The major objective of the study was to develop a procedure to determine the level of service using the RWIS speed measurements. The procedure developed can be used by ODOT to evaluate winter maintenance activities and for winter maintenance decision making. Average traffic speeds for five minute intervals were measured using NuMetrics road sensors and they were related to the pavement and driving conditions. In addition speed data from two other studies was used. The pavement conditions were determined by conducting surveys at rest area buildings using a questionnaire form. It was found that the average traffic speeds were significantly lower during a major snow event even when periodic plowing and salting was done. The average speeds decreased almost linearly for the period of the snow storm reached the minimum and then climbed back slowly towards higher speeds. The speeds appear to be a fairly sensitive measure to judge the condition of the pavement. The motorist judgments about the pavement condition and their perception of the safety of driving decreases during a rather severe winter storm which is mirrored in the speed decrease. It appears from the survey that about two thirds of the motorists judge the deterioration of the road conditions and the inadequate level of road maintenance during a winter storm as bad or moderately bad. The responses obtained for the car and the truck drivers are fairly close to each other indicating that both groups can judge bad road condition equally well. The observed road conditions appear to influence the drivers in terms of how they subjectively feel about the level of safety and stress experienced during driving in the winter storm. A simple procedure was developed for winter maintenance management to determine the condition of the road (freeways) based on the average speeds observed by the RWIS sensors. If the average winter speed of the traffic is equal or greater than the historical established wet/salted pavement speed, the level of service is considered adequate. According to the Swiss study, the wet/salted surface winter speeds are about 85% of the dry surface speeds for freeways and 96 % for city streets. If the average winter speed is below the wet/salted surface speed, the level of service is considered inadequate. Any speed less than 50% of the wet/salted surface speed indicates fairly bad road conditions and an extremely inadequate level of service. It should be noted that the winter pavement conditions can be highly dynamic. Depending on the rate of accumulation of snow, frequency of the snow plowing, length of the snow plow route, the pavement condition can improve and deteriorate a number of times during a winter storm. The level of service can get worse even with maximum snow plowing and salting effort in a situation with a high rate of snow accumulation. The winter speeds observed as a percentage of the average dry surface speed can be correlated with the level of service. A relatively more fine graduation of the level of service as a function of the percentage of the average dry surface speed is proposed in the recommendations of the report.

This study addressed pretreatment protocol for winter maintenance of roadways using brine. Information regarding pretreatment was assessed from surveys of personnel in state departments of transportation and county garages in Ohio. Field durability studies of various applications of brine were conducted on Portland cement concrete and asphalt concrete pavements in Ohio. Over three winter seasons, weather events and resulting pavement conditions were documented during pretreatment and during the subsequent events using visual and limited road grip tester (RGT) assessment. In addition, extensive laboratory studies were undertaken to supplement the field investigations. Integration of the findings resulted in a decision tree to aid in operational planning and pretreatment.

Road Weather Information System (RWIS) pavement sensors from three vendors were evaluated under controlled conditions in a climate chamber. The chamber had a maximum cooling rate of about 5.6°C in the first 15 minutes (about 15.8°C in the first hour) and could cool from a room temperature of about 22°C to a minimum temperature of approximately -17°C in about 15 hours. One vendor supplied an active (cooling/heating cycle) sensor with an active-passive sensor as a combined system, and the other two supplied passive sensors. Each sensor was installed in a concrete block cut from a bridge deck and positioned in a large climate-controlled chamber. The sensors were tested for accuracy and precision of temperature, freezing point determination, chemical percentage or index, surface status, and liquid depth, where appropriate. Testing parameters included salt type (pure water, 7%, 13%, & 19% NaCl in water, 17% & 30% CaCl2 in water, and two mixtures of the two salts in solution), and liquid depth (0.5 mm, 1.5 mm, 3 mm, 6 mm). The temperature in the chamber was reduced until the liquid froze or a minimum temperature below -15°C was attained. Independent temperature probes were placed on and near the block for comparison to sensor readings. Results obtained for the three sensors varied considerably. The surface status (wet/dry) reported by Vendor C's sensor was accurate at the beginning of each run 100% of the time, but accurate only 81% of the time at the end of the run. For Vendor A the corresponding accuracy rates were 97% and 92%. Vendor B's more detailed scheme was accurate only 19% of the time at the beginning of each run and 47% at the end. The surface temperature reported by each passive or active-passive sensor during cooling of the climate chamber was found to lag behind that measured by the independent probes placed in the liquid and on the surface of the block. The average maximum lag in temperature and the average time at which it was observed for each vendor were: Vendor A: 5.6°C at 150 minutes, Vendor B: 4.0°C at 157 minutes, and Vendor C: 7.0°C at 128 minutes. Vendor A's active sensor freezing point values varied considerably from expected values, and also changed during runs. The chemical percentage reported by Vendor B's passive sensor showed an appropriate variation with NaCl concentration, though at higher concentrations the initial reported concentration was low; this bias was also reflected in the freezing point values computed by the system. Percentages were less accurate with CaCl2 or NaCl-CaCl2 mixtures, for which the system was not calibrated. Vendor C's chemical index did not appear to correlate with salt concentration at all. Because of these results, none of these sensors in their present state is recommended for use at the present time. Any future sensor deployment should be contingent on the ability of the chosen sensor to perform those functions considered necessary by ODOT.

The Ohio Department of Transportation (ODOT) has initiated pretreatment as an integral part of a winter management strategy. Currently forty gallons per lane mile of 23% salt brine (NaCl) by weight is applied at a minimum frequency of two times per week when conditions warrant. In order for ODOT to develop the most effective plan for pretreatment, an in-situ study to provide data on decay of brine on trafficked pavement was needed. Objectives included a survey of other state DOT's pretreatment protocols, laboratory studies to discern brine concentrations that precluded ice formation, brine decay with traffic and time on several pavements, and correlation of laboratory and field data. Ten of the 28 state DOTs responding to the survey regarding pretreatment protocol applied NaCl two to 24 hours prior to a storm; two states used surface type, traffic volume, and air temperatures for decision making. The survey reinforced the need of laboratory and field studies. In the laboratory, release temperatures of the ice/surface bond at various brine concentrations were obtained utilizing conductivity and physical observation techniques. Laboratory tests with the field brine measurement instrumentation (SOBO-20 by Boschung Megatronic AG) provided correction factors for the field data on AC and PCC pavements. Sodium chloride brine was applied and measured in-situ in mass per area at five field sites (ATH-50 PCC, ATH-50 AC, DEL-23 PCC, DEL-23 AC, and ATH-33 AC) encompassing at least four sections at each site. Initial losses and decay due to time/traffic were obtained. Of the five test sites, AC (micro seal), AC (NOVA chip), and a transversely grooved PCC pavement provided statistically valid data to develop residual decay equations as a function of time/traffic. Field decay of brine was incorporated into laboratory brine/ice/specimen bonding temperature findings to determine the effective ice prevention temperatures as a function of time/traffic for AC and PCC at standard application rates.

TRB's National Cooperative Highway Research Program (NCHRP) Synthesis 344: Winter Highway Operations examines changes that occurred between 1994 and 2004 to practices and strategies used to control the impacts of winter weather on the safe and efficient movement of traffic.

"This synthesis report will be of interest to transportation agency maintenance engineers, managers, and operators and others involved with roadway snow and ice control including safety engineers, traffic engineers, and law enforcement agency personnel. It presents information on the state of the practice in managing roadway snow and ice control considering both rural and urban locations. The document describes the developments that have occurred during the past 20 years to improve winter maintenance. This report of the Transportation Research Board discusses winter maintenance policies and provides examples for state, city, and county agencies. Included in the discussion of winter maintenance policies are issues such as: levels of service, public relations, liability for services, and experimenting with new policies. Additional information is included on estimating winter maintenance benefits and costs; personnel and management issues; weather information systems; and materials, equipment, and facilities for winter maintenance."--Avant-propos.

Snow and ice management is the single largest expenditure in the maintenance budget for the Ohio Department of Transportation (ODOT) with an annual cost including labor, equipment, and materials reaching approximately $86 million (ODOT, 2013). One method to increase the efficiency of operations is the implementation of a Global Positioning System (GPS) and Automatic Vehicle Location (AVL) resource management system. Phase One of the project, was to implement 22 trucks with systems that work with the hydraulic system on the truck. During this phase, the research team was able to implement one system that works with both primary hydraulic systems in the ODOT fleet. With success in Phase One, Phase Two was developed to scale-up the fleet from 22 to 187 trucks throughout the state. The deliverables of this project are the details of installing and troubleshooting the system and the website developed for ODOT and with ODOT. The website allows real-time data and historical data for the truck location and road conditions. The website is able to rely various data platforms depending on what ODOT wants to implement. The cost of the systems will vary depending on the level of detail the end users would like to receive. More sensors requires more capital, maintenance and calibration cost.