The Development of a Pavement Management System at Fermilab Through the Implementation of Micro PAVER 5.1

 

Lakelle D. Pritchett, Civil and Environmental Engineering Department, Southern University A&M College, Baton Rouge, La 70813

 

 

I.  Abstract

The objective of this research is to create a Pavement Management System, consisting of empirically based pavement assessments and distress identifications, leading to a more objectified future plan for road maintenance and repair at Fermilab through integration of Micro PAVER 5.1.  In preparing to use Micro PAVER 5.1, it was necessary to create a data inventory.  This inventory’s primary focus was on pavement area, pavement material composition, and last known construction date of the roads to be tested.  After obtaining this information, network division, a system separating roads into networks, branches, and sections and furthermore into sample units, was created in order to do rankings of the roads.  With the needed inventory data collected and distress identification completed, a Pavement Management System for Fermilab is soon to be developed with the allowance of additional time. 

 

II.  Introduction

This study places emphasis on the realm of transportation as related to civil engineering.  Civil engineering encompasses several different disciplines within the one field such as environmental, geotechnical, structural, transportation, and water resources.    The most important elements of transportation- planning, street maintenance, and traffic, is in constant use and under pressure of widely fluctuating demands and intense citizen expectations.  At Fermilab, the pavement is in such use when relied upon to carry out the demands of all individuals responsible for optimizing the success of high-energy physics.  The mission of Facilities Engineering Services Section (FESS) is to establish and maintain a dependable base from which high energy physics and other Fermilab programs can be accomplished without interruption.  In keeping focus of this mission, a question of the need for the maintenance and repair of roads on the laboratory’s site was considered.  In compliance with the mission of FESS, it was concluded that the best way to produce an accurate Pavement Condition Index, PCI, and in turn a working pavement management system, would be to use the Army Corps of Engineers’ new program, MicroPaver 5.1.  Micro PAVER is a tool implemented by a pavement manager to better organize a pavement inventory.  The U.S. Army Corps of Engineers developed the Pavement Condition Index, PCI, to rate pavement distress conditions.  This index ranges from 0 to 100, with 100 being a pavement in perfect condition, with no distresses, and 0 a pavement in hazardous condition and in need of repair.  PCI is expressed in terms of a visual condition assessment, which identifies the existing distress type, severity level and quantity on a particular portion of pavement.  It can be calculated with a PCI calculating program, or by inputting the collected distress data into the Micro PAVER database.  The version 5.1 creates an accurate inventory database by observing and correcting existing pavement conditions and predicting future conditions, all while maintaining a realistic budget.  The objective of this research is to create a Pavement Management System, consisting of empirically based pavement assessments and distress identifications, leading to a more objectified future plan for road maintenance and repair at Fermilab.  A Pavement Management System, PMS, is a valuable tool used to alert a pavement manager of the critical points of a pavements life cycle  (Shahin, 2000).  Micro PAVER uses Geographical Information System (GIS) to create viewing and pavement inspection and work plan capabilities available to its users.  Fermilab roads are in dire need of such a regimen to enhance the overall process of road maintenance and repair. 

III.  Theory

Without the presence of documented data, the evaluation of future costs for repair is usually underestimated in addition to a misappropriation in the use of funds for repair.  Also, in the absence of documentation, the potential for duplication of error exists.  If documentation were accessible, a company might easily understand the extent of the procedure in considering the maintenance and rehabilitation for one road over that of another.  This idea introduces a need for a process called the ad hoc approach used when a company implements a system for maintaining and repairing pavements based on the amount.   This approach serves as a prediction for the pavement to be maintained and repaired.  The method for creating a prediction model using MicroPaver is as follows:  define the pavement family, filter the data, conduct data outlier analysis, develop the family model, and predict the pavement section condition.  This methodology was designed to predict PCI vs. time; this allows a more accurate account of the amount of time a pavement will stay in perfect condition, or conversely the amount of time left before reconstruction will need to take place.  The standard PCI scale gives a numeric association to the appearance of pavement condition.  A failed pavement has a PCI rating of 0 to 10; very poor- 11 to 25; poor- 26 to 40; fair- 41 to 55; good- 56 to 70; very good-71 to 85, and excellent pavement has a PCI rating from 86 to 100.  The custom PCI rating scale, a more generic scale, measures an unsatisfactory pavement as a PCI from 0 to 55; a degraded- 56 to 70 and an adequate pavement- 71 to 100.

IV.  Data Collection (Experimental Details)

In preparing to use Micro PAVER 5.1, it was necessary to create a data inventory.  This inventory’s primary focus was on pavement area, pavement material composition, and last known construction date of the roads to be tested.  After obtaining this information, network division, a system separating roads into networks, branches, and sections and furthermore into sample units, was created in order to do rankings of the roads.  Primary ranking assignments were given to the roads to indicate the extremity of the pavement conditions of Fermilab roads.  Distinctions between the pavements were based more specifically on an empirical assessment of the degree of distresses such as, alligator, transverse, longitudinal, and edge cracking, bleeding, and corrugation.  According to M.Y. Shahin in the text, a network is the main pavement to be analyzed.  Network definition is the process of dividing an installation’s pavements into a hierarchical order in order to facilitate inspection and maintenance planning.  In this assessment the network Id was called Fermi and the network name was Fermi roads.  Branches of a network were identified as being any of the individual streets; Inbound and Outbound Pine and Wilson Streets, Batavia, A-1, D, and B Roads.   In the MicroPaver system, a branch is given an alphanumeric name called the branch name and an alphanumeric code called the branch ID.  The original Fermilab road names were used as the branch names and the branch IDs were simply abbreviations of the branch names.  Sections, which are more precise accounts of pavement differentiation, are smaller divisions of a branch.   In dividing sections of a branch, the pavement structure, traffic, construction history, pavement rank, drainage facilities and shoulders, and condition are of interest.  However, for this experiment, more emphasis was placed on the pavement structure, rank, condition, and construction history.  Pavement structure is related to the type of materials, the thickness of that material and the overall pavement in that section.  It is necessary to ensure that the section materials and thickness be the same throughout a given section.  In the case of a large section, a sample unit was created to help better analyze pavement characteristics. When there were differences in elevation in the data collection, a new sample unit was created.  In doing the pavement inventory, it was difficult to maintain the division between data collection and distress identification.  As the data collection was carried out, the pavement’s distresses were immediately identifiable.  Samples of distress information are taken by pacing the pavement.  Pacing is a technique used in elementary level surveying in civil engineering as a way to measure distances.  By pacing, a surveyor is able to become familiar with pavement characteristics, such as distress type and pavement condition, while still measuring.  The dimensions of the pavements were determined through measurements obtained while using a measuring wheel.  This measuring wheel was capable of measuring up to 99,999 feet.  The measuring wheel was used to measure the vertical and horizontal distances of the pavement, which were then inputted into the Micro PAVER database and were used to compute the calculated pavement area.  Micro PAVER introduced a systematic random sampling that is used to help determine the sample units that should be inspected.  It was found that there is usually a 10 to 25 percent degree of sampling.  To ensure accuracy of distress quantities, it was suggested that every sample unit be surveyed.  Every sample unit was measured in this data collection to guarantee such accuracy, and the random sampling selection process was used later to determine which sample units best depicted the distress qualities of the entire pavement. 

It was typically found that the pavement on Fermilab’s site consists of asphalt with a two-inch aggregate.  Aggregate is defined as a granular material used with cement to form concrete or mortar.  It is important to remember that concrete is identified as a substance that contains cement to avoid confusion between the two.   The 18-inch subbase consisted of 12 inches of Crushed Aggregate, CA-1 and 6 inches of CA-6 (Figure I).  CA-1 is large limestone that helps to make up a significant part of the sub aggregate base layer.  The remaining component of the subbase layer was composed of CA-6.  CA-6 is simply CA-1 that has been ground into finer pieces of limestone.  This is what served as the main support for the pavement, and is what holds the pavement up from impacting force.  In considering the force the pavement withstands, it is important to contribute some form of elasticity in the tension and compression of the pavement.

Young’s Modulus refers to the ratio of stress to strain of a material measured in GPa or N/m².  Young's modulus can be used to predict the elongation or compression of an object when the stress is less than the yield strength, or applied strength, measured in pounds per square inch (psi) of the material.  Elasticity is an objects ability to return back to its original state after undergoing tension or compression.  The closer an object can get to its original state, the more elasticity an object possesses.  Fundamental concepts of the properties of elasticity in pavement maintenance are best illustrated in a process called the Freeze Thaw Cycle.  The Freeze Thaw Cycle is a condition in which, the fallen ice causes the pavement to shrink and in turn widens cracks in the pavement.  A rubber or elastic material, which not only fills the widened crack, but also prevents further widening and extends the pavement life expectancy, is used.  After the ice thaws, the asphalt is left expanded.  This problem only intensifies the severity pavement cracking.  In addition to the aforementioned effects of the cycle, the products used to rid pavement of ice play a role in the deterioration of pavement.   Products such as rock salt, potassium sodium chloride, calcium chloride, or ammonium sulfate should not be used on asphalt.  These products aid in absorption, a process in which aggregate soaks up moisture and becomes unprotected against stress and strain.  With this in mind, it is safer to use the rubber filling to help extend pavement life expectancy because it reduces the chance for a chemical change or reaction.

There are other deterioration prevention methods to maintain an operable pavement.  Chip and seal or seal coating is a surface treatment process in which a thin coat of asphalt with stone chips is used to add an extended life, approximately five-years, to the surface.  The primary purpose of chip sealing is to minimize surface water infiltration and prevent dust while serving as a surface capable of withstanding all weather conditions.  Because chip sealing has a short average life, the process can be repeated without great concern of coating build-up.  Some advantages of implementing such elements of a pavement management system are the lowered costs, and in this instance avoided traffic detours and road closures. 

A shapefile image was downloaded into the Micro PAVER.   The PAVERGIS program was used to correlate the information inventory data inputted with shapefile image. 

V.  Results

Table I.  Sample Inventory Data for the Branches of Network Fermi

 

This table shows the branches of network Fermi and the severity levels, distress types, found in particular sample units within a section.  It is possible to have only one section in a branch, which is the case for Wilson Street.   The portion of Wilson Street measured serves as a branch and also as a section.  Due to the extensive length of this section, Wilson Street had been divided into sample units to better describe the distresses of the section.  In most cases, the severity levels were measured as high severity for the distresses found.  The severity levels of the distresses found were compared to the guidelines found in the PAVER Asphalt Distress Manual.

 

 

Table II.  Weekly Average Traffic Counts

Branches	Weekly Average Traffic Counts
Outbound Batavia	1344
Inbound Batavia	1234
Wilson	772
A-1	436
D	2471
Inbound Pine	954
Outbound Pine	998

 

           

 

 

 

 

The results of the traffic counting done for one week show that the branch or road with the least amount of traffic encountered was A-1 Road.  It was surprising to find that the traffic count for both inbound and outbound Pine Street was low in comparison to that of other roads that do not contain entrance gates.  D Road, located Northeast of Wilson Hall was found as having the most traffic in a given week.  These studies were made possible through the use of the DuPage County Department of Transportation’s traffic counting devices.

 

 

 

 

 

 

 

 

Figure 1.1 Typical depiction of subbase layer of Pine Street

 

VI.  Discussions  

The PAVER Asphalt Pavement Distress Manual gives the guidelines for measuring the severity levels and identifying the types of distresses found.  The following was found to be true based on these guidelines.

When the mixture of tar and sand are made in the seal coating, the tar may begin to exude or “bleed” out of pavement to the surface.  Tar is a bituminous cementing or binding material obtained as a residue from the destructive distillation of coal.  This type of asphalt distress is classified as bleeding.  As evident on Road B, bleeding was due to the effects that heat had on a pavement.  To help minimize the amount of tar seeping through the pavement and to serve as a cooling aid, sand was sometimes placed over the seal-coated area.  Bleeding is typically found on roads that have been recently constructed.  Although older roads are less likely to exhibit high severity levels of bleeding, these older roads are combated with problems of deterioration due to road usage.  It was found that the problem of bleeding could be slightly ameliorated by the presence of trees, which provide shading for the pavement.  The presence and abundance of a viscous material found on the surface of the asphalt measures the severity levels of bleeding of the pavement.  For high levels of bleeding, the bituminous material on the pavement sticks to shoes and tires on several weeks of a year given a high temperature.  Medium and low severity is evident with there is a stickiness on only a few weeks of a year and only a few days of a year, respectively.

Blocking was found on A-1 Road.  Blocking occurs when cracks are found in pavement, which have interconnected and created what looks like square or rectangular blocks.  The severity levels of the cracks are determined by measuring the area of the blocks.  While blocking is not load associated, it does indicate a hardening of asphalt.  Block cracking occurs over an entire area not subjected to loading.

Alligator cracking was a difficult distress to observe in that in cases of medium to high severity it was comparable to highly severe edge cracking, which was the extremely evident on Inbound Pine Street near Kirk Road.  Another difficulty experienced in the distress identification process was the coexistence of differing severity levels within one given sample unit.  Alligator or fatigue cracking is a series of interconnecting cracks caused by fatigue failure of the asphalt concrete surface under repeated traffic loading. Cracking begins at the bottom of the asphalt surface, in this case, an aggregate base where tensile stress and strain are highest under a wheel load. The cracks surface initially as a series of parallel longitudinal cracks. After repeated traffic loading, the cracks connect, forming many sided, sharp-angled pieces that develop a pattern resembling alligator skin.  Alligator cracking occurs only in areas subjected to repeated traffic loading, such as wheel paths. 

Corrugation occurs when there is a “rippling” in the pavement and is measured by ride quality.  Corrugation, also known as “washboarding,” is a series of closely spaced ridges and valleys (ripples) occurring at fairly regular intervals, usually less than 10 feet along the pavement. These ridges or ripples are perpendicular to the direction in which traffic travels. This type of distress is usually caused by mobile traffic combined with an unstable pavement surface or base.  High severity corrugation is easily felt when riding in a vehicle such as a Cushman.  In doing the ride quality assessment, it was suggested that a sedan or a car typically used in local traffic.  This sedan also was to be driven at the posted speed, which at Fermilab is 40 miles per hour (mph).  The Cushman used had a maximum speed of 8 mph.  Although a sedan was suggested it was found that for a more accurate pavement analysis a Cushman was perfect for detecting the pavement distresses.

Potholes were found within the longitudinal cracking on Outbound Pine Street.  In most instances the Outbound Pine Street branch, the longitudinal cracks also contained highly severe alligator cracking, which created the potholing.  Potholes are usually small, less than 30 inches in diameter and have concaved depressions in the pavement surface. They generally have sharp edges and vertical sides near the surface of the hole. When high-severity alligator cracking creates holes, they should be identified as potholes, not as weathering.  Weathering and raveling are the wearing away of the pavement surface due to a loss of asphalt or tar binder and dislodged aggregate particles. These distresses indicate that either the asphalt binder has hardened appreciably or that a poor-quality mixture is present. In addition, raveling may be caused by certain types of traffic, e.g., tracked vehicles. Softening of the surface and dislodging of the aggregates due to oil spillage are also included under raveling.  These two distresses, weathering and raveling, were not commonly found in the branches of the Fermi network.  Raveling however was found in Sample Unit 5, Section 2, on Inbound Pine.   This distress must be carefully observed, and not mistaken simply as high severity edge cracking.  Raveling is differentiated in that the pieces of pavement have been broken up and can be pulled up from the rest of the pavement.

Edge cracks are parallel to the outer edge of the pavement and are usually within 1 to 1.5 feet of this outer edge. Edge cracking was found on several branches in the Fermi network.  For example, Wilson Street, B Road, D Road, and Inbound Pine Street all exhibited high severity edge cracking.  This distress is accelerated by traffic loading and can be caused by frost-weakened base or subgrade near the edge of the pavement.  This is comprehensible in considering the conditions under which the pavement withstands in the falls and winters in Illinois.  The area between the crack and pavement edge is classified as raveled if it is broken up in some cases to the extent that pieces are removed.

      Longitudinal cracks are parallel to the pavement's centerline direction.  This is the mother distress, in that it leads to the development of several other distresses, such as blocking, alligator cracking and potholing.  On the seven branches of the Fermi network, there existed some high and medium severity longitudinal cracking on all of them.  Longitudinal cracking may be caused by poorly constructed paving lane joint, shrinkage of the Asphalt Concrete (AC) surface due to low temperatures or hardening of the asphalt and/or daily temperature cycling, or reflective cracks caused by cracking beneath the surface course, including cracks in the case of Portland Cement Concrete (PCC) slabs.  Most pavements will experience this type of distress.  Coupled with longitudinal cracking, in the PAVER distress manual is transverse cracking.  Transverse cracks extend across the pavement at approximately right angles to the pavement centerline and are not usually load-associated.

                  To better understand the causes of the distresses, a minor integration of traffic counters was incorporated in the analysis of the data obtained.  Traffic counters were provided through the DuPage County Department of Transportation and were installed on several branches of the Fermi network, and other roads on site.  These counters were installed to provide a measure of the amount of traffic on the selected roads on site.  In some cases traffic count is measured by Vehicle Magnetic Imaging (VMI), which measures a motor vehicles’ magnetic influence passing through the earth’s magnetic field.  Such counters were not used in this data collection, rubber tubing was used to do traffic counting by volume.  The traffic counters used consisted of a rubber tube about 50 feet long, which was adjustable to fit any desired length, and a metal box that contained the actual counter.  In some instances, the tube was short and needed to cover only one lane and a separate short tube was used to cover the other side with opposing traffic.  A nail was hammered into a metal nut attached to the rubber tube by a clamp.  After the tube was nailed in place, duct tape was used to reinforce the rubber tube to the pavement.  The device was capable of giving a traffic count for bidirectional traffic, multi-lane traffic, and numerous other features.  The traffic counters were installed during the rain and are capable of performing traffic counts in poor weather conditions.  The traffic counters were placed at several locations that were believed to be heavy populated with traffic.  This experimentation lasted for one week.

VII.  Conclusion

With the needed inventory data collected and distress identification completed, a Pavement Management System for Fermilab is soon to be developed with the allowance of additional time.  The following are recommendations for doing similarly related data collections: Contact persons affiliated with a newly implemented program before use and establish and maintain a relationship with programmer and others experienced in working in the related field throughout research.  In this data collection, a Palm Vx Palm Pilot was used while collecting data in the field.  Such an instrument is suggested when doing field measurements to help keep collected data organized.

VIII.  References

Dowling, Norman E.  Mechanical Behavior of Materials.  1999

 

Shahin, M.Y.  Pavement Management for Airports, Roads, and Parking Lots.  1994.

 

http://www.lecol.com/HTML/catalog/NuMetrics/HiStar.htm 

 

http://www.cecar.army.mil/paver/History.htm