Assessment of the expected construction company's annual work volume using neural network and multiple regression models

Mohamad H.H.¹,  Ibrahim A. H.²  And  Massoud H.H.³

¹Associate Prof., Construction Engineering Dept., Faculty of Engineering, Zagazig University, Egypt .

²Assistant Prof., Construction Engineering Dept., Faculty of Engineering, Zagazig University, Egypt .

³Ph.D. Student, Construction Engineering Dept., Faculty of Engineering, Zagazig University, Egypt .

Corresponding author Email: hazem_mm1969@yahoo.com

Accepted 14 June 2012

Abstract

Annual work volume of any construction company can be considered as an important indicator for the company's financial performance. Business success heavily depends on the ability of financial executives to maximize the company's net profit and annual work volume. Consequently, the firm financial managers should continuously strive to maximize their company's annual work volume. Modelling the company's annual work volume can help financial management to investigate the serious effect that the different financial conditions can have on the expected annual work volume of their companies. Stated differently, financial managers can make sure that business operations of their companies are running in a successful manner. For example, inadequate working capital may interrupt the normal operations of the business which impairs the company's annual work volume and consequently its profitability. To elaborate more, excessive levels of current assets may have a negative effect on firm's work volume and profitability whereas a low level of current assets may lead to lower level of liquidity and stock outs which results in difficulties in maintaining smooth operations that leads to a corresponding decline in the annual work volume.  The objective of this research is to develop a mathematical model for the assessment of the expected construction companies' annual work volume. First, the main factors affecting firms' annual work volume were identified based on a comprehensive literature review. Next, pertinent data regarding these factors were collected. Such data are mainly concerned with the companies' financial statements as well as the economic environment. Then, two different annual work volume models were developed using the Multiple Regression (MR) and the Neural Network (NN) techniques. The validity of the proposed models was also investigated. Finally, the results of both MR and NN models were compared to investigate the predictive capabilities of the two models.  

Keywords: Construction Company's Annual Work Volume, Neural Network, Multiple Regressions.