Interim Report I, edited by T. C. o. S. University. Anhui: Anhui Province Communications Department.

26. Rhodes, A. H. 1997. Highway pavements: use of tapered sections to extend design life. Engrs Transp.,:8.

27. Rob Bushman, Curtis Berthelot, Brian Taylor. 2003. Commercial Vehicle Loading in an Urban Environment. In Transportation Association of Canada. Canada.

28. Roy Suptphen, Rick W. Warner. 2003. Commercial Vehicle Accident Reconstruction and Investigation. Introduction, http://www.lawyersandjudges.com/productdetails.cfm?pc=1092.

29. Squires, Michael. 2001. Station Closures: Truck inspections decline. Las Vegas Review-Journal:3.

30. Strathman, James G. 2001. Economics of Overloading and the Effect of Weight Enforcement. Portland: Center for Urban Studies College of Urban and Public Affairs Portland State University.

31. Technical Report: Proposed Harmonized System for Vehicle Overload Control. 2003. Gaborone: Chemonics International, Inc.

32. Technical Specifications for Maintenance of Highway Asphalt Pavement. 2001. edited by J. P. H. Department. Beijing: Transport Department of The Peoples Republic of China.

33. Technical Specifications of Cement Concrete Pavement Maintenance for Highway. 2001. edited by J. P. H. Department. Beijing: Transport Department of The Peoples Republic of China.

34. TranSafety, Inc. 1997. Oregon Study Looks at Potential Pavement Damage from High-Pressure Truck Tires and Single-Tired Axles. U.S.


-164-

ROADS, Road Engineering Journal:4.

35. TranSafety, Inc. 2000. Axle to Grind: Driving Questions About the Limits to Place on Trucks Traveling Our Roads. U.S. ROADS, Road Engineering Journal:8.

36. Williams, A. R. 1998. Vehicle, tyre, road intercation-the vision for the future. Engrs Transp.,:6.

37. Xiao-ming, huang. 2003. Concrete Pavement Design. Beijing: China Communications Press.

Documents

Truck Overloading Study in Developing Countries and ...eprints.qut.edu.au/28561/1/Ying_Chan_Thesis.pdf · Conclusion and Recommendation 7.1 ... Traffic volume of truck at bypass route