Fuzzy Agent based model framework for BIM Adoption for Facility Management in the New Zealand Construction Industry from Stakeholders Collaboration Perspective

I. INTRODUCTION AND PORTRAYING THE CURRENT PROBLEM
The potential benefits of building information modeling (BIM) for facility management (FM) have been acknowledged by the architecture, engineering, construction, and operations (AECO) industry. The remarkable benefits of 6D BIM are to deliver precise, timely, and appropriate information all over the lifecycle of an entire portfolio of facilities, not just during the design and construction phases (Naghshbandi 2016). BIM-FM has the potential to add value to the building asset throughout its lifecycle by providing semantically rich visual representations of facilities (Enegbuma et al. 2014). Although BIM has become more commonly adopted by the AECO participants and its capability to support FM has become unequivocally evident, its widespread acceptance and adoption have remained an unsolved issue globally. In New Zealand also, BIM adoption is still in its infancy, and the level of BIM uptake, especially amongst small and medium-sized enterprises (SMEs), which account for 97% of the total construction companies, is indispensably low (Harrison and Thurnell 2015). Therefore most construction projects are not BIM-centric, especially in "post-construction" phases such as facilities management. (Tran et al. 2012).
According to scholars, a vital step to reap the benefits of BIM in the facility management phase is to determine the problems and concerns that affect the excellent performance of facility operation and maintenance activities among the stakeholders. Stakeholders include project owners, users, project manager, facility manager, designers, legal authorities, employees, contractors, subcontractors, suppliers, sponsors, and the natural environment who, besides organizational rules and established physical processes, have their subjective attributes and behavioral rules.
As mentioned by Liu and Issa (2016), BIM is regarded as a way of designing, constructing, running, and maintaining an asset as a collaborative process. Collaboration can be seen as working together in a seamless team for common objectives that deliver benefit to all (Enegbuma et al. 2016). In other words, BIM is the collaboration of various stakeholders at different stages of a facility's life cycle to insert, extract, update or modify building information in the model to support and reflect the roles of that stakeholder (Enegbuma et al. 2014). Although collaboration amongst stakeholders who must adjust their technologies, processes, and capabilities to align with BIM requirements has the highest impact on BIM uptake, in research by Doan et al. (2020), the insufficiency of collaboration and coordination was confirmed by key construction practitioners as a substantial challenge/barrier on the road to BIM uptake in New Zealand's AECO industry. In another related research, Okakpu et al. (2020) confirmed that a lack of collaboration is among the risk factors that caused a decline in BIM adoption among New Zealand construction firms. Therefore, there is a need to deeply understand BIM-FM stakeholders and their interactions, thereby constructing a framework to allow managers to handle them more effectively. In this regard, identifying and then removing BIM stakeholders' collaboration barriers could play a pivotal role in eliminating BIM-FM use hurdles and conquering its full advantages.
Agent-based modeling (ABM), a modeling technique that is relatively new to the field of construction management, has been used to simulate complex systems of interacting agents (Berglund 2015). In light of subjective uncertainty in BIM-FM stakeholders' collaboration, there is a need to consider more complex agent models in human behavior and social relationships. Fuzzy logic is useful in vague and uncertain environments, as it is the case in the study of human societies. Therefore, the modeling can be done using a fuzzy agent-based model (FABM). FABM is defined by four components: a FABM model architecture, a set of agents specified by their attributes and behaviors, a set of agents' relationships, and methods for interaction (Kedir et al. 2020; Raoufi and Fayek 2018)
II. PREVIOUS WORKS
While the positive value that BIM can add to FM practice has been described in detail and long-anticipated, the facility management (6D) aspect, BIM-FM, has not caught adequate attention worldwide, and the 6D phase has barely been addressed in current BIM research. Edirisinghe et al. (2017) reviewed 207 BIM published studies; their study revealed that there are very few papers on BIM in FM. In the New Zealand context, which is the focus of the current research proposal, academia has mainly concentrated on potential benefits and challenges of BIM applications (e.g., Davies et al. 2017; Doan et al. 2020; GhaffarianHoseini et al. 2017; Harrison and Thurnell 2015; Stanley and Thurnell 2014). Despite a gradually increasing number of published works since 2007, which are generally limited to identifying benefits and barriers, BIM adoption research in New Zealand has received insufficient attention from scholars. This neglect hinders the understanding of BIM stakeholders, ultimately harming the development of the BIM-FM sector.
Specifically, when it comes to BIM stakeholders, a few published works can be found. Recently Okakpu et al. (2019) empirically compared the real potential benefits between traditional network and BIM network for a real-time refurbishment case study project through agent-based simulation modeling of stakeholder interaction. In another related work, Okakpu et al. (2020) identified peculiar risks associated with refurbishment projects for BIM adoption from stakeholder perceptions in New Zealand. Although these works are first steps toward BIM stakeholder’s management in the New Zealand AEC industry, their focus is on building refurbishment projects with characteristics that differ fundamentally from those in FM practices. Additionally, without integrating the fuzzy concept to the agent-based simulation modeling known as FABM, this technique cannot handle the subjective uncertainty in the variables and the stakeholder’s relationships in construction systems. Finally, even though the literature review showing a lack of effective collaboration among stakeholders hinders BIM development in New Zealand, research has ignored BIM adoption issues specifically from the stakeholder's collaboration perspective. Hence, it can be inferred that the literature has remained silent about BIM-FM uptake among different stakeholders considering subjective uncertainty exists in the variables and the relationships of the BIM stakeholders. Therefore, there is a research gap, about impact of dynamic and complex interactions between BIM-FM stakeholders' behaviors in BIM-FM adoption.

III. RESEARCH AIM AND OBJECTIVES
The principal aim of this research is to examine the current collaboration frameworks and develop an optimized alternative framework using fuzzy agent-based simulation modeling of New Zealand BIM-FM stakeholders from a collaboration perspective with particular attention to the subjective uncertainty and behavioral rules that exist in construction systems and societies relationships. Based on this research aim, four objectives are defined as follows:

(1) identify of key agents and activities along with perceived drivers of and barriers to BIM-FM adoption among New Zealand BIM-FM stakeholders from a collaborative perspective; (2) simulate and visualize dynamic and static interactions of agents by developing a fuzzy agent-based simulation model of stakeholders interactions capable of prioritizing and ranking the adoption barriers taking into account the impact of stakeholders behaviors as a whole; (3) formulate an optimized collaborative framework that can be adopted by New Zealand BIM-FM practitioners toward increasing BIM-FM uptake in AECO industry; and (4) demonstrate the application of the framework to case studies.
Finally, the research question can be designed as follows:
What is an optimized collaborative framework from a collaboration perspective capable of considering both static and dynamic impacts of stakeholders?

IV. SUMMARISED RESEARCH METHODOLOGY

This study involves a three-stage of data collection, analysis, and interpretation to address the research question. The research consists of three studies to address the key research question. The role of the studies are as follows:

Studies Ⅰ and Ⅱ:
Since BIM-FM is in its early stages, the aim of these studies is to provide a range of data that is currently unlikely to be found by either reviewing the available literature or field survey. Study Ⅰ involves a questionnaire-centered survey of international BIM and FM researchers. Study Ⅱ involves semi-structured interviews with New Zealand subject-matter experts. For the purpose of the research, in studies, Ⅰ and Ⅱ two sets of data will be collected. Set 1 includes identification of agents, activities, and institutions of the different players in BIM-FM, which will be extracted from a questionnaire with key representatives of each of the stakeholders. Set 2 includes data about existing collaborative practices of BIM-FM in the New Zealand AECO industry using semi-structured interviews with New Zealand subject-matter experts and with key representatives of each of the stakeholders.

Study Ⅲ:
Based on gathered data in the studies Ⅰ and Ⅱ, in the Study Ⅲ, using a methodology framework adopted from Raoufi and Fayek (2018), first, the Fuzzy Agent-Based Model Architecture will be determined, then, the basic structure of agents, including the types of attributes and behaviors of each agent in the model, will be defined. After which, agent static or dynamic interactions will be defined, then to account for and incorporate the subjective uncertainty involved in agent behavioral rules, using fuzzy rules, agent behavioral rules will be defined. Finally, the simulation experiment will be performed using ANYLOGIC and MATLAB soft wares. The research will end with an optimized collaboration framework for BIM-FM practitioners in New Zealand, proposing practical solutions for BIM-FM adoption, which its application will be examined using various case studies.
The proposed study that can be classified as a semi-qualitative data-driven stakeholder analysis is novel concerning (1) the application of FABM in the analysis of stakeholders' complex behavior in BIM-FM, (2) visualization of the stakeholders' interactions in BIM-FM, (3) Identifying BIM-FM uptake barriers from stakeholders' collaboration perspective.
V. POSSIBLE OUTCOMES
The research will provide valuable insights into BIM-FM uptake barriers. The results could be considered baseline information for the companies and government to have effective strategies towards BIM stakeholder's management and, ultimately, BIM adoption. Using the proposed research outputs, it is hoped that the BIM policymaking landscape can be enhanced; this can help to tackle the BIM stakeholders related challenges and pave the way toward the further adoption of BIM in the New Zealand construction industry.