Incorporation of Reliability Into Nigerian Empirical Mechanistic Pavement Analysis and Design System (Nempads)

This work was aimed at providing the most appropriate Nigerian Empirical-Mechanistic Flexible Pavement Analysis and Design method (NEMPAD) by incorporating Reliability, using First order reliability method (FORM), considering all the input variabilities, uncertainties, and seasonal variations, with the aid of Matlab to express the algorithms. A frame work was developed for computation of component reliability index (R.I), system reliability index (R.I), and probability of failure. Inputs parameters considered includes: traffic load, material properties, and environmental conditions. Four seasons were considered in the design namely: Hunturu(1 week Dec-2 week Feb), Bazara(2 week Feb-2 week May), Damina(2 week may-1 week Oct), and Kaka(1 week Oct-1 week Dec) seasons, for the NorthWest geographical zone, and the cumulated damage was computed for each season. Comparism was made between deterministic and reliability methods. The results indicates that at coefficient of variation (COV) of 25% , a surfacing thickness of 100mm (representing wearing and binder course), 165mm thickness of Base course material, and 180mm thickness of sub base course material, are adequate, while at coefficient of variation (COV) of 42% a surfacing thickness of 100mm (representing wearing and binder course), 200mm thickness of Base course material, and 250mm thickness of sub base course material, are adequate, as against  the deterministic  method cited in Olowosulu(2005),where higher thicknesses were used, under the same traffic load and conditions. The results also indicated that highest level of damage with respect to fatigue was recorded during Bazara season due to the effects of Temperature, while highest level of damage with respect to Rutting was recorded during Damina season due to the effects of water in the subgrade.

INTRODUCTION

Traditionally, flexible pavement thickness design has been accomplished throughempirically based procedures. One well-known procedure, the American Association ofState Highway and Transportation Officials (AASHTO,1993) method, was based upon theAASHTO Road Test held in Illinois between 1958 and 1960. The designprocedure utilized empirical relationships developed from the AASHTO road test and istherefore limited to the conditions of that test. In fact, all empirically based methodsshare the same common disadvantage in that they are limited to the conditions andobservations of the particular road sections on which the procedure was based. This factmay require engineers to extrapolate outside the original inference space, which could beproblematic.As a result of this limitation, many pavement sections failed prematurely while other sections far outlived their design lives(Rutherford, 2012)

Conversely, Empirical-Mechanistic (M-E) design is more robust since it combinesthe elements of mechanical modeling and performance observations in determining therequired pavement thickness for a set of design conditions. The Empirical-Mechanistic Empirical-Mechanisticbased method of pavement design is based on the mechanics of materials,which relates inputs such as wheel loadsto output such as pavement response.The response is then used to predict pavement distresses(including cracking)and other performance based laboratories and field testing outcomes.In essence, M-E design has the capability of changing andadapting to new developments in pavement design by relying primarily on the mechanicsof material (Timmet al, 1999).