Carbon management of infrastructure performance: Integrated big data analytics and pavement-vehicle-interactions

Abstract

As a crucial part of the transportation system, roadway network provides mobility to the society and is vital for the economy. At the same time it contributes significantly to the environmental footprint during its construction, operation and maintenance. Hence, the sustainable development of our Nation's roadway system requires quantitative means to link infrastructure performance to lifecycle energy use and greenhouse gas emissions. Recent developments in mechanistic models of roughness- and deflection-induced pavement-vehicle interaction aim at providing such engineering estimates. Herein, it is demonstrated that these models when implemented at a network scale are a powerful basis for big data analytics of excess-energy consumption and carbon dioxide emissions by integrating spatially and temporally varying road conditions, pavement properties, traffic loads and climatic conditions. A novel ranking algorithm is proposed, that allows upscaling of the local carbon dioxide emissions due to pavement vehicle interaction to the size of state-wide or national sustainability goals. Implemented for 5157 lane-miles of the interstate highway system in the State of Virginia, sections contributing significantly to carbon dioxide emissions are identified. It is shown that the proposed ranking algorithm based on the inferred emission that exhibits a power-law distribution, provides the shortest path for greenhouse gas emissions savings per maintenance at network scale. That is, maintaining a few lane miles allows for a significant synergetic improvement of both infrastructure performance and environmental impact of the interstate network and helps transportation agencies in making economic and environmentally sustainable decisions.