article · 2003

DATA IMPUTATION STRATEGIES FOR TRANSPORTATION MANAGEMENT SYSTEMS

Cited by 18Abstract and details only

Abstract

Many states have implemented large-scale transportation management systems to improve mobility in urban areas. These systems are highly prone to missing and erroneous data, which result in drastically curtailed data sets for analysis and real-time operations. As a general rule, the surface transportation industry does not impute missing or erroneous traffic data. Other disciplines have recognized the importance of addressing missing data and, as a result, methods and software for imputing missing data are becoming widely available. The objective of this research is to find an automated imputation strategy for imputing missing data in a traffic data archival system. This research addresses the feasibility and applicability of imputing missing traffic data, and performs an analysis on five imputation techniques. The first technique examined in this study is historical average imputation. The next two techniques are new heuristic approaches designed specifically for missing traffic data. Finally, the last two are statistical imputation procedures, which have been developed and implemented in various other fields. These approaches for imputing missing data are implemented and compared in this study. The results of this study indicate that of the tested models, the EM algorithm generates the most accurate imputations. For applications where the accuracy of a single measurement is not a high priority, simpler techniques such as the historical average imputation may work adequately. Imputed data were tested in several applications to access the propagation of imputation error. The result of this test indicated that the data could be used in applications without drastically changing the outputs of the application.

Cite this paper

APA 7

Conklin, J. H., & Scherer, W. T. (2003). DATA IMPUTATION STRATEGIES FOR TRANSPORTATION MANAGEMENT SYSTEMS.

MLA 9

Conklin, James H., and William T. Scherer. “DATA IMPUTATION STRATEGIES FOR TRANSPORTATION MANAGEMENT SYSTEMS.” 2003.

Chicago

Conklin, James H., and William T. Scherer. 2003. “DATA IMPUTATION STRATEGIES FOR TRANSPORTATION MANAGEMENT SYSTEMS.”

Harvard

Conklin, J.H. and Scherer, W.T. (2003) ‘DATA IMPUTATION STRATEGIES FOR TRANSPORTATION MANAGEMENT SYSTEMS’.

BibTeX

@article{conklin2003data,
  title = {{DATA IMPUTATION STRATEGIES FOR TRANSPORTATION MANAGEMENT SYSTEMS}},
  author = {Conklin, James H. and Scherer, William T.},
  year = {2003}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.