REVISED BETA CRITERIA FOR THE CBR AIRFIELD PAVEMENT DESIGN METHOD

ABSTRACT

The empirical California Bearing Ratio (CBR) design method for flexible airfield pavements has been a mainstay for practicing engineers for decades.  Its simplicity and proven performance makes it easy to use and reliable.  However, design scenarios for increasingly larger aircraft have revealed a significant design shortfall at these extremely heavy and multi-wheeled loading conditions.  This has driven a redevelopment of the CBR method to underpin it with a mechanistic load response of vertical stress at the subgrade surface instead of an equivalent single wheel load at the surface of the pavement.  This response is correlated to an empirical failure model of pavement life developed from full-scale test data.  This new design method has eliminated limitations in the original procedure that resulted in overdesigned pavements for large aircraft.  However, any new method always leaves room for improvement.  This thesis explored the development of the current failure model and identified an opportunity to refine it.  The primary target of this modification was the concentration factor, an empirical modification to the Boussinesq stress distribution theory used in calculating the load response.  The test data used to develop the current failure model was previously analyzed with a constant concentration factor for all test points.  However, a more representative relationship of the concentration factor as a function of the subgrade CBR was available to better model soil behavior.  After collecting additional test data beyond that used to develop the current failure model, a new failure model was developed that included the concentration factor as a function of CBR.  In addition, it was shown, through comparison of nonlinear regression statistics, that a failure model developed using the concentration factor as a function of CBR provided failure model with a better fit to the

test data than a model developed with a constant concentration factor.

 

                                                     TABLE OF CONTENTS

List of Tables ………………………………………………………………………………………………………………… vi!

List of Figures ………………………………………………………………………………………………………………. vii! List of Abbreviations and Acronyms ………………………………………………………………………………… ix!

1. Introduction ………………………………………………………………………………………………………………… 1!
   • Summary ………………………………………………………………………………………………………. 1! 2. Research Goals ………………………………………………………………………………………………. 1! 1.3. Relevance ……………………………………………………………………………………………………… 2!

1.4. Scope ……………………………………………………………………………………………………………. 4!

2. Prior Research/Literature Review ………………………………………………………………………………….. 5!
   • Breadth of Existing Research …………………………………………………………………………… 5!
   • Background Concepts …………………………………………………………………………………….. 5!
     • Coverages ……………………………………………………………………………………….. 6!
     • California Bearing Ratio (CBR) …………………………………………………………. 8! 2.3. Equivalent Single Wheel Load ……………………………………………………………. 9!
   • The Classic CBR Design Method …………………………………………………………………… 11!
     • Origins ………………………………………………………………………………………….. 11!
     • Definition ………………………………………………………………………………………. 13! 3.3. Shortfalls ……………………………………………………………………………………….. 15!
   • Beta Criteria; the Revised CBR Design Method ………………………………………………. 16!
     • Definition ………………………………………………………………………………………. 16!
     • Development of the Beta Criteria Model ……………………………………………. 20! 4.3. The n = f(CBR) Model …………………………………………………………………….. 23!

3. Research Approach and Methods ………………………………………………………………………………… 26!
   • Data Collection ……………………………………………………………………………………………. 26! 2. Stress and Beta Calculations ………………………………………………………………………….. 27! 3.3. Beta Criteria Curve Fitting …………………………………………………………………………….. 29!

3.4. Model Testing and Verification ……………………………………………………………………… 31!

• Statistical Analysis ………………………………………………………………………….. 32!
• Design Thicknesses …………………………………………………………………………. 33! 4.3. Test Data ……………………………………………………………………………………….. 34!

3.5. Selection of Model ……………………………………………………………………………………….. 34!

4. Results and Analysis ………………………………………………………………………………………………….. 35!
   • Data Calculation …………………………………………………………………………………………… 35! 2. Stress and Beta Calculations ………………………………………………………………………….. 38!

4.3. Curve Fits ……………………………………………………………………………………………………. 42!

• Resulting Curves …………………………………………………………………………….. 42! 3.2. Concentration Factor Effects ……………………………………………………………. 47!

4.4. Model Testing and Verification ……………………………………………………………………… 50!

• Statistical Analysis ………………………………………………………………………….. 50!
• Design Thicknesses …………………………………………………………………………. 51! 4.3. Test Data ……………………………………………………………………………………….. 60!

4.5. Model Selection …………………………………………………………………………………………… 65!

5. Summary, Findings, Conclusions, and Recommendations ………………………………………………. 67!
   • Summary …………………………………………………………………………………………………….. 67! 2. Findings ……………………………………………………………………………………………………… 67! 5.3. Conclusions …………………………………………………………………………………………………. 68!

5.4. Recommendations ………………………………………………………………………………………… 68!

References ……………………………………………………………………………………………………………………. 70!

Appendix: Test Data ……………………………………………………………………………………………………… 72!

1. Introduction

1.1. Summary

The California Bearing Ratio (CBR) flexible airfield pavement design procedure is an empirical design method that has been in use since the 1940s.  The empirical design input of the subgrade CBR results in a simple design procedure requiring few resources to implement.  Due to recently identified shortcomings caused by the introduction of very large aircraft, the US Army Corps of Engineers (USACE) Engineer Research and Development Center (ERDC) recently redeveloped the method to account for these shortfalls, resulting in a mechanisticempirical failure model that retains the simplicity of the original CBR input for the design process.  The mechanistic feature of the failure model is the vertical stress at the top of the subgrade induced by the aircraft landing gear.  This stress is a function of the loading conditions and the properties of the material through which the stress is distributed, quantified by the concentration factor.   This factor is a function of the soil type and is taken into consideration in the current CBR design process.  However, it was only partially accounted for in development of the failure model itself.  This research will further describe the development of the CBR design process, the theory behind the redeveloped method, and then redevelop the current failure model with full inclusion of the concentration factor.

1.2. Research Goals

The goal of this research is twofold.  First, the research objective is to develop new beta criteria for the CBR airfield pavement design procedure by including the concentration factor as a function of subgrade CBR in the test data points used in development of the criteria curve.

Second, the research hypothesis is that this new curve will provide a better fit to the test data, and will subsequently allow better confidence in pavement thickness design.

1.3. Relevance

In the modern era of computational ease enabled by computers, coupled with the everdeepening understanding of, and capabilities to characterize, geotechnical materials, the use of an empirical design procedure such as the CBR method may seem obsolete.  However, the use of such a method is actually considered valuable by many in the airfield pavement field, primarily in the Department of Defense (DoD).  This is due to the simplicity in application of the design procedure as well as its ability to easily accommodate a worldwide myriad of geographical locations, environmental conditions, and construction materials.  Also, the International Civil Aviation Organization (ICAO) has selected the CBR procedure for use in the Aircraft Classification Numbers (ACN) and Pavement Classification Numbers (PCN) system.  This system is used to classify the relative strength of airport pavements as well as the relative damage imparted by an aircraft in order to allow pilots and airport managers to quickly and easily determine aircraft operability.

The US military is broadly recognized as the world’s preeminent fighting force.  Although this can be attributed to numerous factors such as shear size, technological mastery, and superb training, one of the most significant enabling factors is air power, provided by all service branches.  The US military is alone in the ability to transmit mass amounts of military strength, be it combat operations or disaster and humanitarian assistance, anywhere in the world, at anytime, a concept known as global reach.  This global reach is made possible through a network of air bases, aerial refueling, and aircraft carriers.  The backbone of this network is the presence of existing airbases throughout the United States and allied nations around the globe.

However, at times, new airfields are required where none exist, and are often needed quickly.  These locations can be austere in terms of environment and material availability, or in terms of logistical support such as electrical supply, computer access, and material testing capabilities.  These factors lend perfectly to the use of the empirical CBR design method.  From the structural designer’s perspective, all that is needed is one piece of testing equipment, the dynamic cone penetrometer (DCP), and a design chart for the design aircraft.  All other design aspects, primarily materials and construction methods, are considered in specifications that have been adjusted from six decades of experience throughout the entire world.  Putting the simple structural CBR design method together with the specifications provides a method that has proven capable of quickly, and with minimal resources, handling tropical environments, arctic conditions, and arid deserts.

Despite this aptness to urgent and austere design scenarios, the CBR method is perfectly suited to permanent air base design needs.  Since the failure mode of pavements is rarely catastrophic, the primary two concerns in pavement design are minimal (life-cycle) cost and achieving the desired pavement life, functionally and structurally.  As such, good design models will neither over-design, causing excess up-front costs, nor under-design, causing early failure and excessive maintenance, repair, or rehabilitation costs.  Due to certain assumptions made in the current empirical CBR failure model, it may be possible to produce a revised model that more closely fits the available test data and a more reliable design thickness across the full spectrum of the model’s potential applications.

1.4. Scope

The scope of this research was an analytical redevelopment of the CBR design process subgrade failure model.  Due to the cost and complexity of full scale pavement testing, no actual testing was done in this research to develop the model.  Instead, historic test data was collected to do so.  The general steps required to complete this research are as follows:

1. Search for existing flexible airfield pavement test reports.
2. Evaluate and consolidate available test data for adequacy to this research.
3. Process the test data via appropriate methods to produce failure model data points.
4. Fit several general forms of potential failure models to the data points.

Evaluate the produced models using various model testing and verification methods.

REVISED BETA CRITERIA FOR THE CBR AIRFIELD PAVEMENT DESIGN METHOD

Sharing is caring!