Development of Roughness Prediction Models for Laos National Road Network

Mohamed Gharieb,Takafumi Nishikawa

doi:10.3390/civileng2010009

Abstract

The International Roughness Index (IRI) has been accepted globally as an essential indicator for assessing pavement condition. The Laos Road Management System (RMS) utilizes a default Highway Development and Management (HDM-4) IRI prediction model. However, developed IRI values have shown the need to calibrate the IRI prediction model. Data records are not fully available for Laos yet, making it difficult to calibrate IRI for the local conditions. This paper aims to develop an IRI prediction model for the National Road Network (NRN) based on the available Laos RMS database. The Multiple Linear Regression (MLR) analysis technique was applied to develop two new IRI prediction models for Double Bituminous Surface Treatment (DBST) and Asphalt Concrete (AC) pavement sections. The final database consisted of 83 sections with 269 observations over a 1850 km length of DBST NRN and 29 sections with 122 observations over a 718 km length of AC NRN. The proposed models predict IRI as a function of pavement age and Cumulative Equivalent Single-Axle Load (CESAL). The model’s parameter analysis confirmed their significance, and R2 values were 0.89 and 0.84 for DBST and AC models, respectively. It can be concluded that the developed models can serve as a useful tool for engineers maintaining paved NRN.

Highlights

IntroductionConnectedness with adjoining countries is of pivotal significance [2]. As the country has no harbor and has only 3.5 km of the railway link at the Thai border, the road network is an essential means of transportation to link Laos with its neighbors and within the country [3]
As a landlocked country [1], the Lao People’s Democratic Republic’s (Lao PDR)connectedness with adjoining countries is of pivotal significance [2]
The Laos road network is split into six classes: (I) National Roads (NRs) link the national capital to other provincial capitals and commercial centers and reach the global borders; (ii) Provincial Roads (PRs) link provincial capitals to district centers, river ports, tourist and important historic sites of the province; (iii) District Roads (DRs) link district commercial centers to rural areas; (iv) Urban Roads (URs), which are internal to cities and towns; (v) Rural Roads (RRs) link rural areas and utilities serving the rural houses; and (vi) Special Roads (SRs) serve definite purposes, such as connecting to sightseeing areas [5,6]

Summary

Introduction

Connectedness with adjoining countries is of pivotal significance [2]. As the country has no harbor and has only 3.5 km of the railway link at the Thai border, the road network is an essential means of transportation to link Laos with its neighbors and within the country [3]. In 2004, as a result of limited budget and growing challenges in pavement maintenance and rehabilitation strategies, the Road Management System (RMS) was developed for NRs based on Highway Development and Management (HDM-4) models with aid from the World Bank [24]. The calibration and application of the Mechanistic–Empirical Pavement Design Guide (MEPDG) or HDM-4 models by highway agencies demand detailed and precise distress data [36] Such data records are not quite available for Laos yet, making it difficult to calibrate the HDM-4 IRI prediction model for local conditions. The domestic roughness deterioration models for various pavement categories have to be developed covering the influence of Laos’ local conditions, which would have direct implementation without any calibration factors based on accessible data in the Laos RMS. The developed models’ primary objective is to evaluate and predict the Laos NRN’s condition to assist the responsible authorities in making consistent and cost-effective decisions related to pavement sections’ maintenance and rehabilitation

Prediction Method

Data Separation

Data Screening

Pavement Age

Findings

Regression Model