LDA-Based Model for Measuring Impact of Change Orders in Apartment Projects and Its Application for Prerisk Assessment and Postevaluation

Abstract

This study analyzed the impact of change orders on construction projects by using the loss distribution approach (LDA), which is used in finance to analyze operational risks, and proposed a framework for applying the LDA to construction. It examined 9,028 change orders for 237 apartment housing projects in South Korea. Prior to the analysis, the change-order factors were classified into five categories and the work types into eight categories in order to form a risk matrix. The severity and frequency distributions of each cell were calculated based on the risk matrix. Because the amount of loss can vary depending on the construction cost, severity was defined as the amount of loss relative to the construction cost. Using this, the authors calculated the loss distribution of each cell and combined them to derive the total loss distribution. The analysis results confirmed that field conditions had the largest overall impact on change orders. This was ascribed to the limitations of estimating and reflecting various situations that can occur at construction sites during the initial design stage. A high loss of change orders was also confirmed to occur in the finishing work and mechanical, electrical, and plumbing (MEP) work. Because the finishing work consists of various work types, multiple subcontractors proceed with their tasks simultaneously, which makes coherent coordination relatively difficult and increases the frequency of changes to the design. For MEP work, the frequency of design changes is high because various interference phenomena may occur depending on the complex pipelines and wiring diagrams. This paper proposes an LDA comprising a prerisk assessment and postevaluation and feedback method. The preresponse is measured as a cost contingency calculation tool. In other words, response strategies are proposed against expected and unexpected losses at three levels after the total loss distribution has been derived by aggregating various risks: namely, cost contingencies at the project level, equity capital at the company level, and insurance at the catastrophic level. The proposed postevaluation and feedback method evaluates the target project after its completion by identifying points of cost overrun in the cumulative distribution constructed in advance with the LDA and by analyzing detailed items for each risk factor in the same manner. Thus, this method can be used to verify the ability of a subcontractor.