DEVELOPMENT OF PAVEMENT MANAGEMENT SYSTEM FOR INDIAN NATIONAL HIGHWAY NETWORK

Abstract

In India, the growth of road traffic in the post independence era has been quite unprecedented, both in terms of goods and passenger traffic. The vehicular traffic has increased manifolds from 3 lakh to 450 lakh, during the period 1951-2001. But unfortunately, the growth in the road network has not been commensurate with the huge growth in vehicular population, which has increased from 4 lakh km to just 33 lakh km, during the same period. The traffic axle loadings have also been much heavier than the specified limit of 10.2 tonnes. Consequently, the existing highway network is being heavily stressed, resulting in accelerated deterioration of road pavements leading to their premature failure. But the maintenance and rehabilitation measures of pavements are based on subjective judgment and past experience of the highway engineers only. Therefore, there is a need of developing a scientific approach towards determining the maintenance and rehabilitation requirements of pavements. In such a situation, development and practice of an efficient Pavement Management System (PMS) would provide objective information and useful analysis to ensure consistent, and cost-effective decisions related to preservation of the existing highway networks. A comprehensive survey of the research literature suggests that at present, many highway authorities in developed countries are using a systematic and objective method to determine pavement condition and programming maintenance in response to observed conditions. In many of the developing countries, PMS is in various phases of working process with diversified approaches as per the respective needs and problems of each country. In India, some efforts have been made towards development of Pavement Management Systems for different categories of roads viz. National Highways, State Highways, Urban Roads and Rural Roads. The 'Ministry of Road Transport & Highways' (MORT&H), Govt, of India, has reviewed the above efforts and recommended the need for development of a comprehensive pavement Management System for National Highway network of India. Modern tools like HDM-4 and GIS have been recommended to be used comprehensively, to make the PMS decision-making process much more quicker and simpler. In the last twenty years, the pavement management systems have improved significantly due to the advances in computer technologies, and a series of PMS tools have been developed. But most of the tools have been developed keeping in view the requirements of a particular highway agency, and the conditions prevailing in a particular country or a geographical region. Hence, these tools lack universal acceptance and implementation. But one of the universal tools available for pavement management analysis is the Highway Development and Management System (HDM-4). The HDM-4 is a new international standard model, which can be used in assisting pavement managers to predict economic and technical outcome of possible investment decisions concerning maintenance management of pavements. The objective of this study is to develop a Pavement Management System for an identified National Highway network. The study is limited to National Highways having bituminous surfaces only. In the present study, only those parameters have been considered which are found suitable to the identified National Highway network. The parameters, related to rigid and unsealed pavements as well as those related to extreme climatic zones have not been considered. In view of its universal recognition and application, HDM-4 system has been used for the development of pavement management system for the identified National Highway network. A methodology has been proposed for the development of pavement management system for a representative National Highway network of India. The highway network selected for the present study comprises of a number of segments of the five National Highways viz., NH-58, NH-72, NH-72A, NH-73 and NH-74, which are located within the administrative boundaries of Dehradun and Hardwar districts of Uttaranchal State, and Saharanpur and Muzzafarnagar districts of the State of Uttar Pradesh in India. These highways are considered to be the representative National Highways of the country. The selected highway network has been divided into 22 homogeneous pavement sections. Division of highway network into sections has been done on the basis of homogeneity with respect to: soil type, terrain, traffic volume, climatic conditions, and pavement condition characteristics. These 22 pavement sections have been grouped into three Maintenance Serviceability Levels, i.e. High (> 10000 AADT), Medium (5000 - 10000 AADT), and Low (<5000 AADT), as per the volume of traffic carried bythem at present. Major part of the data, especially on road inventory, pavement condition, and traffic have been collected by using equipments, such as, Fifth Wheel Bump Integrator, Benkelman Beam, Portable Falling Weight Deflectometer, Portable Skid Resistance Tester, and Straight Edge, etc. Other data, such as, pavement construction and maintenance history data, vehicle fleet data and cost data have been collected through secondary data sources like Public Works Department (PWD) records, and the relevant publications of the Indian Roads Congress (IRC), and the Ministry of Road Transport & Highways (MORT&H), Government of India. The applicable maintenance and rehabilitation (M&R) activities and the corresponding intervention criterion, as per the three maintenance serviceability levels have been defined. All the collected data has been stored in 'National Highway Network', 'National Vehicle Fleet', and 'Maintenance Work Standards' databases, created in HDM-4. The data has also been integrated with GIS package ArcView to create a geographically referenced pavement database. The HDM-4 pavement deterioration models for cracking, ravelling, potholes, and roughness, have been calibrated to suit the local conditions, by making use of the Indian deterioration models, as developed in the Pavement Performance Study conducted in India, under the Research Project sponsored by MORT&H, Government of India, and the data collected from the identified National Highway network. The calibration factors obtained in this study for various deterioration models are: Cracking Initiation (Kcia = 0.43), Cracking Progression (Kcpa = 1.25), Ravelling Initiation (Kvi = 0.37), Ravelling Progression (Kvp = 0.52), Pothole Initiation (KPj = 0.45), Pothole Progression (Kpp = 0.95), and Roughness Progression (Kgp = 0.85). The calibration factors in respect of Rutting, and Skid Resistance models have been assumed as equal to unity, as per the provisions of the HDM-4, since these two deterioration parameters were not considered part of the above mentioned Pavement Performance Study. The calibrated deterioration models have been used for time series prediction of distresses on all pavement sections of the identified National Highway network. The HDM-4 pavement deterioration models, considered in this study have been validated by comparing the values of distresses predicted by the HDM-4 model with those observed in the field. The validation has been carried out through % variability, iii coefficient of determination (R2), and T test. Variability of 10.8 to 28.2 percent for cracking area, 15.4 to 39.8 percent for ravelling area, 0 to 66 percent for pothole units and 2.1 to 15.1 percent for roughness have been obtained. The variability obtained for rut-depth, and skid resistance ranges from 12.3 to 40.5 percent, and 2.5 to 30 percent respectively. The above variations were expected and accepted for such complex phenomenon of pavement behaviour under varied conditions of soil type, pavement composition, terrain, traffic loading, and climatic conditions. Also, the regression analysis has been carried out and correlation equations have been developed for the above distress values. The Coefficient of Determination (R2) values for cracking, ravelling, pothole, roughness, rut-depth, and skid resistance models have been obtained as 0.98, 0.79, 0.88, 0.97, 0.97, and 0.67 respectively. The R2 values show good agreement between observed and predicted distress values. The T test has been carried out to find out the significance of difference between the observed and predicted distress values. The f values calculated for cracking, ravelling, potholing, roughness, rutting, and skid resistance models are 0.572, 0.456, 1.588, 0.614, 1.199, and 1.250 respectively. The tabulated 't' value for 5% level of significance is 1.76, and all the calculated 'f values are less than this value. This indicates that the difference between the observed and predicted distress values is not statistically significant at 5% level of significance. Therefore, the calibrated HDM-4 deterioration models can be used for prediction of distresses and the development maintenance management strategies for the selected National Highway network. The application of the developed PMS methodology has been made for project level and network level PMS, for the National Highway network. The 'Project Analysis' application module of HDM-4 has been used to carry out Project Level PMS analysis. An optimum maintenance and rehabilitation strategy for a pavement section has been determined on the basis of highest NPV/Cost ratio, amongst a number of pre-defined M&R strategies. This M&R strategy comprises of application of overlay (40 mm Bituminous Concrete) at a pre-defined intervention level (activated when Roughness >2.8 m/km IRI). An optimum improvement strategy of widening of an existing single lane pavement section has been determined. As per this improvement strategy, there is no immediate need of addition of an extra lane to the existing pavement section. The widening of pavement section may be deferred at iv least for two years and may be taken up in the year 2005. This will result in a saving of about Rs. 7 million in highway agency costs. The remaining service life (RSL) for five pavement sections of the National Highway network has been determined which varies from 1 to 4 years, indicating that reconstruction on most of the pavement sections would be inevitable in the near future, in the absence of any sound maintenance management policy. The adoption of condition responsive maintenance strategies in comparison with the scheduled type of maintenance strategies indicate a saving of more than 33% in highway agency costs, over an analysis period of 20 years. The 'Program Analysis' application module of HDM-4 has been used for the network level pavement management analysis. The life-cycle cost analysis of the highway network has been carried out, and an unconstrained works programme has been prepared for a analysis period of 10 years (i.e. 2003 - 2012). The total budget requirements for maintenance management of the whole highway network at a pre defined optimum serviceability level have been determined, which comes out to be Rs. 1475.8 million. In India, the usual level of fund allocation for the maintenance management of National Highways does not exceed 60% of the requirements. Therefore, an optimized and prioritized works programme, has been developed for 60% (i.e. Rs. 885.5 million) budget. The prioritization of pavement sections has been done on the basis of decreasing NPV/Cost ratio. The funds requirements for maintaining whole of the highway network at the highest level of serviceability (Rs. 1871.8 million) is about 27% more than those required for maintaining it at the optimum level of serviceability. However, it would be possible to maintain the same highway network at a medium level of serviceability at about 74% of funds (Rs. 1089.1 million). But since the budget availability is normally 60%, it would just be possible to maintain this highway network at the lowest level of serviceability. The effect of reducing budget levels and the effect of deferring maintenance works on the overall condition of the highway network has also been determined. The average roughness value of the highway network increases from 3.2 m/km IRI to 4.4 m/km IRI, with gradual reduction in budget levels from 100% to 50%. The average roughness value of the highway network will not change much if the maintenance is delayed by one year, but in case the application of maintenance activities is deferred by two years, the roughness value will rise very sharply to 6 m/km IRI in the year 2005. Afurther delay of one year will cause rise in roughness value up to 8 m/km IRI in the year 2006. Both these scenarios of reducing budget allocations, and deferring maintenance, will result in very high vehicle operation cost for the road users. Geographic Information System (GIS) adds tremendous functionality to a pavement management program not only in graphical output but also in analysis and reporting. In this study, the results obtained through the network level pavement management analysis have also been presented through applications of GIS. This helps in graphical identification of the candidate pavement sections, due for maintenance during the analysis period of 10 years. Query analysis has been conducted in GIS for graphical identification of those pavement sections, which would be due for maintenance in the current year, using the following criteria: Traffic level 5000 - 10000 AADT, Roughness >= 4 m/km IRI, and Total crack area >= 10%. Eight pavement sections have been identified on the spatial map of the identified National Highway network, which fulfill these criteria. The condition characteristics of all the pavement sections in the future years, 10-year M&R works programme, annual fund requirements for carrying out M&R works, and the total life-cycle budget requirements, in respect of all pavement sections of the National Highway network have been presented in the GIS environment. The Pavement Management System methodology, developed in this study would be useful for the highway agencies in planning pavement maintenance strategies in a scientific manner and ensuring rational utilization of limited maintenance funds. Graphical presentations of PMS results will also be useful for gaining better support from decision-makers, for adequate and timely fund allocations for preservation of the highway network. This methodology may be used for preparing the maintenance management programme for the National Highways covered under the prestigious National Highway Development Programme (NHDP) of India. Once this PMS methodology for National Highway network is implemented and made operational for a few years; this would serve as a window to State Governments, particularly for State Highways and Major District Roads. Thereafter, similar kind of PMS may be developed for other categories of roads such as State Highways and Major District Roads, using the inputs of PMS methodology presented in this study. VI