ACCURACY AND CONSISTENCY OF BACKCALCULATED PAVEMENT LAYER MODULI
Y. J. Chou, Robert L. Lytton
Abstract
Y. J. Chou, Robert L. Lytton
Abstract
Backcalculation of pavement layer moduli from deflection measurements has been a focus of recent pavement research because of the need to adopt mechanistic design and analysis methods and the widespread use of nondestructive testing devices such as the falling weight deflectometer and Dynaflect. A number of pavement structural models and computerized procedures have been developed to perform backcalculation. However, the results often vary among analysts because of the assumptions made in each procedure and the different input assigned by individual analysts. Such variation causes concern because engineers will not have confidence using the backcalculated moduli in pavement evaluation or design if the size of the error associated with the backcalculated moduli values is unknown. To better understand the accuracy and consistency of the available backcalculation procedures, backcalculation results from different agencies using various procedures were compared. The results indicate that discrepancies among agencies can be large; however, a few agencies reached good agreement in many cases. The sources of systematic errors and ways to reduce them are discussed. Finally, an expert system approach that uses a specific analyst's knowledge to prepare input for the mechanistic backcalculation program and to interpret results is described.
OpenAlex reports 43 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Backcalculation of pavement layer moduli from deflection measurements has been a focus of recent pavement research because of the need to adopt mechanistic design and analysis methods and the widespread use of nondestructive testing devices such as the falling weight deflectometer and Dynaflect. A number of pavement structural models and computerized procedures have been developed to perform backcalculation. However, the results often vary among analysts because of the assumptions made in each procedure and the different input assigned by individual analysts. Such variation causes concern because engineers will not have confidence using the backcalculated moduli in pavement evaluation or design if the size of the error associated with the backcalculated moduli values is unknown. To better understand the accuracy and consistency of the available backcalculation procedures, backcalculation results from different agencies using various procedures were compared. The results indicate that discrepancies among agencies can be large; however, a few agencies reached good agreement in many cases. The sources of systematic errors and ways to reduce them are discussed. Finally, an expert system approach that uses a specific analyst's knowledge to prepare input for the mechanistic backcalculation program and to interpret results is described.
Key concepts: Falling weight deflectometer, Deflection (physics), Consistency (knowledge bases), Moduli, Engineering, Computer science, Civil engineering, Structural engineering