Time Series and Machine Learning Approach for Forecasting the Demand for Small Equipment, Tools, and Consumables for Industrial Construction Projects

Abstract

The high consumption and utilization of demand for small equipment, tools, and consumables in construction projects underscores the necessity for effective procurement strategies. Accurate estimation of these consumables is crucial for moving toward project completion in a timely manner. With recent advancements in time series analysis, artificial intelligence, and machine learning, these technologies can be employed to formulate predictive models. This research aims to explore the advantages of using time series and machine learning—in combination with historical data from past projects—to identify key factors that impact demand for these consumables, as well as develop an efficient predictive model that analyzes and learns from historical data thereby facilitating precise estimations for future projects. The research involves collecting and analyzing historical data, analyzing current industry practices for estimating requirements for small equipment, tools, and consumables, and implementing time series analysis and machine learning algorithms to forecast demand for various types of consumables in construction projects. This study investigates crucial factors that influence these items, bridging the gap between literature review and industry practices. Finally, this research proposes time series and machine learning models capable of predicting quantities in industrial projects using historical data. The proposed models provide an estimation of monthly requirements for various types of consumables throughout the project, which assists project managers in estimating required quantities, offering them accurate insights to help facilitate effective procurement strategies.