Comp. Meth. Civil Eng., Vol. 2 No. 1 (2011) pp. 95-107 ©Copyright by the University of Guilan, Printed in I.R. Iran

CMCE

Computational Methods in Civil Engineering

Prediction of concrete mix ratios using modified regression theory
D.O. Onwukaa,*, C.E. Okerea, J.I. Arimanwaa, S.U. Onwukab
Civil Engineering, Federal University of Technology, P.M.B. 1526, Owerri, Nigeria. Project Management Technology, Federal University of Technology, P.M.B. 1526, Owerri, Nigeria.
Received 30 March 2011; accepted 17 May 2011 a b

Abstract The strength of concrete is a function of the proportions of the constituent materials, namely, cement, water, fine and coarse aggregates. The conventional methods used to determine the mix proportions that will yield the desired strength, are laborious, time consuming and expensive. In this paper, a mathematical method based on modified regression theory is formulated for the prediction of concrete strength. The model can prescribe all the mixes that will produce a desired strength of concrete. It can also predict the strength of concrete if the mix proportions are specified. The adequacy of the mathematical model is tested using statistical tools.
Keywords: Concrete; Regression theory; Prediction; Mix ratios.

1. Introduction Concrete is a composite construction material consisting of water, cement, fine aggregate (sand) and coarse aggregate. Mix design of concrete is a means of producing the most economical and durable concrete that meet with certain properties as consistency, strength and durability by properly and systematically combining the ingredients (concrete materials) of relative proportions [1]. The strength of concrete is very important because most of the desirable characteristic properties of concrete are qualitatively related to its compressive strength [2]. Various methods have been used to study and/or determine the strength of