Development and Evaluation of Models and Algorithms for Locating RWIS Stations
Bibliographic record
Abstract
Accurate and timely information on road weather and surface conditions in winter seasons is a necessity for road authorities to optimize their winter maintenance operations and improve the safety and mobility of the traveling public. One of the primary tools for acquiring this information is road weather information systems (RWIS). While effective in providing real-time and near-future information on road weather and surface conditions, RWIS stations are costly to install and operate, and therefore can only be installed at a limited number of locations. To tackle this challenging task, this thesis develops various different approaches in an attempt to determine the optimal location and density over a regional highway network. The main research findings are summarized as follows. \n \nFirst, a heuristic surrogate measure based method (SM) has been developed. Two types of location ranking criteria are proposed to formalize various processes utilized in the current practice, including weather and traffic related factors. Consideration of these two types of factors captures the needs to allocate RWIS stations to the areas with the most severe weather conditions and having the highest number of traveling public. A total of three location selection alternatives are generated and used to evaluate the current Ontario RWIS network. The findings indicate that the current RWIS network is able to provide a reasonably good coverage on all location criteria considered. \n \nSecond, a cost-benefit based method (CB) has been proposed to give an explicit account of the potential benefits of an RWIS network in its location and density planning. The approach has been constructed on a basis of a sensible assumption that a highway section covered by an RWIS station is more likely to receive better winter road maintenance (WRM) operations. A case study based on the current RWIS network in Northern Minnesota show that the highest projected 25-year net benefits are approximately $6.5 million with cost-benefit ratio of 3.5, given the network of 45 RWIS stations. \n \nThird, a more comprehensive and innovative framework has been developed by using the weighted sum of average kriging variance of winter road weather conditions. Methodologically, the formulation of the RWIS location optimization problem is foundational with several unique features, including explicit consideration of spatial correlation of winter road weather conditions and high travel demand coverage. The optimization problem is then formulated by taking into account the dual criteria representing the value of RWIS information for spatial inferences and travel demand distribution. The spatial simulated annealing (SSA) algorithm was employed to solve the combinatorial optimization problem ensuring convergence. A case study based on four study regions covering one Canadian province (Ontario), and three US states (Utah, Minnesota, and Iowa) exemplified two distinct scenarios –redesign and expansion of the existing RWIS network. The findings indicate that the method developed is very effective in evaluating the existing network and delineating new site locations. \n \nAdditional analyses have been conducted to determine the spatial continuity of road weather conditions and its relation to the desirable RWIS density based on the case study results of the four study areas. Road surface temperature (RST) was used as a variable of interest, and its spatial structure for each region was quantified and modelled via semivariogram. The findings suggest that there is a strong dependency between the RWIS density and the autocorrelation range - the regions with less varied topography tend to have a longer spatial correlation range than the region with more varied topography. \n \nThe approaches proposed and developed in this thesis provide alternative ways of incorporating key road weather, traffic, and maintenance factors into the planning of an RWIS network in a region. Decision on which alternative to use depends on availability of data and resources. Nevertheless, all approaches can be conveniently implemented for real-world applications.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".