COMPARISON BETWEEN RANKING AND OPTIMIZATION MODELS IN SETTING PAVEMENT MAINTENANCE PRIORITIES

Abstract

The rapid growth in road construction that took place in the past two decades brought about considerable expansion of road infrastructure, which subsequently fell into disrepair through lack of maintenance. The damage is often so severe that ordinary maintenance will no longer suffice and if roads are to be fully restored, rehabilitation or even reconstruction work is necessary, at a life cycle cost three to seven times higher than that of preventive maintenance strategies. To apply such strategies, information concerning the condition of the network, its rates of deterioration, and the impact of different maintenance and rehabilitation treatments on pavement serviceability level, are required. Such components are typically integrated through the application of Pavement Management Systems (PMS). The quality of such PMS outputs is dependent not only on the above data but also on the schemes (models) which utilize this data to produce the prioritized list of short and long term maintenance and rehabilitation activities under the prevailing budget and other constraints. Therefore, the General Authority for Roads and Bridges and Land Transport (GARBLT) decided to develop a comprehensive PMS for Egyptian network. Due to the complexity of such a PMS, its procedure was implemented in a computer program, known as Highway Pavement Management Application (HPMA). HPMA is a computerized information system focused specifically on pavements. It provides access to pavement and road related data in a manner that can be easily retrieved for purposes of editing, reporting, and analysis. The analytical routines allow the user to objectively assess the current status and estimate the rehabilitation needs and/or maintenance needs of the network. Rehabilitation or maintenance programs can be developed for a variety of user-defined budget scenarios. In this research, the data collected through the GARBLT was utilized to test the effectiveness of several maintenance priority setting techniques ranging from a simplified worst first scheme (ranking) to optimization models with different constraint levels using the HPMA program. The outputs of this research are the distribution of the Pavement Condition Index along the surveyed network, the percent of network lengths less than the trigger value, the repaired lengths, and the cost per kilometer. The results indicated the superiority of the optimization models in terms of better overall network condition, more repaired lengths and less cost per kilometer.