Semisequicentennial Transportation Conference Proceedings
May 1996, Iowa State University, Ames, Iowa

A Methodology for Conducting Research into Winter Highway Maintenance

Wilfrid A. Nixon and Leland Smithson

W. Nixon,
Iowa Institute of Hydraulic Research,
University of Iowa,
Iowa City, Iowa 52242-1585.

L. Smithson,
Iowa Department of Transportation,
800 Lincoln Way, Ames, Iowa 50010.

Over the past five years, a number of studies have been conducted at the University of Iowa to examine new techniques for removing ice from pavements. One upshot of these studies has been the development of a system of research which is extremely effective at bringing concepts through to consideration at the operational stage. This system has applicability to a wide range of problems. The research methodology begins with laboratory studies of how ice can be cut from a pavement surface. To study this, a special scraping machine was developed. This piece of equipment is capable of scraping ice from PCC pavement samples, while loads are measured. This system has shown how the cutting edge geometry affects the scraping forces. Further, it allows new geometries to be tried in small scale, before they are placed on trucks. After the laboratory testing, these cutting edges are then used on a truck with an underbody plow blade. The underbody blade is instrumented so that the ice-scraping loads can be determined. At a closed site, water is sprayed onto a PC surface to form a layer of ice up to 12.7 mm (0.5 inch) thick over a test track of approximately 100 m (109 yards) in length. The truck then scrapes the ice from the pavement while the loads are recorded and the test is videotaped. However, it can be argued that even the full-scale field, closed track tests described above are not fully representative of a true ice-roadway situation. To address this, two Iowa Department of Transportation trucks have been instrumented in the same way as the above described test truck. This then provides a clear route from laboratory testing to deployment. Not only can cutting edges be tested in this way, but other aspects of winter highway maintenance can also be considered.

The recent expansion of technology transfer programs in the United States (a growth which is mirrored overseas) indicates a growing realization that the transfer of research results to engineering practice is not a process which can be taken for granted. It has to be planned, and ideally should not be thought of as an addendum to the research in question. Rather, if at all possible, a research project should be designed with the aim of eventually transferring the knowledge gained clearly in mind from the outset.

Technology transfer in itself is not a well defined process, and in times past it has been reduced (often by the research community) to little more than the presentation of papers at conferences. While this is indeed part of technology transfer, it is only a small part. The much more critical part is taking a technique, prototype product, process, or idea that developed from research and bringing that to the marketplace or the workplace. If this is to be the end result of a research project, then thought should be taken as to how it will come about, from the outset of the project.

A critical part of this process is responsiveness to the needs of the end user. Research sometimes has a bad name with engineering practitioners, precisely because the research performed has either failed to solve problems at all, or has solved problems that are not the ones with which the practitioner is concerned. The issue of client responsiveness and fundamental research is even thornier (1). Fundamental research by its nature is far removed from practical application, yet the knowledge gained may be critical to achieving progress in the practical arena. If the end user is to feel served by such fundamental research, it must be set in the context of the end user's reality.

The purpose of this paper is to describe a research process which has attempted to address the issue of practical application from the outset. This has to be done within the terms of a specific research problem (in this case, the problem of the mechanical removal of ice from pavements), but it is hoped that the process described is sufficiently general to be applicable to other transportation research problems as well.


The process used in this research, and which is proposed herein as a model for use in a wide range of transportation areas, has five distinct phases. These are discussed in detail further below. The five phases include problem definition, an investigation of fundamental issues, some form of intermediate testing in which field type data may be collected without full-scale field trials, a process for field testing, and finally full acceptance and implementation of the end product.

It should be readily apparent that not all research projects will go through all five stages, nor should they be expected to. The most obvious example in which this will be the case is when an idea is tried out and found not to give the expected benefits. However, even in such cases, the process described is useful because it focuses on the overall problem which the research is intended to solve. Such a focused approach will limit what might be termed "research for the sake of research," in which a research issue acquires a life of its own and proceeds over many years, regardless of whether it can bring any benefits to the user community which it purports to serve.

Problem Definition

The process of defining a problem is critical to an eventual solution, but part of this process is understanding what forces have created the problem. This stage is also the part of the process which most often causes difficulties between practitioners and researchers. To a researcher, a problem is simply something about which we do not have full understanding. A practitioner sees problems as things which limit the ability to perform specific required tasks. The two do not necessarily overlap at all. As an example from the field of ice mechanics, the compressive strength of ice at high strain rates is not well understood (2), and thus from the researcher's viewpoint is clearly an issue worth studying. From the point of view of removing ice from roads (the practitioner's viewpoint) the compressive strength of ice is a marginal issue at best. This does not mean that it is a poor topic for research, merely that this particular topic is not an area in which both researchers' and practitioners' interests are served.

From the point of view of ice removal from pavements, what issues make this a particular problem to the practitioner? First, ice removal is not easy even using the best of current methods. Put another way, current methods do not provide a totally satisfactory solution. To make the issue even more critical, current methods have drawbacks which are growing in significance. That is, the application of salt to roads is becoming increasingly less acceptable. There are a number of driving forces behind this. Salt causes maintenance problems for the roadway itself, giving rise to significant corrosion problems in the reinforcing materials. It also gives rise to some environmental damage. This latter is true (although to a more limited degree) for sand, which is another critical part of current winter maintenance methods. Thus a second issue for ice removal might be termed "concern" over current methods.

A third driving force for change is economic. There is increasing dependence on "just-in-time" manufacturing in industry, and this requires a road system which can ensure standard delivery times in spite of weather conditions. Thus economics increases the level of service expectations from the end user. Tied in with this are safety issues. Because of excellent work in the past by those responsible for winter maintenance, the general public expects all roads to be open under all weather conditions (a slight exaggeration!). They should be safe no matter how bad the weather.

Other issues also play a role. For ice removal, these may include increasing material costs, problems with keeping a well trained workforce, concerns about changes in operating practice (e.g. contracting out winter maintenance activities), and so forth. All of these factors combine to create the research problem requiring solution, which in this case is "How do we remove ice from the roads more effectively—using less salt and other materials, while still maintaining or improving our level of service?" This process is summarized in Figure 1.

Even at this stage however, the process is still rather diffuse. The problem has to be refined into a specific mode of attack. In the case of the work considered here, that mode was to examine how ice might be removed more effectively using mechanical methods, specifically to determine the best shape for the ice removing cutting edge that is mounted on the mold board. Refining the process is the stage at which the conceptual problem (removing ice)

becomes a series of experiments or studies from which firm conclusions can be drawn (what cutting edge shape is best?).

Fundamental Issues

Although considerations of fundamental issues may not be part of the research process, it is important to be aware that such issues exist. For ice removal these issues are relatively straightforward: How does ice bond to the pavement, and how can we break that bond?

These are not issues that are confined to ice alone, nor are they in any way simple. The issue of adhesion between two bodies has been of considerable interest in the microelectronics and adhesives industries for many years and remains so today. Even measuring the adhesion between two bodies is not a trivial exercise, and it involves complex mathematics and stress analysis to develop a framework in which to pose the questions that are relevant (3).

While these issues do not feed directly into the concern at the heart of the project considered here (removing ice from the road mechanically), there is clearly a relationship. Future developments in this area may well rely on work being done today to develop understanding of ice adhesion.

Intermediate Steps

The key part of the initial research was aimed at determining the optimal shape of a cutting edge for scraping ice from the pavement. To do this required the development of a special testing apparatus which could scrape ice at speeds similar to those used in the field for scraping ice. There were two reasons for imposing such a restriction. First, ice is a rate sensitive material, and thus the rate of scraping is likely to have some effect on the scraping forces measured. Second, practitioners may find it hard to accept that results obtained at scraping velocities of 1 cm s-1 (or about 0.03 mph) are really relevant to scraping which in the field occurs at about 13.4 ms-1 (30 mph).

The apparatus developed to perform the experiments makes use of a hydraulic ram, and a three-component load cell (which can measure loads in three orthogonal directions). Relatively large blocks of concrete with a surface area of 0.031 m2 (48 square inches) are used in the testing. These have a layer of ice up to 12 mm thick (0.5 in) frozen onto their top surface, and are then mounted into the testing machine and scraped across the cutting edge, which is mounted to the load cell. The machine is described more fully in (4).

The apparatus has been used to determine the effects of various geometrical factors on the ice scraping force. One critical factor found is the blade clearance angle (Figure 2). If this clearance angle is not greater than 2_, then the cutting edge will require very high downloads to scrape ice successfully. In turn, these high downloads increase the forward scraping load because of friction between the ice and the cutting edge. Further, these high downloads reduce tire loadings, which results in lower friction and road surface traction, and thus less control of steering. Thus, a good ice cutting edge will either have a well defined clearance angle, or it will be adjustable on the truck in such a way that such a clearance angle can be effectively obtained. Conversely, the rake angle has no significant effect on the measured ice scraping forces.

The major benefit of this system is that it allows new blade designs to be tested under controlled conditions, and relatively cheaply, so that the best shapes can be determined prior to full-scale prototype testing, thus both reducing the number of such full-scale tests that are required and increasing the likelihood of successful full-scale testing. The system has been in use for a number of years now (5) and a wide range of tests have been conducted with it, including some tests with toothed and serrated blades,intended to find the optimal tooth-gap geometry.

Field Trials

Once a laboratory-based system has refined the number of possible solutions, then some sort of field trial must be made. However, field trials have a number of difficulties associated with them. First, they are expensive, since the equipment is in general larger than that used in the laboratory. Second, there are significant safety concerns. Engineering practitioners have a responsibility to the public to ensure safety, and thus there must be some circumspection when experimental studies are conducted at the field scale. Finally, the field is not a controlled environment. The test temperature cannot be set as in a lab, nor can the road surface condition, and many other critical parameters may not only be uncontrollable but may also be extremely difficult to measure in the field. Thus field tests will always tend to be somewhat more ambiguous than laboratory tests.

Realizing these difficulties, in this study an intermediate field test program was developed. In this case, a full-scale truck was equipped both with the blades to be tested and with appropriate instrumentation (6,7) to measure scraping loads. A system was developed by which a sheet of ice with relatively controlled thickness could be formed on a section of closed-off pavement (i.e., no public access was available during the test season). Once formed, and after taking visual records of the sheet and the testing temperatures, the ice sheet was scraped off, the loads recorded, and the state of the sheet post-test also noted.

This system has a number of benefits. First, it combines the controlled situation of the lab (to a degree) with the reality of the field (again, to a degree). This is critical because it allows for full-scale testing of concepts developed in the lab prior to field deployment. Second, it provides useful validating data prior to the full expense of field trials. If a cutting edge which proved promising in the lab does not bear out that promise in the controlled field test, then there is no need to include it in the complete field deployment part of the program. Third, this part of the program allows for the development of a rugged and complete data acquisition system that can then be placed on in-service trucks while the full-scale testing is ongoing. Developing a data acquisition system on an in-service truck is notoriously difficult, because the truck must go out to plow when the weather dictates. In the controlled field situation, testing is somewhat less at the mercy of the weather, and thus development is more easily accomplished.

Final Acceptance

From the field trials, a final step must be taken. This step involves actually testing the cutting edges in the field and comparing their performance. This final step should only be taken after as thorough a review as possible, because the safety of the public is then at stake. This past winter, two Iowa Department of Transportation trucks with underbody blades were instrumented, and data have been collected from which the ice scraping forces can be deduced.

While analysis of these new data has not yet been completed, there is sufficient information to indicate that the trends observed in the previous field trials are indeed observed in the actual trucks operating under storm conditions. This is very encouraging and serves to validate the field trials and in turn the laboratory work also. Whether the new cutting edges should be deployed is a decision which lies beyond the realm of a research project.

The most critical thing about this process, however, is not the end result but rather the system which has been put in place with this single project. This project was limited to cutting edge geometry, but there are numerous other issues concerned with ice removal which can now be tested using the same methodology: Going from lab to field trials to test deployment. This methodology represents a new paradigm for winter highway research and can surely be paralleled in a number of other areas, such as chemical application rates, sensor development, snow plow visibility, and so on. It is to be hoped such developments will be forthcoming.


A methodology has been described whereby the ice scraping performance of cutting edges as a function of their geometry has been measured, first in the laboratory, then in controlled (closed road) field tests, and finally in a limited field deployment. The results of this study have been very promising. However, as important as the specific results have been, perhaps equally as important has been the development of a methodology whereby promising research ideas can be taken from the lab to the field as rapidly (with due care) as possible. Such a rapid research and development methodology can be applied to a number of winter maintenance activities and, no doubt, to other areas as well.


The projects described herein have been funded by the Iowa Department of Transportation through the Iowa Highway Research Board. The trucks used were supplied by Iowa Department of Transportation. This support is gratefully acknowledged. The support of the director and staff of the Iowa Institute of Hydraulic Research has been critical to the success of this research. The U.S. Army Corps of Engineers gave their permission for the field tests to be conducted at the Coralville Reservoir, and this permission is much appreciated.

  1. R. Ettema and J.F. Kennedy. Commercial Projects at University Laboratories: Student Training? Research Adjunct? or Unfair Competition? IAHR Bulletin, Vol. 7, No. 2, 1990, pp. 13–18.
  2. W.A. Nixon. Wing Crack Models of the Brittle Compressive Failure of Ice. Cold Regions Science and Technology, Vol. 24, 1996, pp. 41–55.
  3. A. Whelan and W.A. Nixon. The Interfacial Fracture Mechanics of Ice, Final Report to the FAA, March 1996.
  4. W.A. Nixon and C.-H. Chung. Development of a New Test Apparatus to Determine Scraping Loads for Ice Removal from Pavements. Proceedings 11th IAHR Ice Symposium, Vol. 1, Banff, pp. 116–127, 1992.
  5. W.A. Nixon. Improved Cutting Edges for Ice Removal. National Research Council, SHRP Report, SHRP-H-346, 1993.
  6. W.A. Nixon. Improved Underbody Plowing. Transportation Research Circular, No. 447, July 1995, pp. 42–49.
  7. W.A. Nixon and T.R. Frisbie. Field Measurements of Plow Loads During Ice Removal Operations. Iowa Department of Transportation Project HR334. IIHR Technical Report No. 365, November 1993.

CTRE is an Iowa State University center, administered by the Institute for Transportation.

Address: 2711 S. Loop Drive, Suite 4700, Ames, IA 50010-8664

Phone: 515-294-8103
FAX: 515-294-0467


Iowa State University--Becoming the Best