Applying a Loss Distribution Approach to Evaluate Maintenance Cost of Housing

Abstract

The maintenance of deteriorating buildings is becoming an important issue worldwide. Adequate maintenance not only prevents futher deterioration but also improves the building’s performance and supports the safety of its occupants. In South Korea, the majority of apartment housing was constructed under government introduced to cope with the rapid population growth and economic growth the country experienced in the 1970s. Forty years later, problems are beginning to appear in these aging buildings and the situation is worsening rapidly. The cost of maintaining a building is a major factor in its overall life cycle cost, so managing the scarce resources available for maintenance wisely is vital. Given the growing number of apartment housing blocks that require maintenance and the importance of maintenance in terms of the life cycle cost, effective strategies for minimizing the maintenance cost and avoiding excessive concentration of maintenance costs are urgently needed. The cost of maintaining a building depends on the types of maintenance required and the age of the building. In the case of apartment housing, the range of maintenance costs can be especially wide because the maintenance items themselves vary hugely and the frequency is relatively high. It is therefore necessary to develop a strategy for evaluating a building’s maintenance costs that considers both its life span and the specific maintenance items required based on that building’s historical maintenance data if decision makers are to be able to make accurate and effective decisions that make the best use of the available resources when setting priorities for its maintenance budget. A new Loss Distribution Approach (LDA) based framework was developed for this study that minimizes maintenance costs in terms of a building’s overall life cycle costs by evaluating the cost of maintaining individual buildings based on their historical maintenance data. LDA is a popular statistical methodology that is used where a large amount of real world data is available for estimating the expected loss of an event by assessing a bank or insurance company’s operational risk, for example, thus making it easier for them to establish a systematic cost management strategy. In this study, a maintenance cost evaluation framework based on LDA using historical data was developed to analyze a building’s overall maintenance costs as well as the costs it might incur for a range of scenarios, taking into account both individual maintenance items and the maintenance times required for each. The study is presented here in 5 chapters, with the main contents of each chapter as follows: Chapter 1. Introduction Chapter 2. Literature review Chapter 3. Research methodology Chapter 4. Validating the LDA framework for evaluating maintenance costs Chapter 5. Discussion and conclusions The literature review examines the current status of our understanding of the maintenance cost of major infrastructure elements and buildings, focusing specifically on the processes used for the evaluation or estimation of the maintenance cost, and on the data based probabilistic cost evaluation methodology utilized. In addition, the research on LDA is examined to determine its potential utility in the field of maintenance cost evaluation. In the chapter on research methodology, the development of the LDA based maintenance cost evaluation framework proposed here is presented, including the steps required to adapt it to estimate building maintenance costs. This chapter also describes the collection process for the 7,267 maintenance data events in 1,915 apartment housing complexes in South Korea during the 5 years from 2012 to 2016 used to validate the proposed framework in the following chapter. The final chapter presents the study’s results, discusses their implications, and makes recommendations based on the findings. The new framework for evaluating building maintenance costs using LDA that was verified using historical maintenance data for apartment complexes in South Korea is presented in this chapter. By applying LDA to the assessment of maintenance costs, it is possible to assess maintenance costs in specific situations based on a risk matrix as well as performing a comprehensive assessment of maintenance costs. This approach is expected to help decision makers assign the optimal priority mix for their maintenance budget investment by calculating the statistical value based on the nature of the maintenance item and the number of years post-construction, greatly facilitating the budget allocation process. The findings revealed by the new LDA based maintenance cost evaluation framework are as follows: 1) In the first 11 to 20 years after the completion of an apartment complex, the maintenance cost tends to be above average and the frequency high. This is because the replacement of the mechanical, electrical and plumbing systems and the interior and exterior painting occur during this time period due to natural wear and tear and the normal replacement period. 2) The “building exterior”also exhibited high severity and high frequency for the period 11 to 20 years post-completion. This includes a large-scale construction project covering the entire complex and a relatively short repair cycle, hence the high values for the severity and frequency. 3) The “building interior”showed high severity and low frequency during the first 10 years post-construction. The interior has a high severity due to the simultaneous occurrence of painting the surfaces in each household, including the ceiling, the inner wall, and the staircase. The elapsed years 0 to 10 years represent relatively new construction, so the maintenance frequency is relatively low. 4) “Electricity, safety, elevator & home networks”showed a high frequency from 0 to 20 years. These essential facilities are all related to providing a comfortable life for the occupants, including repairs and the replacement of mechanical equipment. In this study, the usefulness and feasibility of the LDA based maintenance cost evaluation framework was verified through a validation process. Although in this case the validation focused specifically on apartment housing complexes, if it were to be extended to infrastructure elements such as bridges, harbors and tunnels it could not only serve as a useful tool for the national budget allocation but also provide useful feedback supporting the development of more effective maintenance schedules.