Big data analytical framework in managing maintenance management for government office buildings in Malaysia

Abstract

The government sector in Malaysia faces major challenges in managing maintenance data. The development of technology and software for industry 4.0 has produced a vast volume of data, and the increase is very high. The sudden rise of Big Data has left real estate players unprepared to use it effectively. Furthermore, the Computerized Maintenance Management System (CMMS) and Sistem Pengurusan Fasiliti Berpusat (eSPFB) used in the government, especially in Putrajaya, are not working well. Based on the research and monitoring conducted, the data in the CMMS are still incomplete for analysis and projection to assist or support strategic decisions in managing facilities. Although it can produce dynamic dashboarding for decision-making, it does not involve Business Intelligence (BI) in providing real-time analysis or an interactive dashboard to the user, making it easier for newcomers to understand the system. Scattered, insufficiency and inaccuracy of maintenance data have become challenges for the maintenance department, making modelling the process or the management of maintenance activities enormously hard and complex. To mitigate this situation, the government must have a framework that can assist in managing the maintenance data management of public facilities, which encompasses an improvement tool through the dashboard simulation model for enhancing current conventional maintenance practices containing the necessary information to satisfy the stakeholder. Due to problems arising in the management of government building maintenance, especially during the decision-making stage, this research attempts to develop a new approach in managing dispersed and complex domain structures using Business Intelligence. Three objectives drove the study, firstly to identify data management challenges in maintenance management; secondly to determine the existing big data and business intelligence in federal government buildings, and thirdly to develop the big data analytical framework in maintenance management for federal government buildings. The Federal Territory of Putrajaya was chosen as the case study for this research. Three research methodologies were employed to achieve the research objectives, a literature review, a questionnaire survey and expert opinion. Firstly, the literature review identified four barriers to CMMS and eSPFB implementation and eight elements of data management challenges in government buildings. The respondents were asked to choose their level of agreement with the barriers and data management challenges. The respondent involved experts from the maintenance and asset management field, making them reliable and relevant for validating the barriers and data management challenges in maintenance management. Six experts were selected based on purposive sampling. Next, questionnaires were distributed to the target group of 35 supervisors who were selected through random sampling at the Jabatan Kerja Raya, Putrajaya. Data were analysed using IBM SPSS 23. The result showed that 73% of the respondents had difficulties collecting maintenance management data. Lastly, the big data analytical framework was developed, grounded by a dashboard simulation model and validated through expert opinions. The developed framework and dashboard simulation model was recommended as a new approach to replace the existing conventional method. In conclusion, this approach is an added value for the government in making structured knowledge in conveying maintenance data to the users for decision-making and better performance of public facilities by Jabatan Kerja Raya.