Abstract

The construction market is a complex system that is affected by a variety of factors, including macroeconomic indicators, such as inflation, interest rates, and economic growth. For the construction market stakeholders, accurately predicting future construction costs is essential for maintaining financial stability and setting appropriate pricing strategies. Over the past decade, global economic disruptions have significantly impacted national economies, and Egypt is no exception. The Egyptian construction market has experienced record volatility in materials costs and inflation. This has made it challenging for project owners, contractors, as well as cost consultants in the construction market to predict the cost of their projects. Many researchers addressed the issue of rising inaccurate cost prediction by developing different tools to capture the correlation between economic indicators and construction costs, however, transmission of price inflation of construction materials in a volatile economy was something often overseen by researchers. The goal of this research is to investigate the dynamics of inflation transmission between construction materials and utilize the findings to develop and test a data-driven machine learning model for forecasting future costs of key materials. This research is primarily addressing the Egyptian market due to its steadily growth in recent years, driven by the rapid population in the country and the increasing urbanization adopted by country developers and supported by the government of Egypt and the challenges the Egyptian construction sector is suffering from due to the economic challenges and unprecedented price fluctuations. Monthly prices of 16 construction materials in Egypt were collected over the period from 2009 to Q1 of 2025. Employing a robust methodology, the research utilized a Vector Auto regression model (VAR) to model interrelations between different materials and further understand the correlations through a Granger Causality Test. Results of the model were interpreted through a network analysis to describe the specific characteristics of the construction materials in relation to how their prices influence or are influenced by inflation within the network of analyzed materials using centrality measures to quantify materials susceptibility (In- Degree Centrality), transmission capacity (Out-Degree Centrality), and Intermediary role (Betweenness Centrality). Modularity-based clustering further categorized materials based on their price interconnections, revealing inflation transmission pathways across material sectors. Finally, a neural network model, implemented in Python, was developed to predict future material prices by leveraging the identified relationships and inflation trends. This research offers a prominent practical tool for construction stakeholders assessing the predicting materials prices and understanding the dynamics of internal correlation between key materials in the construction industry affecting the overall project budget and performance to enhance price estimations and prepare for increasing costs.

School

School of Sciences and Engineering

Department

Construction Engineering Department

Degree Name

MS in Construction Engineering

Graduation Date

Winter 1-31-2025

Submission Date

9-18-2025

First Advisor

Ibrahim Abotaleb

Committee Member 1

Khaled Nassar

Committee Member 2

Dina Saad

Committee Member 3

May Haggag

Extent

112 p.

Document Type

Master's Thesis

Institutional Review Board (IRB) Approval

Approval has been obtained for this item

Disclosure of AI Use

Code/algorithm generation and/or validation

Download

Available for download on Saturday, September 18, 2027

Share

COinS