| Author: | Peiris, Manimel Peirislage Sanduni Nuwanthika |
| Title: | Establishment of a decision-support system for smart retrofitting for office buildings |
| Advisors: | Lai, H. K. Joseph (BEEE) |
| Degree: | Ph.D. |
| Year: | 2025 |
| Department: | Department of Building Environment and Energy Engineering |
| Pages: | xix, 279 pages : color illustrations |
| Language: | English |
| Abstract: | In an era of rapid urbanization, growing populations and rising energy demands, the need for sustainable and smart solutions has never been greater. Buildings account for a major share of worldwide energy consumption making the built environment a significant contributor to greenhouse gas emissions. In this regard, Smart Building (SB) technology is a promising solution as it allows for more efficient use of resources such as electricity, water and gas, facilitating a comfortable living environment for building occupants and contributing to a more sustainable future. Hence, it significantly helps to meet climate pledges while also addressing other social and economic challenges such as evolving patterns of work and the constant demand for cost reduction. For existing buildings, retrofitting them with SB technology through a process called 'Smart Retrofitting (SR)' appears to be the obvious choice for achieving these goals. SR is the comprehensive process of upgrading pre-existing building systems using appropriate, if not optimal, retrofit techniques to integrate SB technology. It transforms existing buildings into modern, automated structures through strategic applications of Information and Communication Technology (ICT) enhanced Renewable Energy Systems (RES) and Building Energy Management Systems (BEMS). SR achieves this through the integration of advanced technologies such as Internet of Things (IoT)-enabled devices and data driven decision-making algorithms, some of which are increasingly based on Artificial Intelligence (AI). These systems allow for real-time monitoring and control of building operations, creating environments that are not only efficient but also highly responsive to user needs. Despite the transformative potential of SR projects, their implementation in office buildings involves significant challenges, including device interoperability, data integration, and balancing initial investment costs with long-term benefits. These challenges vary significantly between developing and developed regions, where differences in building stock, economic constraints, regulatory frameworks, and technological capacity necessitate distinct approaches to retrofitting decisions. For facility managers and decision-makers, these retrofits require addressing multi-faceted objectives: enhancing operational efficiency, ensuring occupant satisfaction, reducing environmental footprints, and achieving financial viability. To navigate these complexities, a robust Decision-Support Systems (DSS) is essential to provide a structured framework for selecting appropriate retrofit strategies. Developing such a holistic DSS for SR, with the present domain knowledge, is difficult, given the significant gaps in the existing literature and practice. Therefore, the aim of this study is to establish a holistic Smart Retrofitting Decision-Support System (SRDSS) for office buildings. The envisaged SRDSS evaluates and ranks retrofit alternatives for an SR project based on their performance across assessment criteria. The scope of this research is confined to leasehold office buildings, where ownership is retained by a landlord, and the majority of the space is occupied by tenants. To evaluate the model's applicability across different economic contexts, Hong Kong and Sri Lanka are selected as representative cases of developed and developing regions, respectively. The study employed a mixed-method research approach that utilized Multi-Criteria Decision-Making (MCDM) techniques. As an initial step in developing the SRDSS, a comprehensive literature review was conducted to identify the critical criteria to be considered when making SR decisions. A systematically identified set of publications was critically analysed using content analysis. The list of criteria identified from literature was finalized (by excluding any irrelevant criteria and adding any other important criteria) in two stages: a focus group meeting followed by a questionnaire survey. The final set of criteria includes: (1) investment cost, (2) operational cost, (3) financial return, (4) automated risk prediction and response capabilities, (5) reliability of retrofit option, (6) compatibility with existing systems, (7) requirement of competent Facility Management (FM) staff, (8) ease of implementing retrofitting project, (9) indoor environmental quality, (10) user-friendliness, (11) energy saving capabilities, (12) data protection capabilities, and (13) improved access control and surveillance. The importance weights of criteria were then determined using a combination of two methods. The first method, the Analytic Network Process(ANP), generated weights based on expert opinions, while the second method, the Entropy method, generated weights based on empirical data of real-world SR projects. Combining these methods ensures a balanced, accurate and robust set of criteria weights, as ANP captures the interdependence amongst criteria and the Entropy method exploits data variability. Through a series of expert interviews, the final list of 13 criteria was presented to industry experts to establish ANP-based importance weights. Twelve experts were interviewed from Hong Kong and twelve from Sri Lanka, totalling 24 interviews, from which a fixed set of weights was produced for each criterion. Next, the methods for scoring retrofit alternatives under each criterion were developed. These methods were presented to two experts, one from Hong Kong and one from Sri Lanka, to obtain feedback on their practicality. The scoring methods involved direct value inputs and subjective scoring on a Likert scale (1–5), depending on the nature of respective criterion. Direct value inputs were used for quantitative criteria such as investment cost, operational cost, financial return and energy saving capabilities. They were input directly using the quantitative values submitted by the suppliers or calculated by the facility manager's team. Likert-scale scoring was used for qualitative criteria, which include automated risk prediction and response capabilities, reliability of retrofit option, compatibility with existing systems, requirement of competent FM staff, ease of implementing retrofitting project, indoor environmental quality, user friendliness, data protection capabilities, and access control and surveillance capabilities. These qualitative criteria required assigning a subjective score based on the facility manager's perception of each retrofit alternative. Case studies were conducted to develop the SRDSS further. Four office buildings were selected, comprising two from Hong Kong and two from Sri Lanka. The case studies included discussions with the facility managers overseeing the SR projects undertaken at the selected buildings. General project data and detailed data on various retrofit alternatives from different suppliers were collected. Referring to a bespoke questionnaire, each retrofit alternative was scored by the respective facility manager. The retrofit alternative scores were then used to calculate the objective, context-specific weights for each criterion using the Entropy method. Once the Entropy weights were calculated, the weights derived from the ANP method were merged with the Entropy weights to establish a combined weight for each criterion. Afterwards, the 'Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS)' method was used to rank retrofit alternatives by determining each option's closeness to an ideal solution. Following the case studies, a structured validation exercise was conducted to ensure the reliability, accuracy and practical applicability of the SRDSS. The validation exercise elicited feedback from ten industry experts in Australia using a bespoke interview guideline. The expert feedback was addressed by conducting iterative refinements on the SRDSS that increased its practicality, decision confidence and ease of use for facility managers. The validation by Australian industry experts ensured that the SRDSS is credible and adaptable beyond Hong Kong and Sri Lanka. The final outcome of this study is a ready-to-use SRDSS in the form of a Visual Basic for Applications (VBA) macro-enabled Microsoft (MS) Excel tool. Using 13 decision criteria, the SRDSS applies ANP and Entropy for criteria weighting, followed by TOPSIS for selecting the optimal retrofit solution. Applying the SRDSS in the case studies of the two regions, Hong Kong and Sri Lanka, demonstrates its versatility in accommodating regional differences in decision-making for SR projects. By integrating VBA automation in MS Excel, the SRDSS streamlines data entry, calculations and visualization, making it accessible and user-friendly for facility managers without requiring advanced computational knowledge. In addition to advancing data-driven decision-making in SR, the outcome of this study serves as a scalable framework adaptable to various building typologies and global contexts, contributing to the significant betterment of the built environment. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/14217