Effects of Construction Cost and Volume on Construction Time of Educational Projects in Texas

Abstract

Studies indicate that there is a relationship between project cost and construction time for different construction markets. The purpose of this study is to validate the time-cost relationship model developed by Bromilow et al. 1 in context with educational construction projects in Texas. The model was extended to include the magnitude of the projects in terms of gross floor area and project delivery methods to determine whether these variables also have an effect on project duration. Data related to 39 educational projects was obtained for the study. SPSS ® program was for analysis of the data. The statistical technique used for the analysis was stepwise linear regression. The results indicate that when gross floor area is also used an independent variable, construction cost does not have any relationship with construction time for educational projects in Texas. However, the results show a statistically significant relationship between construction time and magnitude of the project, measured by gross floor area, at the level of significance (p-value) of <0.0001. A prediction model of construction time has been developed based on the results of the study. This model will be useful to students taking courses related to cost estimating and construction project scheduling and also to professionals involved with construction industry.

Construction time prediction model for public building projects

“…They argued that construction cost, when decomposed, consists primarily of labor and material costs; labor cost is a function of time, and material cost of a building is a function of gross floor area. A similar model to Bromilow’s time-cost relationships was advocated by different scholars, for instance, (Chan, 1998) and (Choudhury, 2012), of the form T = L A M , where A is the total gross floor area in m 2 , and L and M correspond to the constants K and B in the time-cost empirical model. Chan (1998) and Choudhury (2012) tested the assumed time-gross floor area model and were found it to be significant but marginally worse than the time-cost model in some cases.…”

“…Accordingly, the study incorporated a total sample size of fifty-six (56) public building projects; thirty-one (31) Educational and twenty-five (25) Office building projects completed during the period from 2007 G.C to 2014 G.C. The sample size consisted in this study considered to be satisfactorily relative to the previous similar studies; (Choudhury et al , 2002) consisted of the data of 35 health sector construction projects, Choudhury and Rajan (2003) 55 residential projects, Car-Pušić and Radujković (2009) 17 multistorey building projects, Le-Hoai and Dai Lee (2009) 34 building projects, Choudhury (2012) 39 educational projects, Chan and Chan (2002) data of 56 residential buildings and (Odabasi, 2009) 7 educational building projects.…”

“…The other study made by Walker (1995) suggested that building size measured by gross floor area is a better predictor of construction time performance than the project cost. Love et al (2005) (Chan, 1998) and (Choudhury, 2012), of the form T = L A M , where A is the total gross floor area in m 2 , and L and M correspond to the constants K and B in the time-cost empirical model. Chan (1998) and Choudhury (2012) tested the assumed time-gross floor area model and were found it to be significant but marginally worse than the time-cost model in some cases.…”

“…Variable selection can also be made from reviews of pieces of literature that are claimed as commonly accepted factors (Williams, 2008). This approach is practiced in most researches on time model development; for instance (Hoffman, 2005; Assefa, 2008; Williams, 2008; Helvacı, 2008 and Choudhury, 2013). But, to be useful in predicting construction durations at an early stage, the factors must be easily identifiable or known during the planning phase of a project.…”

“…Most of the previous models have found one of the good measures of project-scope to be the construction cost of the project (Bromilow et al , 1980; Kaka and Price, 1991; Kumaraswamy∗ and Chan, 1995; Chan, 2001; Ameyaw et al , 2012; Otali and Adewuyi, 2015 and others). Few researchers, however considered the gross floor area and/or number of storeys as a project scope factor for time prediction (Chan, 1998; Choudhury, 2012; Kumaraswamy∗ and Chan, 1995; Mackova et al , 2017; Love et al , 2005 and others). Some studies have attempted to improve the accuracy of the BTC model by incorporating predictor variables, i.e.…”