Relationship between BIM implementation and Performance of OSM projects 1

AbstractBuilding information modeling (BIM) and off-site manufacturing (OSM) approaches have been recognized as revolutionary methods in the construction industry. However, limited empirical studie...


70
On the other hand, OSM can be an alternative to the traditional construction method. Scholars have explored 71 OSM's advantages, which can increase buildability, save labor requirements, improve health and safety 72 performance, and reduce waste, noise, and dust (Pons and Wadel 2011;Lu et al. 2012). In developed countries, 73 this construction approach has been used frequently to improve the speed and quality of construction delivery 113 long-winded lead-in times, high cost, and space clash (Ezcan et al. 2013). Integration management occupies an 114 important position of project management, which refers to arranging a set of processes needed by a project 115 orderly and economically (Demirkesen and Ozorhon 2017). Therefore, BIM implementation could provide an 116 integration paradigm to achieve efficient management among OSM stakeholders. Thus, the following 117 hypothesis is proposed: The processes of OSM require highly continuous collaboration among all stakeholders to achieve project 121 success. The building products are well organized, manufactured, labeled, and then delivered on time with a 122 proper logistic arrangement to follow the building assembly process. This is in step with the theory of lean 123 construction, which focuses on partners' mutual assistance, optimal arrangement, and integrated project 124 management for higher productivity (Excellence Constructing 2004). The powerful function of the BIM 125 platform can capture precise information provided for different participants and standardize the format of digital 126 exchange for coordination and interaction with different parties across trades. The organizational integration in 127 the project is vital, and involves experience, strategy, process, knowledge, and innovation. Therefore, it has 128 great potential to promote the performance of OSM and enables unified organization to remain competitive 131 including information of various types of OSM technology. This facilitates the exploration of partnerships and 132 mitigation of market risks (Pan et al. 2012). Some studies stressed that flexible internal change has an important 133 effect on better performance in the OSM projects management process (Söderlund 2010). It seems that 134 integration management could reduce opportunistic behavior and enhance project management performance.

135
Thus, the following hypothesis is proposed:

187
One of the functions of a contract is to minimize transaction costs. The costs accrue from opportunism and 188 information asymmetry under an integral system. In order to avoid opportunism, a correct and matching contract 189 should be constructed, as proper contract management will positively influence project performance (Gong et 190 al. 2005). Therefore, when BIM is used in the OSM industry, a new contract system will be formed. Contracts  that a certain degree of freedom within a contract structure could provide the essential "adaptability" to deal

203
The shared expectation of outcomes and good relationship would drive stakeholders to support each other. Based 204 on the above perceptions, contract management is not restricted to transaction affairs, but is also used to maintain 205 partnerships from long-term development, and finally achieve the cooperation probability. Therefore, the 206 following hypothesis is proposed:

225
A questionnaire survey approach was adopted to collect the data, and SEM was used to analyze the relationship 226 among the variables through the PLS technique. This technique is tested by the cross-validation method of Stone 227 (1974) andGeisser (1974) and used to detect the causal relationships and measure small data samples and 228 skewed distributions (Hulland 2015). Therefore, the PLS method was an appropriate approach to analyze 229 individuals' attitudes that were measured on a five-point Likert scale.

230
The questionnaire was designed based on the constructs in the conceptual model, and they were distributed to 231 industry participants who are acquainted with BIM technology and OSM. Initially, the key variables are 232 identified by a sequence of relative literature, and then the preliminary questionnaire was checked by three 233 industry experts to confirm the clarity of the measurement items as per the context. They serve as the managing 234 director, general manager, and senior project manager of local publicly listed companies, with an average 30 235 years of working experience. Then, a validity evaluation of the scale was conducted and nine items were deleted 236 as they did not satisfy the convergent validity. Finally, the remaining 26 items were formulated as shown in 237 Table 1. In this research, the conceptual model ( Fig. 1) contains five variables, of which four are latent variables https://doi.org/10.1061/(ASCE)ME.1943-5479.0000704 https://creativecommons.org/licenses/by/4.0/ treated as the independent and only exogenous variable (expressed by a circle). This construct relates to the 240 extent to which BIM is applied in OSM projects. Two dimensions, BIM LOD and project phases, can be used 241 to measure the BIM implementation. A multiplicative index provides a more reliable measure than a simple 242 additive index for the extent to which BIM has affected a project. The level of information contained in a BIM 243 model, and its corresponding available data, is described by the common term of "level of development" (LOD).

244
This type of BIM maturity model had been developed by Bew and Richards (2008)   A thorough measurement analysis is conducted in all constructs to reduce measurement error before 276 implementing the hypothesis testing. The analysis includes an evaluation of the reliability and validity for this 277 study. Generally, the internal consistency reliability is estimated through Cronbach's alpha in the research model.

278
The value of Cronbach's alpha should be greater than 0.7 for an acceptable level of reliability (Hair et al. 2006).

279
The validity is reflected by the composite reliability (CR) and average variance extracted (AVE). The threshold 280 values of CR and AVE are suggested to be greater than 0.7 and 0.5, respectively (Fornell and Larcker 1981).     Pearce (2001) Mahama (2006) B5 The partners keep one another informed of changes and events that might affect them.

281
Pearce (2001) B6 In most aspects of our relationship, the parties are jointly responsible for getting things done.
Pearce (2001) Cannon and Perreault (1999) B7 Problems are treated as joint rather than individual responsibilities in the cooperation process.