Missing data in pavement condition and performance records of pavement management systems (PMS) are ubiquitous in practice. Imputation of missing data is often required in the analysis of pavement performance and decision making for pavement management. The traditional methods of handling missing data by pavement engineering professionals include deletion of affected records, and imputation of missing data either by means of interpolation substitution, mean substitution, or regression substitution. Today, the advancement of computer technology has permitted the use of computationally complex stochastic Multiple Imputation algorithms to improve the accuracy of missing data estimates. This paper examines the effects of different available missing-data imputation techniques in handling missing pavement performance data in pavement management systems. The methods of Multiple Imputation are also examined to take into account the stochastic nature of the data imputation problem. Demonstrative examples using actual records from LTPP database are presented to illustrate the relative merits of different missing data imputation techniques.

Presented during Session 16: Data Quality, moderated by Gonzalo Rada, at the 9th International Conference on Managing Pavement Assets (ICMPA9) in Alexandria, VA

Includes conference paper and PowerPoint slides.