Expert System-Based Exploratory Approach to Cost Modeling of Reinforced Concrete Office Building

Full Text PDF PDF
Author(s) Amusan Lekan Murtala | Joshua Opeyemi | Owolabi James D | Tunji-Olayeni Patricia F | Anosike Michael N
Pages 86-93
Volume 3
Issue 1
Date January, 2013
Keywords Expert system, Model, Colinearity, Validation, Office building.

Abstract

Expert system is a conventional method that is in use in cost modeling, considering its advantage over traditional regression method. It is based on this fact, that this study aimed at deploying neural network in cost modeling of reinforced concrete office building. One hundred (100) samples were selected at random and divided into two; one part was used to develop network algorithm while the second part was used for model validation. Neural network was used to generate the model algorithm; the model is divided into 3 modules: the data optimization module, criteria selection with initializing and terminating modules. Regression analysis was carried out and model validated with Jackknife re-sampling technique. The colinearity analysis indicates high level of tolerance and -0.07403 lowest variation prediction quotients to 0.66639 highest variation quotients. Also the Regression coefficient (R-square) value for determining the model fitness is 0.034 with standard error of 0.048 this attest to the fitness of the model generated. The model is flexible in accommodating new data and variables, thus, it allows for regular updating.

< Back to January Issue