Search results

This study aims to establish a new system to predict the defect liability phase (DLP) cost using the Six Sigma methodology, which investigates sources of variations and reduces the error level to 3.4 per million through five phases: define, measure, analyze, design and verify.

After the initial handover of the construction project, the DLP follows the practical completion. During this stage, the contractor is responsible for the remedy of any defects that appeared in the project. Many researchers have studied defect reasons and their associated costs in different industries, while the construction industry remains a green field for this kind of research. The objective of this study was to develop a model to predict the DLP cost. The research methodology adopted the five stages of the Six Sigma cycle: defining objectives, measuring the data, analyzing performance, designing the model and verifying the results. Twenty factors were identified as potential factors affecting the DLP cost. Factors were categorized into two main clusters: project data and organization data. Interviews were conducted with 42 project management experts, who have 8–35 years of experience in construction project management, to rank the 20 factors based on their importance. Simo’s procedure was used to obtain the weight of each factor affecting the DLP cost based on the opinions of the experts. The Pareto principle was used to select the “Vital Few” factors affecting the DLP cost, and six factors were selected. The design of experiments (DOE) was used to establish a dynamic model to predict the DLP cost using a sample of 41 construction projects obtained from the above-mentioned 42 project management experts. The model accuracy was verified using data obtained from a different sample of five construction projects, which were not used to establish the model.

The results showed that among the 20 factors, only six were found to have a cumulative impact of 50% over the cost of the DLP: type of project, project contract value, nationality of the employer, project manager experience, DLP duration and sector of the employer. A model was established through the DOE to predict the DLP cost using the values of the aforementioned factors.

As a natural limitation of using DOE, the newly developed model can be applied to predict the DLP cost based on data within the range of data used during the model development, which means that model is confined within the specific measured values of factors. Furthermore, it will be beneficial for future studies to study the impact of other factors related to the types of materials or equipment used in building the project because it was not considered during this study because of the huge diversities in these factors and difficulties in determining its impact on the DLP cost.

The unique results of using DOE through Minitab software facilitated obtaining of a dynamic model, which means that researchers can modify any value of the six factors and monitor instantly the expected change in the DLP cost, which will allow a better understanding of the impact of each factor on the DLP cost. Moreover, the new model will help contractors to predict the expected DLP cost to be added for their project budget, which will mitigate the risk of cost overrun resulted from the cost of defect rectification.

A dynamic model was established to predict the DLP cost using the DOE. The new model was validated, and the prediction error ranged from −18% to +21%.