Imputation of activity travel sequences from GPS data is dominantly based on epoch-level characteristics and land use data. This study showed that the accuracy of GPS data imputation can be improved by enforcing consistencies in transportation mode use across the day. In particular, a new algorithm was proposed to reduce misclassification error in imputed activity travel diaries. The suggested algorithm identified a hierarchical set of tours and superimposed logically consistent transportation modes at the tour level. Three methods, which maintain some degree of consistency while allowing different transportation modes within a tour, were examined. Method 1 identified the most probable transportation mode for each trip episode separately and then compared the modes across all segments of the tour. Method 2 selected the most probable mode across the whole travel episode. Method 3 selected the most likely main mode in the intermediate portion of a tour as the mode in all trip segments. All methods imputed transportation mode based on the highest number of epochs for which the predicted probability of that mode was the highest. The algorithm was examined by using GPS data recently collected in the Netherlands. Results showed that the new algorithm significantly improved the imputation accuracy of transportation modes. The enhanced algorithm, which partly relaxed the strict assumption of consistency, yielded results even closer to reality.
Enhanced Imputation of GPS Traces Forcing Full or Partial Consistency in Activity Travel Sequences
Comparison of Algorithms
Transportation Research Record
Transportation Research Record: Journal of the Transportation Research Board ; 2430 , 1 ; 20-27
2014-01-01
Article (Journal)
Electronic Resource
English
Semiautomatic Imputation of Activity Travel Diaries
Transportation Research Record | 2010
|Integrated imputation of activity-travel diaries incorporating the measurement of uncertainty
Taylor & Francis Verlag | 2019
|