Gravel roads are an economical choice where traffic volume is low. Many gravel roads were constructed in rural areas in the Kyela district and they were well-served for many years. Besides the gravel roads, the main c...Gravel roads are an economical choice where traffic volume is low. Many gravel roads were constructed in rural areas in the Kyela district and they were well-served for many years. Besides the gravel roads, the main challenge faced is the issue of unpaved road condition deterioration by the behaviors of road materials. The study selected three gravel rural roads with a total of 31.24 kilometers as constructed roads to gravel standard. The aim of this study was to assess gravel roads condition to examine trends and behaviors of gravel roads in rural areas. A descriptive analysis was conducted to determine trends and behaviors of gravel roads. The methods used to collect data were observation and physical measurement. The results show that there was more than one road distress identified and assessed on each study road. It was noted that all the study roads had excellent to poor distress conditions associated with poor road materials and lack of maintenance. This study recommended that road authorities improve the current condition by avoiding the use of poor road materials in order to reduce surface deterioration associated.展开更多
The current limitation in maintenance budget and resources necessitates developing new cost-effective techniques for gravel roads management systems (GRMS). Thus, the Wyoming Technology Transfer Center (WYT2) has star...The current limitation in maintenance budget and resources necessitates developing new cost-effective techniques for gravel roads management systems (GRMS). Thus, the Wyoming Technology Transfer Center (WYT2) has started developing a holistic automated GRMS. Utilizing smartphones in gravel roads data collection is one of the main features in the proposed system. In this study, smartphones were used to collect gravel roads condition data in terms of International Roughness Index (IRI) and corrugation to develop an objective computational method to estimate the riding quality on gravel roads. The developed method will help local agencies to reduce subjectivity in their data collection process and support them with a solid computational justification for their evaluation data and decisions. Two analyses have been carried out to achieve the purpose of this study. Artificial Neural Network ANN method and linear regression were used to develop the riding quality model. The linear regression resulted in a model that has a 0.8242 coefficient of determination (R2) value which means that the developed riding quality model can represent 82.42% of the collected data. The achieved R2 value is considered sufficient for GRMS purposes. In addition, the developed ANN model has a prediction accuracy of 92.5%. The achieved prediction accuracy shows that the ANN model can predict the riding quality significantly better than the linear regression, with 12.5% higher accuracy. Furthermore, thresholds for the gravel roads IRI were suggested and introduced in this study to be the first IRI thresholds for gravel roads in the literature. Based on the suggested threshold, the gravel roads IRI has three classes: smooth, acceptable and rough. The gravel road segment can be classified in terms of IRI to be smooth, acceptable, or rough if its IRI value is less than 284, between 284 and 496, or more than 496 inch/mile, respectively.展开更多
文摘Gravel roads are an economical choice where traffic volume is low. Many gravel roads were constructed in rural areas in the Kyela district and they were well-served for many years. Besides the gravel roads, the main challenge faced is the issue of unpaved road condition deterioration by the behaviors of road materials. The study selected three gravel rural roads with a total of 31.24 kilometers as constructed roads to gravel standard. The aim of this study was to assess gravel roads condition to examine trends and behaviors of gravel roads in rural areas. A descriptive analysis was conducted to determine trends and behaviors of gravel roads. The methods used to collect data were observation and physical measurement. The results show that there was more than one road distress identified and assessed on each study road. It was noted that all the study roads had excellent to poor distress conditions associated with poor road materials and lack of maintenance. This study recommended that road authorities improve the current condition by avoiding the use of poor road materials in order to reduce surface deterioration associated.
文摘The current limitation in maintenance budget and resources necessitates developing new cost-effective techniques for gravel roads management systems (GRMS). Thus, the Wyoming Technology Transfer Center (WYT2) has started developing a holistic automated GRMS. Utilizing smartphones in gravel roads data collection is one of the main features in the proposed system. In this study, smartphones were used to collect gravel roads condition data in terms of International Roughness Index (IRI) and corrugation to develop an objective computational method to estimate the riding quality on gravel roads. The developed method will help local agencies to reduce subjectivity in their data collection process and support them with a solid computational justification for their evaluation data and decisions. Two analyses have been carried out to achieve the purpose of this study. Artificial Neural Network ANN method and linear regression were used to develop the riding quality model. The linear regression resulted in a model that has a 0.8242 coefficient of determination (R2) value which means that the developed riding quality model can represent 82.42% of the collected data. The achieved R2 value is considered sufficient for GRMS purposes. In addition, the developed ANN model has a prediction accuracy of 92.5%. The achieved prediction accuracy shows that the ANN model can predict the riding quality significantly better than the linear regression, with 12.5% higher accuracy. Furthermore, thresholds for the gravel roads IRI were suggested and introduced in this study to be the first IRI thresholds for gravel roads in the literature. Based on the suggested threshold, the gravel roads IRI has three classes: smooth, acceptable and rough. The gravel road segment can be classified in terms of IRI to be smooth, acceptable, or rough if its IRI value is less than 284, between 284 and 496, or more than 496 inch/mile, respectively.
