Learning curve modelling of gas pipeline construction in Egypt

Abstract

It is commonly accepted that production rates for repetitive tasks will improve with acquired experience and learning. Several mathematical models, or learning curves, have been proposed to investigate improvement in productivity as a function of the number of units produced. Deciding the best-fit learning curve model for construction activities is a managerial challenge. In this paper, the best-fit learning curve model for describing past performance of gas pipeline construction in Egypt is investigated. Data were collected from eight real-life projects, which are constructed in different types of land, under various weather conditions with different sizes, lengths, and pipe diameters. Only labour-intensive activities are considered in the present work. Cumulative average data is used to represent collected data in order to ensure curve smoothness as well as to avoid scattered data. The commercial Statistical Package for Social Science (SPSS) is used to determine Pearson's coefficient of correlation. The cubic model is found to be the best fitting curve to describe welding activities in gas pipeline construction activities in Egypt. This work helps both academics and practitioners in deciding the best-fit learning curve model(s) for gas pipeline construction. This will aid in making realistic estimates of the time and cost of repetitive projects.