# Testing Regression Models to Estimate Costs of Road Construction Projects

No Thumbnail Available

## Date

2014-11

## Authors

## Journal Title

## Journal ISSN

## Volume Title

## Publisher

Addis Ababa University

## Abstract

At the outset of the project, when the scope definitions are in the early stages of development, little
information was available, yet there is often a need for some assessment of the potential cost. The
owner needs to have a rough or approximate value for the project’s cost for purposes of determining
the economic desirability of proceeding with design and construction. Special quick techniques are
usually employed, utilizing minimal available information at this point to prepare a conceptual
estimate. Little effort is expended to prepare this type of estimate, which often utilizes only a single
project parameter, such as square meter of floor area, or span length of a bridge. Using available,
historical cost information and applying like parameters, a quick and simple estimate can be
prepared.
The objective of this study is to develop conceptual and preliminary cost estimating models for
asphalt road construction projects using historic data using statistical tools such as spss, and Rsoftware’s,
based on sixteen sets of data collected in the Federal Road Projects. As the cost estimates
are required at early stages of a project, considerations were given to the fact that the input data for
ANNOVA F-test regression analysis to develop the cost models could be easily extracted from
sketches or scope definition of the project. As a result in this study Six regression cost estimating
models are developed to estimate the total cost of road construction project; among these models two
include bid quantities, and four include project size ( i.e. road length and road width) as input
variables. The coefficient of determination (r2) for the developed models is ranging from 0.65 to 0.98
which indicate that the predicted values from a forecast models fit with the real-life data. The values
of the mean absolute percentage error (MAPE) of the developed regression models are ranging from
±16.3% for preliminary cost estimating and to ±38.9% for conceptual or ball park method of cost
estimation, the results compare favorably with past researches which have shown that the estimate
accuracy in the early stages of a project is between ±25% for preliminary method of cost estimating
that can be related to specific characteristics of known sections or areas of the project and ±50% for
conceptual method of cost estimating where early informed guesses made when virtually no
drawings exist.
The research finding shows how regression models based on the significant variables or bid
quantities can be used to develop regression models as tools in forecasting future road construction
cost that carry much greater reliability than the previous estimated value. The paper introduces the
development of cost estimating techniques and principles from historic data in the archives from both
a client and consultants viewpoint both in the early stage of pre-tendering or the planning phase and
project-level
Keywords: Cost estimating, Regression Model, Early cost estimate

## Description

## Keywords

Cost estimating, Regression Model, Early cost estimate