Determining a Threshold of the Damaged Road Which Categorized to Fail of Service In Accordance with Road User Satisfaction

Determining a Threshold of the Damaged Road Which Categorized to Fail of Service In Accordance with Road User Satisfaction

Marsinta Simamora 1) And Jati Utomo Dwi Hatmoko 2)

1Civil Eng.Depart, Kupang State Polytechnic, East Nusa Tenggara, Indonesia.

2Civil Eng.Depart, Diponegoro University, Central Java, Indonesia.

Abstract

The non-toll national road (ntNR) management in Indonesia indicated that it is less optimal. One of the indicators is the increase of the road management budget which is not in line with the increase of the significant benefit. One of the alleged as a possible cause is no difference between the roads which severely damaged and failed, so it is difficult to apply laws and regulations that exist. This study aims to determine a threshold of damage road which failed of serviceability function in accordance with road user satisfaction. The research was conducted using a qualitative approach. Respondents' opinions against the service on the damaged road segments were measured using a Likert scale, which is adapted to service standards of ISO2631, while the value of damage was measured using the standard AASHTO 1993. The result showed the road that was failed of service was at a value CDV > 87.01 or PCI <12.99. By obtaining a threshold of the damaged road which failed of service, then, the road assessment system can be improved by adding the label "failed". As the implication of this study, the Law of Indonesian Republic No.18/1999 and the Government Regulation of Indonesian Republic No.29/2000 can be applied in the ntNR management system.

Keywords: management, damage road, failure, serviceability, threshold, and road user.

1. Introduction

   The non-toll national road (ntNR) management in Indonesia indicated that it is less optimal. One of the indicators is the increase in the budget for road management is not in line with the increase of the significant benefit. Based on the data of Public Communication Center at the Department of PublicWorks in 2013, the cost per km of ntNR management in 2011and 2012 was IDR 725,304,900 and IDR 1,045,890,329, respectively. If we assumed that in general ntNRs have 2 lines and 1US$=IDR 9,500 (at the time ago), then ntNR management costs per km per lane in 2011 and 2012 was US$ = 38.174

   and US$ = 55.047, respectively. Pavement condition index (PCI) before the treatment in 2011 and 2012 was 50.16 (medium category) and 50.13 (medium category), respectively. After the treatment, PCI value in 2011 and 2012 was 50.13 (medium category) and 52.67 (medium category), respectively.

   The change of the the PCI after treatment on the road can be used as a indicator to confirm whether the benefits derived optimal or not (Khurshid et al., 2011; Dong and Huang, 2012; Fwa and Farhan, 2012). By using that concept, in 2011 the benefit gained almost nothing. That can be seen from the PCI after treatment was lower than before, where a decline in PCI was at 0.03. While in 2012, the benefit of treatment was at 2.54, there was an increase in the PCI but it was not optimal.

   Jimenez and Mrawira (2009) mentioned that in order to increase the PCI from 60 to 80 in the class of hight raffic category performed minor maintenance actions at a cost of US$ 56,000 per km/lane. When compared to the cost of ntNR management on average per km/lane in 2012 which was almost equal to the cost of handling standard proposed by Jimenez and Mrawira (2009), should have obtained a minimum benefit value of 20 (80- 60), but the fact showed that the benefit obtained only 2.54, so far from expectation which is 20.

   The question is why the ntNR management is not optimal. One of the alleged as a possible cause is no difference between the roads which severely damaged and failed. At present, in the road evaluation system, the road condition is only grouping in 4 categories, namely: 1) good, 2) moderate, 3) light damage, and 4) severely damaged. This situation might have an impact on: 1) the possibility of errors on road handling (e.g. reconstruction should be done but in fact should not), 2) can never be assessed against the accountability of road damage, in accordance with law and regulation that were exist, because the Law of Indonesian Republic (LOIR) No.18/1999 and the

   Government Regulation of Indonesian Republic (GROIR) No. 29/2000 state that only the buildings that "failed building" have the responsibility such as sunction. In fact, bad road may have performed below minimum standards, which are often depicted with terminal performance index (IPt) [Zhang et al., 2010]. Noon (2009) mentioned that performance under the minimum acceptance is categorized fail.

   Indeed, the decline in road condition (deterioration) can not be avoided and inevitable (Hudson et al., 1987; Zhang et al., 2012). Pavement deterioration is influenced by three main factors, namely: 1) pavement design and materials (surface layer, base, sub base layers, and subgrade, 2) vehicles load (traffics), 3) environment, and 4) maintenance (Hong and Prozzi, 2010; Al-Mansour et al., 1994). Even Hudsonet al. (1987) mentioned that the failure may occure due to naturally of aging, where the structural capacity decreases to exceed a minimum threshold that resulted in the destruction of.

   However, the damage of the road pavement should be accountable and should not go unpunished. Never again use the principle if it breaks do repair. Castro and Fernandes (2004) mentioned in getting more objective and accountable information needed investigations regarding to: i) what is going on, ii) where they are occurred,iii) how it happened, iv) who or what caused it, v) who is responsible for what happened, vi) what is the cost of repair or reinstatement, and vii) what compensation imposed on the parties concerned. In NntR management system, the principle of Castro and Fernandes (2004) is considered to be able tracing such as the cause, the parties concerned and what the impact of the NntR management which is less optimal.

   Unfortunately, at this time, the existing evaluation system yet can be used to assess the accountability of road damage in accordance with the existing rules. That is caused due to the damaged roads are not mentioned as a problem in legal systems in Indonesia, especially by law of construction services. According to LOIR No.18/1999 and GROIR No.29/2000, that building can be sanctioned when the building can be categorized into building failure. Therefore, the accountability of heavely damaged road cannot be audited during the heavily damaged road cannot be categorized as part of building failure. For this reason, it is needed the effort to determine the characteristics of road damage as part of building failure.

   The system thinking in formulating the characteristics of road damage as part of building failure is using the pavement performance approach

   as described in pavement design. According to Zhang et al. (2010), the road pavement performance is consisted of the structure and serviceability performance. Pavement service performance is a form of pavement services to users which is affected by the intensity of the holes, cracking and rutting, while the structure performance is the level of support to the traffic load (Mustafavi et al., 2013; Mariani et al., 2012). Whilst, Parket al. (2007) mentioned that the road is how the service can provide a sense of security and comfort for the rider at design speed.

   By using the pavement performance approach is expected the assessment on the severely damaged road can fulfill the rules of a domain ontology for construction knowledge (DOCK). According to El- Diraby (2013), DOCK is a constructive and pragmatic claim to knowledge of construction. DOCK aims to provide a platform to build a form or level ontology; facilitating the construction of information system on existing fundamental ontology according domain. Based on the DOCKs principle (El-Deraby, 2013), the heavily damaged road will be evaluated regarding to: 1) the damaged road that failed of service and 2) the damage road that failed of structure, in accordance with the principles of pavement design (AASHTO, 1993 ). This study aims to determine a threshold of the damaged road that failed to serve using road users satisfaction.

   The criterion used to failed of service is the performance which it is under of minimal acceptance (Noon, 2009). Simamora and Hatmoko (2013) revealed at least 33.3% experts agreed that riding unsafe and uncomfort were as boundaries of failed of service. Park et al (2007) mentioned that the road which is not safe and convenient used driving at the design speed that is categorized as functional failure. In the other hands, there are several indicators related to serviceability function, i.e: safety and comfort (Park et al., 2007; Regulation of the Public Works Minister of Indonesia Republic No.14/PRT/M/2010), and satisfaction (Regulation of the interior Minister of Indonesia Republic No. 6/2007). In the previous study, to explain satisfaction or ride comfort was used the intensity of vibration when driving (Yang et al., 2009; Falou et al., 2003;Uys and Els, 2007; Els, 2005; Cantisani and Loprencipe, 2010; Corriere and Vincenzo,2012).

2. Literatur Review

In this paper, several topics had been reviewed regarding to: 1) present serviceability index (PSI) according to AASHTO, 1993, 2) PSI value on service level, and 3) the relationship between service level

(according to he road user satisfaction) and the level of damage road.

PSI according to AASHTO, 1993

Generally, the method of flexible pavement component analysis is using a nomogram flexible pavement design (AASHTO, 1993;Pt-T-01-2002-B). Before using the nomogram is usually determined:

1. Traffic estimates in future periods (W18), the accumulation of the load at the end of life design

2. Reliability (R)

3. Overall standard deviation (S0)

4. Resilent modulus effective subgrade

5. Design serviceability loss (PSI = IP0-IPt) Pavement thickness in these guidelines are based on

Santos and Ferreira (2012) described a set of value of IPo (PSIo) and IPt (PSIt) in accordance with roads classes as showed at Table 1.

Table 1. A set of value of PSIo and PSIt

Source: AASHTO 1993 (Santos and Ferreira, 2012)

Generally, the cognitive mean of the IP can be seen in Table 2, in which according to Hudson et al. (1987) that the IP is equal to the PSI.

Table 2. Relation between Serviceability and

P avement (surface) index (PI)

the relative strength of each pavement layer by using

No Pavement (surface)

Serviceability

Eq. 1, as follows :

SN = a1D1 + a2D2m2 + a3D3m3 (1)

where: SN = structural number; a1, a2, a3 = coefficient of relative strength of pavement materials; D1, D2, D3 = thickness of each layer of pavement; m2, m3 = drainage coefficient; Number 1,2,3 respectively for the surface layer, base layer, and subbase layer.

Besides using a nomogram, SN can also be calculated using Eq.2 (AASHTO 1993; Zhang et al., 2010; Pavement interactive, 2008 dan Pt T-01-2002-

1. as follows:

   LogW18 = ZR S0 + 9,36 Log SN + 1 0,20 +

   Log PSI

   . index (PI)

   1. 4 5 Very good

   2. 3 4 Good

   3. 2 3 Moderate

   4. 1 2 Bad

5 0 1 Very bad Source : Sukirman (1992) and Pt-T-01-2002-B

Serviceability indicators

Regulation of interior minister of Indonesia Republic No. 6/2007 state that the service indicator is customer satisfaction. In terms of the road user, the customers were road users. While, Zeithaml et al. (1990) stated that customer satisfaction can be measured by several variables as in Table 3.

Corresponding between the PSI values and level of

0,40+ ,

0,40+ ,

0 f

0 f

IP IP 1094

SN +1 5 19

+ 2,32LogMR 8,07 (2)

serviceability

Several previous studies have performed to

where: W18= estimated number of equivalent standard axle loads 18-kip; ZR = normal standard deviation as a function of the confidence level (R), ie, assuming that all input parameters used are mean values (Sebayang et al., 2008); S0= standard error of traffic prediction and performance; MR= effective modulus of subgrade; PSI= IP0 -IPt = loss of service

= Difference between the initial design serviceability index (IP0) and terminal design serviceability index (IPt) [Pt T-01-2002-B]; IPf= failure serviceability index (Pt T-01-2002-B) = 1.5. Usually, IP0 = 4.2; IPt

= 2 (Pt T-01-2002-B); PSI = 4.2 – 2 = 2.2

PSI value paired to level of service

measure the road service using objective and subjective approach as shown in Table 4.

Determination of damage road value

Suswandi et al. (2008) said that PCI can be used to evaluate the road condition in supporting of decision making. Mentioned that the severity level is the damage level of each type of damage that it is classified into low severity level (L), medium severity level (M) and high severity level (H). PCI is calculated by the following terms: 1) determining density of damaged, 2) determining the deduct value (DV), 3) determining the total of deduct value (TDV), and 4) determining the corrected deduct value

In this section the intent of the PSI value is presented as a service received by road users.

(CDV).

= 100% (3)

PSI values and services

where:Ad = the total of damage for each unit of measurement segment (m2); As = area of unit of measurement segment (m2).

Table 3. Dependent and explanatory variables of customer satisfaction

Table 4. PSI (IP) value paired with objective/subjective measures

DV is value reduction for each type of damage obtained from the relationship between the density curve and deduct value. Deduct value is also distinguished by degree of damage for each damage type. TDV is the total value of individual deduct value for each type of damage and the level of the damage to a unit of study. CDV obtained from the relationship between the curve of the TDV and CDV values with selecting of curve and corresponding to the number of individual deduct value that have a value greater than 2. If the value of CDV is known, then the PCI value for each unit can be determined by the Eq.4:

PCI(s) = 100 (4)

where: PCI(s) = pavement condition index for each segment, CDV = corrected deduct value for each segment. For over all value of PCI used Eq.5.

Relation between Level of Service and Road Surface Damage

According to Hudson et al. (1987), the relation between the PSI and the value of road conditions is shown by Eq.6.

PSI = C + A1R1 + + B1D1 + B2D2 +

± e (6)

where; C = coefficient (5.03 to 5.41 for flexible pavement and rigid pavement); A1 = coefficient (-

1.91 to -1.80 for flexible pavement and rigid pavement); R1 = roughness = log (1 + SV), SV = average slope variation obtained with tools CHLOE profilometer B1 = coefficient (-1.38 to 0 for flexible pavement and rigid pavement); D1 = function rutting (RD), RD = rutting depth; B2 = coefficient (-0.01 to –

0.09 for flexible pavement and rigid pavement; D2 = function deterioration (C + P); C + P = amount of cracking and patching obtained with procedures

PCI = PCI ()

N

(5)

AASHO (American Association of State Highway Officials) Road Test; e = error

Where: PCI = value overall PCI; PCI (s) = Value PCI for each segment; N = Number of segments.

0 0

Figure 1. Significance of PCI and PSI on pavement evaluation (Hudson et al., 1987)

In Fig. 1, we can see that there is a relation between the PCI and PSI in which PCI describes the physical condition of the road and PSI describes the service level.

1. Research Method

   The study was conducted through two main terms, ie: 1) determining the damage value (use Eq.3,4, and 5) used evaluating roads level of service,

   2) determining the level of service on each road segment (there are 4 segments) with different levels of damage. Term of analyses: 1) testing the validity and reliability of the research instrument, 2) determining the relationship between the damaged road and the composite service, 3) determining the damaged road value that was failed of service.

   .

   Research operational variables

   The operational variables in this study can be seen in Table 5. The basic relations between the variables could be described as follows:

   1. Yi = f(Yi1, Yi2, Yi3); Yi, Yi1, Yi2, Yi3 in scale 1-5

   2. CDV = Xi = f (Xi1, Xi2, Xi3); density of Xi1, Xi2, Xi3 in (0-100%)

   3. Yi = f (Xi)

2. Results and Discussion

   Validity and Reliability of research instrument

   To determine a accurate and stable research instrument were conducted testing against the validity and reliability. The tests were performed for a series of research instruments using 41 respondents who were picked by random sampling. A number of respondent in the trial,suggested were about 20-30 people (Muhidin and Abdurahman, 2007).

   Table 5. A series of research operational variables

   Dimension Operational variable Notation Measurement

   standards

   Dep. CDV Xi AASHTO,

   Damaged

   Variable

   1993

   road

   Ind.

   Variables

   PH Xi1

   CRACK Xi2 RUT Xi3

   Dep. Satisfaction Yi AASHTO,

   Variable 1993;

   Intensity of vibration

   Yi1

   ISO 2631;

   Serviceability

   Ind.

   (tangiable)

   Zenithaml et

   Variables

   Safety and comfort

   Yi2

   al. (1990)

   (reliability/assurance) Speed (assurance) Yi3

   Instrument validity test

   In Table 6, it can be seen the results of the validity test. The validity of the instrument was tested using a student-t by comparing t count and t table. The results shown that t count for Yi, Yi1 and Yi2 are 10 385; 11,124, and 6.516, respectively. Those were greater than t table = 2.022, while t count for Yi3 is 1,681 that is lower than t table. When t count > t table, the instrument is valid and if t count < t table that is invalid. Based those requirement, Yi (service/satisfaction), Yi1 (vibration), and Yi2

   (safety and comfort) are the valid instrument and Yi3 (speed) is the invalid instrument.

   Instrument reliability test

   Reliability of a valid instrument can be measured using Cronbach Alpha (r) as can be seen in Table 7. Accordance with the results of validity and reliability test can be mentioned that the variables which are valid and reliable those are:

   1. Yi (satisfactory)

   2. Yi1 (intensity vibration)

   3. Yi2 (safety and comforts)

      Table 6. The instrument validity test

      Variable	Validity test method
      	Factor analysis method			Pearson Analysis method
      	KMO MSA	Require	Remark	rcount	rtablel	tcount	ttable	Require	Remark
      Yi	0.583	KMO > 0.50	valid	0.883	0.308	10.385	2.022	thitng>ttabel	Valid
      Yi1	0.594		Valid	0.910	0.308	11.124	2.022		Valid
      Yi2	0.837		Valid	0.869	0.308	6.516	2.022		Valid
      Yi3	0.165		invalid	0.243	0.308	1.681	2.022		invalid

      Table 7. The Instrument reliability test

      Variable	Cronbach Alpha (r) count	r table	require	Remark	t-test
      		tcount			ttable	require	Remark
      Yi	0.649	0.308	rcount>rta ble	Reliable	5.327	2.022	thitung>tta bel	Reliable
      Yi1	0.703	0.308		Reliable	6.173	2.022		Reliable
      Yi2	0.877	0.308		Reliable	11.398	2.022		Reliable

      Table 8. Data measurement of road damage and serviceability functions

      Segment-i	Xi	CDV	PCIs = 100-CDV	Respondent (N-people)	Component of the road service
      					Yi	Yi1	Yi2
      1	X1	61,55	38.45	95	2.94	3.66	2.83
      2	X2	70,82	29.18	102	2.38	3.09	2.47
      3	X3	80,17	19.83	110	2.02	2.45	1.87
      4	X4	90,34	9.66	101	1.72	1.73	1.69

      Table 9. Mean value of Yi, Yi1,and Yi2 in accordance with CDV

      CDV-61.55% Segmen-1 ( i=1)			CDV-70.82 % Segmen-2 ( i=2)			CDV-80.17 % Segmen-3 ( i=3)			CDV-90.34 % Segmen-4 (i=4)
      Mean		Katego ry	Mean		Kategory	Mean		Katego ry	Mean		Kategory
      Y1	2.94	quite satisfying	Y2	2.38	unsatisfying	Y3	2.02	unsatisfying	Y4	1.72	Very unsatisfaying
      Y11	3.66	A little vibration	Y21	3.09	Real vibration	Y31	2.45	severe vibration	Y41	1.73	vibration very severe
      Y12	2.83	quite safe and comfort	Y22	2.47	unsafe and uncomfort	Y32	1.87	unsafe and uncomfort	Y42	1.60	Very extremely unsafe and uncomfor

      Yi= serviceability for segmen-i; Yi1= vibration intencity for segmen-i; Yi2= safe and comfort for segmen-i.

      The research data

      The research data presented in Table 8.

      Additionally, from the data in Table 8, level of service are labeled on each road segment in accordance with the level of damage as can be seen in Table 9.

      Relations between CDV and composite services (Yci)

      The relations between the CDV and the composite services (Yci) can be constructed based on the average value of the service. The average value of the composite service is:

      Yci = Ln 12.139 0.022CDVi

      or

      Yci = 12.139. e0.022 CDVi (11)

      where: R2adj = function determination after adjustment = 0.996, Yci = the composite service on the i-th segment, CDVi = the corrected deduct value on the i-th segment.

      CDV value that failed of service

      The final part of this study is to determine the

      Yci = (Yi +Yi 1+Yi 2)

      3

      (10)

      CDV value which was failed of service. Based on the failed of service parameters, e.i: 1) Very unsatisfying,

      Table 10. The composite service (Yci)

      2) Vibration very severe, and 3) Very extemely unsafe and uncomfortable and using Eq.11, the CDV

      Segmen-i	CDV	Yi	Yi1	Yi2	Yci	value which was failed of service is determined on
      						the interval scale of service 1Yci1.79, as can be
      1	61.55	2.94	3.66	2.83	3.14	seen in Table 11. At the Table 11, it can be seen that
      2	70.82	2.38	3.09	2.47	2.65	serviceability is failed when Yci =1.79, and at the
      3	80.17	2.02	2.45	1.87	2.12	same time the CDV is at 87.01. Based on data from

      4 90.34 1.72 1.73 1.60 1.69

      By using Eq.10, the composite service (Yci) can be determined and the results are as in Tab.10. Additionally, using the SPSS 17th ver., the best relation between the composite service and the CDV was found in exponential function form as following:

      Table11, the relations between CDV and Yci can be seen in Fig.6.

      .

3. Conclusions and Recommendations

   As the purpose of this research intended is determining a threshold of damage road that it was failed of service,then:

   1. The parameters of failed of servive are:

      1. Very unsatisfying

      2. Vibration very severe

      3. Very extemely unsafe and uncomfortable

   2. The limit of the dameged road that was failed of service is CDV > 87.01 or PCI < 12.99.

      With the failed threshold on the damaged road serviceability, then the badly damaged road are not

      always further categorized on the heavily damaged but may be held to account under the LOIR No. 18/1999, using the clause"building failure". However, to ensure a failed of service road can be categorized as part of building failure, it is necessary to test the ability of road structure whether its a failure or not. So that, the damaged road can be categorized as part building failure when it is fail service and structure.

4. Future Work

To determine failed structurally on damaged road is in progress at authors research.

Table 11. CDVvalue that failed of service

Figure 6. CDV vs Yci

References

1. Al-Mansour A. I., Shinha K. C., and Kuszeck T., 1994. Effect of routine manintenance on flexible pavement condition. Journal of Transportation Engineering, Vol. 120 (3), 65-73.

2. AASHTO (American Association of State Highway and Transportation Officials), 1993. Standar Specifications for Transportation materials and Methods of Sampling and Testing Part I: Specification, 19th edition, Washington, D. C

3. Cantisani, D., and Loprencipe, G., 2010. Road roughness and whole body vibration: Evaluation tools and comfort limits. Journal of Transportation Engineering, Vol. 136 (9), 818-826.

4. Castro, P. M. S. T., and Fernandes, A. A., 2004. Methodologies for failure analysis: a critical survey. Journal, Material and Design 25 (2004) 117-123.

5. Corriere F., and Vincenzo D. D., 2012. The reil quality index as an indicator of the Global Comfort in optimizing safety, quality, and

   efficiency in railways rails. www.sciencedirect.com

   Elsevier, Journal of Procedia-Social and Behavioral Sciences, 53 (2012), 1090-1099.

6. Dong Q., and Huang B., 2012. Evaluation of effectiveness and cost-effectiness of asphalt pavement rehabilitation utilizing LTPP data. Journal of Transportation Engineering, Desember, 2011, doi

   :10.1061/(ASCE) TE.1943-5436. 0000378

7. Els P. S., 2005. The applicability of ride comfort standards to off-road vehicles. www.sciencedirect.com, Elsevier, Journal of Terramechanics, 42 (2005), 47-64.

8. Falou W. E., Duchene J., Grabisch M., Hewson D., Langeron Y., Lino F., 2003. Evaluation of driver discomfort during long-duration car driving. www.sciencedirect.com. Elsevier, Journal of Applied Ergonomics, 34 (2003), 249-255.

9. Fwa T. F., and Farhan J., 2012. Optimal multi-asset maintenance budget allocation in highway asset management. Journal of Transportation Engineering, July, 2011, doi : 10.1061

   /(ASCE)TE.p.1943-5436. 0000323.

10. Government regulation of the Republic of Indonesia No. 29/2000 on the Implementation of Construction Services

11. Hong F., and Prozzi J. A., 2010. Roughness model accounting for heterogeneity based On in-service pavement performance data. Journal of Transportation Engineering, Vol. 136 (3), @ASCE, p. 205-213.

12. Hudson, R., Hass, R., and Uddin, W., 1987. Infrastructur management: Integrating design, construction, maintenance, rehabilitation and renovation, Mc Graw-Hill, New York.

13. Jimenez L. E. A., and Mrawira, D., 2009. Road performance modelling and management system from two condition data points: Case study of Costa Rica. Journal of Transportation Engineering, Vol. 135 (12), 999-1007

14. Khurshid, M. B., Irfan, M., and Labi, S., 2011. Optimal performance threshold determination for highway asset intervention: Analitycal framework and application. Journal of Transportation Engineering, Vol. 137 (2), 128-139

15. Mariani M. C., Bianchini A., Bandini P., 2012. Normalized truncated levy walk applied to flexible pavement performance. Sciverse Sciencedirect, Elsevier, Transportation Research Part C, 24 (2012), 1-8.

16. Noon R. K., 2009. Scientific method: Aplication in failure investigation and forensic science. CRC Press, Taylor & Francis Group, Boca Raton, London, New York, International Standar Book Number: 13:978-1-4200-9280-6, p. 180+xix

17. Park K., Thomas, N., E., and Lee, K. W., 2007. Applicability of the international roughness index as a predictor of asphalt pavement condition. Journal of Transportation Engineering, Vol. 133 (12),706- 709

18. Public Communication Center of Public Work Ministry, 2013. The roads condition and non-toll national road budget management in years 2009- 2013.

19. Pt-T-01-200-B. The guidance of flexible pavement design according to Public Work Ministry.

20. Regulation of public works minister of the Indonesia Republic No. 14/PRT/M/2010, about Minimum Service Standards, Division of Public Works and Spatial Planning.

21. Regulation of interior minister of the Indonesia Republic No. 6/2007 on Technical Guidelines for Preparation and Determination of Minimum Service Standards.

22. Simamora M., and Hatmoko J. U. D., 2013. Verifying Damaged Road Failure Criterion: A Study Using 30 Experts. Proceeding The 6th Civil Engineering Conference in The Asia Region: Embracing the Future through Sustainability in Indonesia. ISBN 978-602-8605-08-3

23. Sukirman S., 1992. Flexible pavement on highway. Publihsed Nova, Bandung.

24. Suswandi A., Sartono W., and Hary C. H., 2008. Evaluation of road distress use Pavement Condition Index (PCI) method to support decicion making. A case study: Yogjakarta South ring road. Forum Teknik Sipil No. XVIII/3-September 2008, pp. 934-946

25. The laws of the Republic of Indonesia No. 18 of 1999 on Construction Services.

26. Uys P. E., and Els P. S., 2007. Suspension settings for optimal ride comfort of off-road vehicles travelling on roads with different roughness and speeds. www.sciencedirect.com ,Elsevier, Journal of Applied Terramechanics, 44 (2007), 163-175.

27. Yang Y., Ren W., Chen L., Jiang M., Yang Y., 2009. Study on ride comfort of tractor with tandem suspension based on multi-body system dynamics.

    www.sciencedirect.com. Elsevier, Journal of Applied Mathematical Modeling, 33 (2009), 11-33.

28. Zeithaml V. A., Parasuraman A., and Bary L. L., 1990. Delivery quality service balancing costumer perception and expectation. New York, The Free Press.

29. Zhang, H., Keoleian, G. A., and Lepech, M. D., 2012. Net-work level pavement asset management system integrated with life cycle analysis and life cycle optimization. Journal of Infrastructure

    Systems, doi. 10.1061/(ASCE)/S 1943-555x 0000093

30. Zhang Q., Mills-Beale J., and You Z., 2010. Hand book of Transportation Engineering: Pavement

Engineering I, Flexible Pavement. 2nd edition, p. 11.1-11.30.