A Model for Identifying the Most Effective Contract Type in Construction Companies

Abstract

Construction managers that are conversant with project management knowledge areas handle the construction project. Intelligent means discerning or looking about, and these construction managers are well-versed in their fields. However, the majority of construction projects in construction companies fail around the world, particularly in low-income countries such as Ethiopia, due to a lack of construction project managers who are unfamiliar with contract management that are used without considering the company's situations and project contexts. The goal of this research is to create a machine-learning-based prediction model for finding the most effective contract type in construction companies. Data from Ethiopian construction enterprises and associations (particularly from the Amhara National Regional State Job & Training Bureau and its branches in Addis Zemen, Woreta, and Debre Tabor) were collected using a purposeful sample strategy. Support vector machines scored 93 %, decision trees scored 92.30 %, and K-nearest neighbors scored 91.20 %, and logistic regression scored 88 % on the testing data set, respectively. Furthermore, in comparison to the other classifiers, Support vector machines outperformed the best in the experiments. The study suggests gathering more building project datasets from construction businesses and comparing them to our findings to determine the most effective contract type.