Potentiality of tree variables as predictors in pavement roughness progression rate modelling

ABSTRACT

This paper presents statistical evidence of roadside vegetation’s contribution to pavement roughness progression rates. Detailed statistical and regression analysis of the roadside vegetation data collected via satellite imageries and road roughness data collected via high speed road profiler was performed. Elaborative investigation on interaction between roadside vegetation and waveband roughness progression has provided a clear indication of tree variable’s contribution on road deterioration. Statistical parameters such as moderate Pearson correlation coefficient (r) values, low mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE) values, high Willmott’s index of agreement (d) were obtained for training and validation datasets, that depicted the potentiality of tree variables as predictors in pavement roughness progression rate modelling. Statistical evidence showed that effect of trees on road deterioration was more noticeable on long wavelength roughness progression rates. This can be justified via prevailing soil moisture interaction in expansive soil deposits subjected to moisture withdrawal of deciduous trees in arid climate conditions. Overall, the findings of this paper exemplify on the necessity of considering the presence of roadside vegetation in road deterioration analysis, and suggesting the scope of improvement for prediction performance.

KEYWORDS:

• Expansive soil
• gilgai
• roadside vegetation
• IRI
• roughness progression rates
• pavement deterioration

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Md Yeasin Ahmed

Dr Md Yeasin Ahmed has completed his PhD from the Department of Civil and Construction Engineering, Swinburne University of Technology, Australia. His research interest includes pavement deterioration modelling, expansive soils, soil moisture interaction, influence of vegetation on light structures. He has published several journals and conference papers in his interest area. Currently he is working in the Australian construction engineering industry as a Pavement Asset Management Engineer/ Consultant and directly involved in many prestigious projects relevant to road deterioration data capture and management, deterioration modelling, road asset management, dilapidation studies, road maintenance and rehabilitation, and capital works program preparation for the existing assets.

Robert Evans

Dr Robert Evans is currently a Senior Lecturer in the Department of Civil and Construction Engineering, Swinburne University of Technology. He is responsible for the undergraduate courses in Geomechanics and Geotechnical Engineering within the Civil Engineering degree program. His qualifications comprise of a Bachelor of Engineering (Civil) with Honours, a Master of Engineering (by research in Civil Engineering), a Doctor of Philosophy (Civil Engineering), a Graduate Diploma in Management and a MBA in Management Technology. Before entering academia, he worked for Vicroads as well as a Geotechnical Consultant at Piper and Associates. His research areas of interest are soil mechanics, the behavior of expansive soils on light structures, analysis of pavement roughness by waveband analysis and deterioration modelling of pavements. He has published several journals and conference papers in his interest area.