Development of Zonal-Specific Semivariograms for a Strategic RWIS Network Optimization: Case Study

Abstract

This paper presents a study aimed at developing zonal-specific semivariograms for zones with different climates using regionalized random variables for a strategic road weather information system (RWIS) network implementation and optimization in a large region. Zonal semivariograms modeled in this study were explicitly compared with regional semivariograms to demonstrate the (dis)similarity in their underlying spatial structures. Large-scale RWIS location and density optimizations were conducted with two groups of semivariograms developed in terms of their weather characteristics, namely regional and zonal, and were conducted to compare outcomes and illustrate their distinct features. A case study based on the existing RWIS network in Ontario, Canada, was used to show the application of the proposed method. The findings indicate that there are very different spatial autocorrelation patterns between regional and zonal-specific semivariograms, thereby emphasizing the need for a strategic zonal-specific RWIS implementation plan. The results of different planning scenarios for optimizing RWIS network also reveal that although the optimal locations are insensitive to the underlying spatial structure (i.e., semivariogram) used to optimize the network, the optimal density is found to be very sensitive to such, providing important yet useful decision-making guidance for improved efficiency and effectiveness of overall winter road maintenance programs.