Pavement sustainability index for highway infrastructures: A case study of Maryland

Abstract

Pavement deterioration creates conditions that undermine their performances, which gives rise to the need for maintenance and rehabilitation. This paper develops a mathematical multi-linear regression analysis (MLRA) model to determine a pavement sustainability index (PSTI) as dependent variable for flexible pavements in Maryland. Four categories of pavement performance evaluation indicators are subdivided into seven pavement condition indices and analyzed as independent variables for each section of pavement. Data are collected from five different roadways using field evaluations and existing database. Results indicate that coefficient of determination (R²) is correlated and significant, R² = 0.959. Of the seven independent variables, present serviceability index (PSI) is the most significant with a coefficient value of 0.032, present serviceability rating (PSR) coefficient value= 0.028, and international roughness index (IRI) coefficient value= ‒0.001. Increasing each unit value of coefficients for PSI and PSR would increase the value of PSTI; thereby providing a more sustainable pavement infrastructure; which explains the significance of the model and why IRI will most likely impact environmental, economic and social values.

Keywords

pavement indices / sustainability / pavement performance / flexible pavements

Introduction

For the past couple of decades, there have been global concerns on climate changes, energy usage, environmental impacts, and financial limitation as they affect highway transportation infrastructures, both in developed and developing countries. Engineers and policy makers have been looking into different ways to planning, designing, constructing, operating, and maintaining pavements in order to sustain these highway infrastructures. Pavement infrastructures are known to be very vital and important in safely transporting people and goods from one point to another, which makes it one of the fundamental components of any transportation system both in the United States of America and the world in general.

Although many researchers have undertaken various aspects of pavement research, such as pavement evaluation, maintenance, pavement management systems, to name a few, however, few research has been carried out on pavement sustainability index (PSTI). The purpose of this paper is to illustrate some major transportation sustainability indicators (such as, environmental, social and economic impacts) and correlate them to pavement performance/evaluation indices (such as serviceability, structural capacity (SC), skid resistance and pavement surface distress) [¹]. The paper addresses the problem of sustainability in the context of pavement performance (pavement conditions) in a statistical multi-linear regression analysis (MLRA) framework. The facility is divided into shorter manageable sections and each section is analyzed and correlated in order to obtain a correlation index. The motivation behind this paper therefore is to develop a comprehensive model that determines the relationship between pavement conditions (indices) and sustainability indices-known as the universally accepted “triple bottom line” or (“three pillars”) of sustainability [¹].

Prior to maintaining any highway infrastructure such as pavements and bridges, the preferred mode of maintenance is selected based on the need and availability of funds for the project. The next deciding factor is whether to maintain or fix the problem of the highway infrastructure or do-nothing. The planning and maintenance strategies depend largely on the type of highway infrastructure to be maintained or rehabilitated. Another important factor to be considered is the environmental and social impacts such as: types of materials used, environmental pollution, traffic congestions, delay, etc. The problem statement of this paper is to develop a statistical multiple-linear regression model that would establish a relationship between PSTI and pavement performance evaluation indices (PPEI) for highway infrastructure in Maryland by modifying selected relevant existing statistical models. These modified models will reflect the pavement condition properties and variations using MLRA.

Pavement serviceability

Pavement serviceability is defined as the ability of the pavement to carry the accumulated traffic axle loads it is designed to serve under the existing pavement condition. The present serviceability index (PSI) is indirectly proportional to traffic loading; i.e., at the initial construction stage, the PSI is in very good condition of 4.5; and as traffic demand increases, the PSI level decreases to acceptable range of 2.0. At this level, a maintenance or rehabilitation is required to increase the pavement PSI level close to the level of its initial construction [²]. There are two major methods for determining pavement serviceability; the PSI and the international roughness index (IRI).

PSI is a combination of mathematical formulation that is obtained from measuring pavement sections used to predict the present serviceability rating (PSR). The American Association of State Highway and Transportation Officials (AASHTO) Road Test originally developed the mathematical equations for PSI for both flexible and rigid pavements, respectively [³].

(1)PSI=\mathrm{5.03}-\mathrm{1.9}\log (\mathrm{1}+SV)-\mathrm{0.01}\sqrt{C+P}-\mathrm{3.38}{RD}^{\mathrm{2}},

(2)PSI=\mathrm{5.41}-\mathrm{1.8}\log (\mathrm{1}+SV)-\mathrm{0.09}\sqrt{C+P},

where SV = slope variance (rad²), P = bitumen patching in ft²/1000ft², RD = average rut depth in inches, C + P = the relative extent of cracking and patching in the wheel-path (ft²/1000ft²).

A literature review on PSI indicates a different model developed by [⁴] to estimate the PSI from IRI for asphalt pavements. This model is used to estimate the PSI for the paper case studies.

(3)PSI=\mathrm{5}+{\mathrm{0.2397}x}^{\mathrm{4}}+{\mathrm{1.771}x}^{\mathrm{3}}-{\mathrm{1.4045}x}^{\mathrm{2}}-\mathrm{1.5803}x,

where x = log(1+SV), SV = 2.2704 IRI².

Correlations between PSR and IRI

There have been several correlations between PSR and IRI and expressed mathematically in Eqs. (4) and (5) respectively:

(4)PSR={\mathrm{5}\mathrm{e}}^{-\mathrm{0.18}IRI},

(5)PSR={\mathrm{5}\mathrm{e}}^{-\mathrm{0.26}IRI}\mathrm{.}

The smoother the riding pavement surface, the higher the value of PSR. The PSR is a subjective ride based on individual observation of pavement conditions; while the IRI is based on the average rectified slope (ARS)—a filtered ratio of a standard vehicle’s accumulated suspension motion (measured in millimeters, inches, etc.) divided by the distance travelled by the vehicle [⁵]. For the purpose of this research paper, existing IRI data will be extracted from Maryland State Highway Administration (MDSHA) database to determine the PSR. Equation (6) gives the relationship between IRI and ARS.

(6)IRI=ARS\times \mathrm{1000.}

Skid resistance (SR)

Skid resistance (surface friction) is a vital component of pavement performance evaluation. It is important to consider SR during pavement design and rehabilitation process in order to provide safe pavement surfaces to road users. According to Highway Research Board, National Academy of Sciences, Washington D.C., SR is defined as the force that develops when the tires are prevented from rotating on pavement surfaces. SR can cause skid related incidents or accidents if inadequately designed. SR is dependent on four factors; pavement surface texture, the tire tread, water in the interface between the tires and the pavement surface, and amount of slippage between the tires and surface. The SR decreases when the pavement is wet, which increases the rate of accidents (safety concerns). Also, aggressive tire treads and dry pavement surface tend to increase SR. Safety issue may be due to the wetness of pavement surfaces because there is friction reduction between the axle tires and the road surface. The amount of slippage (slip speed) between the tires and pavement surface is defined as the relative speed between the contact of a tire and the pavement surface. The slip speed is zero before the application of the breaks, and slip speed equals the instantaneous speed when the vehicle wheels are locked while it is in motion.

There are different field techniques of measuring SR; the four most common are: the locked wheel tester, fixed slip speed, variable slip speed, and the side force in yaw mode. Details of the various techniques are not within the scope of this research and therefore will not be discussed. Details of the measuring devices can be obtained from American Society Testing and Materials (ASTM E 274) [⁶,⁷,¹²]. In order to determine and analyze SR for this paper, existing skid number (SN) data are obtained from MDSHA database. The measurement of SR can be quantified using either skid number or friction factor; expressed mathematically:

(7)\mathrm{friction}\hspace{0.17em}\mathrm{factor}\hspace{0.17em}f=F/L,

(8)\mathrm{skid}\hspace{0.17em}\mathrm{number}\hspace{0.17em}SN=\mathrm{100}f,

where F = frictional resistance to motion in plane of interface, L = load perpendicular to interface.

Structural capacity (structural deflection)

Structural evaluation is used to determine the adequacy of any pavement to support traffic without developing structural distress. Evaluation of pavement adequacy falls into two categories—the deflection category and the effective thickness category. This research focuses on the first category due to traffic loading. Literature reviews indicate that pavement condition is one of the most vital issues for road users [⁸]. The quality of a road is primarily judged by its roughness through subjective ratings by the road users (AASHO road test). There may be adequate relationship between roughness and structural adequacy depending on the quality of the pavement and the axle loads placed on it. For instance, a pavement with low SC can have a good ride quality if it does not exceed its design capacity; but a poorly designed pavement can deteriorate faster, which may result in pavement distress thereby resulting in pavement roughness [⁹].

Structural deflection due to axle loadings is one of the four components for evaluating and measuring pavement performance and is being considered in this research. It also plays an important role in evaluating pavement structures because the size and shape of pavement deflection is a function of factors that cause pavements to deteriorate. Such factors include, but not limited to the following: axle loads due to traffic volumes, temperature, and moisture due to expansion and contraction of the pavements; and pavement structural section. There are various methods and techniques for measuring pavement structural deflection; such as backcalculation, impact load deflection (falling weight deflectometer (FWD)), the static deflection, and steady state deflection.

Effects of pavement conditions on road users and agency

Pavement conditions have significant effects on road users and agencies. These effects are in different forms; such as increase in travel time, delay/congestion, environmental impacts, social impacts, health impacts, economic impacts, etc. In this research paper, the focus is based on the qualitative analysis of triple dimension sustainability; i.e., the environmental, social and economic impacts. Under each dimension, the research paper considers emissions, water, accidents, health, user costs, and agency costs. Some of these impacts are quantifiable as well as qualitative. The quantifiable impacts are discussed briefly to illustrate how pavement conditions affect them; while the qualitative impacts are analyzed using existing models to illustrate their impacts. Islan and Buttlar [¹⁰] performed life-cycle cost analysis to compare user costs vs. agency costs related to pavement roughness and found that agency costs were small compared to roughness-related user costs over the life of the pavement. They concluded that the agency investment in increased maintenance and rehabilitation activities would have a 50-fold return in the form of reduced user costs. The authors buttressed their argument by using hypothesized benefits in pavement system sustainability through reduced user fuel costs, and reduced tire wear and increased remaining life of the pavement.

Roughness is quantified using IRI and PSR; with IRI being the most important index to determine performance. The following subsections present the effects of pavement conditions on the triple sustainability dimensions: pavement condition vs. fuel consumption; pavement condition vs. speed; pavement condition vs. gas emission; pavement condition vs. vehicle operating costs (VOC).

Importance and selection of sustainability Indicators

Indicator can be defined as something that makes one to know the level where one is at the moment, how far, and the next step to take to get to the next level where you would like to be in order to archive the intended goals. In the technical perspective, an indicator alerts an agency or organization of a problem before it is too late to fix the problem and it also helps to recognize the solution to fix these problems. A sustainable indicator directs to the areas of weakness between environment, economic, and social values and they are inter-woven (Table 1).

Methodology

There are a number of factors affecting pavement performance, including the demand and traffic variability that cause pavements to deteriorate. The research paper is to build a PSTI for highway infrastructures maintenance: case study of flexible pavements in Montgomery County and Frederick County, Maryland. The model is of two components, quantitatively and qualitatively. In order to achieve the objectives of the research paper, it was first necessary to select some numbers of roads in Maryland and later narrowed the selection to five roads based on availability of existing structural deflection data (i.e., availability of FWD data) which was the main determinant factor of the model. The selection of these roadways is based on the five categories of pavement performance evaluation. The approach is to first of all determine the sections with pavement conditions (poor, fair, good, and excellent) using a five-man panel to rate each pavement section on a scale of 0–1 (0, being worst condition and 1 indicates excellent condition) to determine the PSTI.

A multiple regression analysis (MLRA) model was first performed for a numerical example to test the proposed model before it was applied to the case study areas under consideration. MLRA is proposed for the research model because of the multiple variables involved in the research paper and these variables include PSTI as the dependent variable, i.e., the conditions of these pavements are subjectively evaluated and rated by five panel evaluators. The PPEI (IRI, PSI, PSR, structural adequacy index (SAI), skid resistance index (SRI), cracking index (CKI), and rutting index (RTI)) are considered as the independent variables using existing data from MDSHA database and data obtained from field evaluations. Figure 1 is a simple flow chart that illustrates the research paper model.

Five different roadways (Table 2) are randomly selected and divided into sections of 500 feet apart for evaluation. A five-member survey panel evaluates and rates each section of the pavement individually and the mean is obtained to determine the PSTI and PSR respectively using proposed guidelines in Table 3. The five panel members with technical background in pavement maintenance are chosen and instructed to consider the pavement condition, such as defeats, comfort of ride ability, etc. and rate on forms provided if acceptable or not. The field data collected for PSTI are analyzed to determine the significance of the relationships between pavement performance evaluation (PPE) indices (i.e., IRI, PSI, and PSR, skid resistance index (SRI), rutting index (RTI) and cracking index (CKI). In order to implement this task, a MLRA model is proposed to analyze, quantify, weigh, and correlate. The methodology for the proposed model consists of the following steps:

(1) determination and collection of relevant data as they pertain to the research objectives and availability of data from MDSHA database;

(2) determination of the pavement performance indices using existing models from AASHTO 1993 and Asphalt Institute Manual Series No. 17, [⁵] respectively to obtain some parameters such as representative rebound deflection, design rebound deflection, structural deflection, etc;

(3) development, Analysis and Validation of a statistical regression model for the PSTI on selected sections of the roadways under consideration using existing data from MDSHA dataset, AASHTO 1993, and AI MS-17, 1983 respectively;

(4) development and analysis of mathematical and graphical relationships between PSTI and PPE indices;

Utilization of Analysis of Variance to analyse the regression results obtained.

Panel evaluation of pavement condition

A five-member panel is used to survey each five selected roadway under consideration to collect data for the PSTI using modified asphalt rating form (Asphalt Institute Manual Series No. 17, [⁵]. Each roadway is surveyed and evaluated based on pavement conditions. Each evaluator is instructed and thoroughly briefed on how the roads are to be rated and the purpose of ratings before the survey was started. The following rules are considered during evaluation:

• the evaluators should only use the current pavement condition to rate each section;

• the driving speed (drive at a speed of 5 mph less than the posted speed limit). For example, if the posted speed limit is 30 mph, the evaluator will drive at 25 mph. This is to allow the evaluators properly capture and rate the conditions of pavement;

• type of traffic mix (trucks, cars, pickups, vans, buses, etc.);

• evaluators should ignore isolated conditions such as bus pads, bridges, rail road crossings, traffic calm bumps, etc;

• the condition of pavements (i.e., type of cracking, rutting, patching, etc.);

• riding condition of pavements (noise and smoothness, skid resistance from pavements);

• weather (i.e., is pavement wet or dry);

• consideration of the roadway configuration;

• record time and date of survey;

• record type of vehicle used for survey;

• the evaluator should be able to determine if each section of pavement rated is acceptable or not.

The five randomly selected roadways based on available deflection data are each divided into equal sections of 500FT long and each section is evaluated and rated. The PSR of each section is rated subjectively on a scale of 0–5 (0 implies very poor; while 5 implies very good) using the standard rating form; while PSI for each section is calculated using Eq. (9b). The guideline set for the PSTI of each section is on a scale of 0–1 (0 refers to very poor and 1= very good).

After the completion of the evaluation, all ratings are tabulated for each roadway under consideration and the mean calculated. For this research, the calculated subjective mean for the PSR is compared with the calculated objective PSR using Eq. (9a) and the PSI is also obtained using Eq. (9b) below. Since the ratings for both PSRs differ by less than approximately 0.3, the panel rating is considered to be satisfactory (Asphalt Institute Manual Series No. 17, [5]. The data are recorded on each PSR and PSTI forms respectively by the raters based on above evaluation rules and the mean recorded.

(9a){{\displaystyle PSR}}_{\mathrm{obj}}={\mathrm{5}\mathrm{e}}^{-\mathrm{0.0041}IRI},

(9b)PSI={\mathrm{5}\mathrm{e}}^{-\mathrm{0.0038}IRI},

where e= the exponential,

(10a){{\displaystyle PSR}}_{\mathrm{sub}}=X={\displaystyle \frac{{{\displaystyle \Sigma }}_X}{n}},

(10b)PSTI={\displaystyle \frac{{{\displaystyle \Sigma }}_r}{n}},

where X, r = individual rating values assigned by each panel evaluator, n = total number of rating panel evaluators.

Data analysis

This section introduces the procedures for determining and collecting the data. Majority of the research data are obtained from MDSHA office of maintenance (OMT) existing database; while PSTI data are obtained from field survey and evaluation of the pavement sections under consideration. The existing database is imported in form of excel spreadsheet from MDSHA OMT database based on availability of structural capacity deflection data as a determinant factor for all five roadways under consideration for year 2013. The database contains some general relevant information for structural deflection of the pavement sections, Data considered for calculating the SAI in this research paper are highlighted in bold:

• reference Station locations

• beginning and ending mile points

• route name and number

• deflection data

• year of data collection

• number of load drops

• average Annual Daily Traffic (AADT)

• equivalent single-axle load (ESAL)

• county codes for the different roads

• number of lanes

Evaluation of pavement condition sections for the study areas consists of the four categories of pavement performance. The research work analyses the serviceability (roughness, PSR or pavement serviceability index), structural capacity (SAI), skid resistance (frictional number) and pavement distress (pavement condition index—cracking index & rutting index).

The method for measuring the pavement condition for the purpose of this research is carried out by subjective process by driving through the selected roadways in order to determine the PSTI on a scale of 0–1 (1 being very good, while 0 is rated as very poor) based on the current state of the pavement distress conditions (very good, good, fair, poor and very poor). Majority of data for the research analysis were obtained from MDSHA (OMT) existing database and field data are analyzed using existing models from AASHTO and AI, MS-17, 1983 respectively. The flow chart in figure 1 illustrates how the model is analyzed to determine the relationships between PSTI and pavement performance indices; and can be used by any agency.

Model formulation

The methodology approach in developing the research paper framework consists of using existing models from AASHTO, 1993 and AI, MS-17, 1983 respectively to determine the independent variables. The model analysis is performed using existing data from MDSHA database and field investigation data to develop the framework. The proposed framework is a linear regression analysis which can be used to incorporate the relationships between the variables in order to predict the PSTI for the proposed sustainability index. The research involves two variables—PPEI as the independent variables and PSTI as the dependent variables. The PSTI is the dependent variable (a predictive variable) and it is a function of PPEI (Eq. (11)).

Regression modeling

Once the data have been collected and analyzed, the MLRA is used (Microsoft Excel software analytical tool) to model the relationships between the major selected variables, based on the basic (AASHTO, 2001) MLRA equation [¹¹]:

(11)PSTI=f(IRI,PSR,PSI,RTI,CKI,SRI,SAI).

Assuming: PSTI = Y (dependent variable), PPE= x_1,x₂ , x₃ ,......... x_n (independent variables), expressed as a function,

(12)Y=f(X),

therefore,

(13)Y=f({{\displaystyle x}}_{\mathrm{1}},{{\displaystyle x}}_{\mathrm{2}},{{\displaystyle x}}_{\mathrm{3}},\mathrm{.........}{{\displaystyle x}}_n),

(14)Y={{\displaystyle \beta }}_{\mathrm{0}}+{{\displaystyle \beta }}_{\mathrm{1}}\times {{\displaystyle X}}_{\mathrm{1}}+{{\displaystyle \beta }}_{\mathrm{2}}\times {{\displaystyle X}}_{\mathrm{2}}+{{\displaystyle \beta }}_{\mathrm{3}}\times {{\displaystyle X}}_{\mathrm{3}}\mathrm{......}+{{\displaystyle \beta }}_{p-\mathrm{1}}\times {{\displaystyle X}}_{p-\mathrm{1}}+e,

where, Y= dependent variable, X1,2,.. p1= independent variables, β0= regression constant (linear regression intercept), β1,2,.. p1= variable constants (slope for linear regression); and e= random error.

This research work considers three objectives while developing the MSLRA model: estimate the unknown parameters (β), hypothesized the proposed linear model, and check if the model is good for predicting the dependent variable Y.

Model accuracy

This section discusses the proposed model accuracy to check the accuracy of the model. Once the developed model is checked for fitness, it is important to determine if the designed model is good for predicting the dependent variable. In order to perform this important task, a hypothesis test is carried out on all the regressions (β–parameters) constants (excluding β₀ ), using the following Ep. (15); and F-test Ep. (16) based on the null ( H₀ ) and alternative hypothesis (H_a):

({{\displaystyle H}}_{\mathrm{0}}):{{\displaystyle \beta }}_{\mathrm{1}}={{\displaystyle \beta }}_{\mathrm{2}}=\mathrm{...}{{\displaystyle \beta }}_k=\mathrm{0},

H_a: one of the parameters should defer from zero; therefore, F = model square /error mean square,

F={\displaystyle \frac{R^{\mathrm{2}}}{\mathrm{1}-R^{\mathrm{2}}}}{\displaystyle \frac{n-(k+\mathrm{1})}{k}}.

Rejection region if F>F_a, where N = number of observations, K = number of parameters in the model (exceptβ₀), R² = multiple coefficient of determination, and A = significance level.

Model framework

Figure 1 illustrates the format of the proposed model for the research. The PSTI is the dependent variable which is a function of the four pavement performance evaluation categories of PS, SC, SR, and PCI. Each category is further subdivided to obtain the indices (IRI, PSR, PSI, RTI, CKI, SRI, & SAI) in order to determine their relationships using multiple statistical linear regression analysis (MSLRA) and analysis of variance (ANOVA). Each independent variable, such as PSR, PSI, SAI are determined using existing models in the literature reviews; whilst dependent variables, such as IRI, RTI, CKR, and SRI are existing raw data obtained from MDSHA database.

In order to adequately fit the linear regression, a linear correlation has to exist between the dependent and independent variables and these variables are assumed to have influence on sustainability and can be used to estimate PSTI. Each of the independent variable was tested against the PSTI in order to determine how correlated they are. The results obtained from the regression analysis showed that correlation between the variables vary depending on the condition of the road being analyzed.

The seven independent variables were determined using existing models from literature reviews to generate a regression model to estimate PSTI. Microsoft Excel Data Analysis Tool for Multiple Regression was used to perform the statistical analysis. The data used for this research contained varying observations based on the deflection data as the determinant factor for selecting roads. These observations were performed on 5 roadways in Montgomery and Frederick Counties in Maryland. The proposed regression model for this research is shown in Eq. (11) and further modified as shown in Eq. (17).

(17)PSTI={{\displaystyle \beta }}_{\mathrm{0}}{{\displaystyle +\beta }}_{IRI}\times IRI+{{\displaystyle \beta }}_{PRS}\times PSR+{{\displaystyle \beta }}_{PSI}\times PSI+{{\displaystyle \beta }}_{RTI}\times PTI+{{\displaystyle \beta }}_{SRI}\times SRI+{{\displaystyle \beta }}_{CKI}\times CKI+{{\displaystyle \beta }}_{SAI}\times SAI+e.

Results

The PSTI is the important information in this research that is weighted on a scale of 0-1 in the statistical analysis. It is the response variable in the regression model and the data are collected from field evaluation of the roads using five-man experienced technicians as evaluators and the average weighted scores obtained are used to determine the pavement sustainability indices.

Among the information gathered during evaluation were the pavement conditions; such as how smooth the rides are, cracks, rutting, patching, what the weather was like, dry or wet pavement, roadway configuration, volume of traffic, type of vehicle used for evaluation, speed while collecting data, noise from pavement surface during evaluation.

The regression model indicates a strong relationship between dependent (PSTI) and the independent variables. A linear regression was performed on the overall case study data which yielded a strong coefficient of determination, R² of higher significant value of 96% (Fig. 2). To further explain this value, 97% of the sum of squares in PSTI can be associated with the variation in the independent variables. Equation (18) below and the regression excel output tables are given below respectively (Tables 4–6).

(18)PSTI=\mathrm{0.6952}-\mathrm{0.0013}IRI+\mathrm{0.027725}PSR+\mathrm{0.032036}PSI-\mathrm{0.01165}RTI+\mathrm{0.00013}CKI+\mathrm{0.0005}SRI-\mathrm{0.00344}SAI+e.

Discussion and conclusions

The coefficients for the five case studies in Tables 4–6 indicate that there is a strong correlation relationship between dependent PSTI and the seven independent variables with R² of 96%.

This means that 96% of the sum of squares in dependent variables (PSTI) can be associated with the variations in the seven independent variables. It can be said that these coefficients are elastic, in other words they can be moved around depending on the signs and or values associated with the variables.

In the t-stat, IRI negatively impact PSTI with a value of –18.24 and a coefficient value of 0.0013; whilst PSI and PSR have higher positive coefficient impacts on PSTI respectively compared to other variables.

The P-values and the intercepts for PSI and PSR are significant at 95% level respectively. PSI has a stronger coefficient of 0.0320 than the rest variables, followed by PSR with coefficient of 0.0277. This means increasing each unit value of coefficients for PSI and PSR would increase the value of PSTI. Thus concentrating resources in raising the PSI and PSR values will drastically increase the PSTI, subsequently providing a more sustainable pavement for the agency and the public in general. This makes the model to be a useful and significant one in determining the relationship between the independent variables and the sustainability index (dependent variables).

The analysis presented here is based on available data from MDSHA database and research field evaluations. The developed model will help support pavement managers/engineers to make rational decisions to enhance pavement performance and sustain the infrastructure.

Subsequently, the model provides framework for transportation agencies which will enable them meet the basic needs of pavement maintenance and improve the quality of life of pavements; while the natural resources they depend on are maintained and enhanced for their benefits, and the future generations without compromising safety, health and system efficiency.

Future works

Due to lack of data availability and time constraint, the triple-bottom-line sustainability dimension is a possible area that should be considered and expanded quantitatively in the future.

References

[1] Obazee-Igbinedion, S O, Jha M K, Owolabi O. Building sustainability index for highway Infrastructures: A case study of flexible pavements. Journal of Biourbanism, 2013(2013): 1–2

[2] Hveem F N, Carmany R M. The Factors Underlying the Rational Design of Pavements Proceedings. Washington, D.C.: Highway Research Board, 1948

[3] Jayawickrama P W, Prasanna R, Senadheera S P. Survey of state practices to control skid resistance on hot-mix asphalt concrete pavements. Transportation Research Record, 1996, 1536, 52–58

[4] Gillespie T D. Everything you always wanted to know about IRI, but were afraid to ask. 1992

[5] Asphalt Institute. Asphalt overlays for highway and street rehabilitation. Manual Series No. 17. 1983

[6] American Association of State Highway Officials. The AASHO road test, pavement research, Report 5. 1960

[7] Graham, J. Internal Traffic Control Plan. Kansas: Midwest Research Institute, 2005

[8] Carvalho R, Stubstad R, Briggs R, Selezneva O, Mustafa E, Raamachandran A. Simplified techniques for evaluation and interpretation of pavement deflections for network-level analysis. Report No. FHWA-HRT-12-023. 2013

[9] Carey, W N, Irick, P E. The pavement serviceability-performance concept, HRB Bulletin 250. 1960

[10] Islam S, Buttlar W G.Effect of pavement roughness on user costs. Journal of the Transportation Research Board, 2012, 2285: 47–55

[11] Obazee-Igbinedion S O. Building sustainability index for highway infrastructures: A case study of flexible pavements. Dissertation for the Doctoral Degree. Baltimore: Morgan State University, 2015

[12] American Society for Testing and Materials. Standard test method for skid resistance of paved surfaces using a full-scale tire, ASTME 274. 1995

[13] Papagiannakis, A T, Masad E A. Pavement Design and Materials. New York: John Wiley & Sons, 2007

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany