SPR-4716: Lessing Density Requirement and Adjusting Density Pay Factors for Asphalt Pavement in Poor Sublayer Condition
Asphalt compaction is an essential task that can affect the quality of the constructed asphalt pavement (e.g., cracking, crumbling, and uneven surface). On moving sublayers (e.g., subgrade in poor condition or soft cold in-place recycling asphalt), it is difficult to compact asphalt as the sublayers absorb compaction energy. The higher density of up to 98 %Gmm results in a higher pay factor (as per INDOT Standard specification, 2022). The contractors are paid by pay factor based on the density of the surface layer., hence, contractors target a higher density. If contractors aim for a higher density on poor sublayer conditions, it can cause issues such as density segregation and aggregation crushing. The problem is that we do not know the scientifical and statistical correlation between sublayer conditions and the density of the surface, and the density pay factor in the current specification does not consider sublayer conditions. Therefore, the goal of this research is to investigate a correlation between asphalt density (%Gmm) and sublayer conditions. And to recommend adjusting the density pay factor in the specification depending on the correlation.
Point cloud segmentation and GIS spatial analysis
The collection and analysis of existing built environment data can help urban and rural planning departments to build a reasonable infrastructure distribution. How to collect and process data more accurately should be a key consideration. Lidar scanning as a high-precision data acquisition method can provide maintenance personnel with an effective digital copy of our real world. This project proposes a framework that uses point cloud segmentation algorithms to extract specific types of infrastructure and then uses GIS (geographic information system) to analyze the spatial distribution of infrastructure based on the extracted results to help inspectors efficiently check the rationality of infrastructure spatial distribution design. This study uses fire hydrants and buildings on the Purdue University campus as the object of study and attempts to automate the analysis to obtain the closest distance between the hydrants and the buildings. Point cloud data were collected using Trimble X7 and 43 hydrant stations were collected with the surrounding buildings and environment. Then, point cloud segmentation was used to extract point cloud instances of fire hydrants and buildings, and then the point clouds were converted into Raster data for import into GIS for further processing to achieve nearest distance estimation.
Water Distribution Pipe
The water distribution network is an essential infrastructure that can provide good quality water for human life. In many countries, most water distribution network systems have a variety of defects that negatively affect water quality due to their long service history. Proactive inspection to assess the current defects and conditions of water distribution pipes is a critical effort for establishing the appropriate strategies to keep the water distribution network in the best condition. However, the existing condition assessment using proactive inspection is not only time-consuming work that requires repetitive visual checking but also subjective work as it depends on the opinion of skillful experts.
This project aims to develop multiple platforms of the automatic defect detection system for water distribution pipes based on the computer vision approach to improve the condition assessment process. The proposed platforms can contribute to not only reducing the time and workforce required for the condition assessment process but also enhancing the accuracy of condition assessment. This project will benefit to introduce the technology for smart water facility management in the future.