Realistic Life-Cycle Cost Analysis with Typical Sequential Patterns of Pavement Treatment through Association Analysis

Abstract

Advancements in data collection, digital data storage, and database management technologies have dramatically increased the capabilities of state highway agencies to generate and to store highway project data. The Oklahoma Department of Transportation (DOT), along with other state DOTs, stores a large amount of highway project data throughout the life cycle of highway projects. Although much of the budget is allocated to collecting and maintaining pavement data sets, the amount of benefit from extracting information and knowledge from the data is limited. This study applies association analysis in data mining to the historical treatment data set available in the Oklahoma DOT to identify the typical sequential patterns of treatment activities. Then, a realistic life-cycle cost analysis (LCCA) model based on typical sequential patterns of pavement treatment activities is developed and compared with the traditional LCCA performed by RealCost through an example. The case study indicates that the results of the realistic LCCA can be significantly different from those of the traditional approach. Because life-cycle cost models developed in this study are based on the actual treatment strategies performed by the Oklahoma DOT, it is expected that the life-cycle cost is closer to actual costs than the traditional approach. Using the realistic LCCA model developed in this study, state DOTs will be able to develop more realistic pavement maintenance and rehabilitation strategies and budgets.