DEVELOPING COST PREDICTIVE MODEL FOR GREEN BUILDING PROJECTS IN TERTIARY EDUCATIONAL INSTITUTIONS IN SOUTH-WESTERN NIGERIA

Abstract

A green building is developed with the aim of achieving the objectives of sustainable development of social, economic and environmental concerns. The study was aimed at developing a cost predictive model with a view to enhancing decision making for green building development among stakeholders in South-Western Nigeria. The study adopted the pragmatic approach using the mixed design methods. The study population comprised the built environment professionals in the Physical Planning Units (PPUs) of the Federal Universities and Polytechnics in the study area totaling 69 and Tertiary Education Trust Fund (TETFund) projects executed from 2011- 2017. A structured questionnaire adopting the LEED v4: Building Design & New Construction rating tool on a five point Likert‘s scale was adopted for data collection. Secondary data on design parameters and the elemental cost details were collected for the development of the cost predictive model. The methods for data analysis included the Mean Item Score (MIS) for ranking the level of awareness of green building requirements; Kruskal Wallis Rank Sum test for the hypothesis testing, Man-Whitney U test and Least Significant Difference (LSD) for the Post- Hoc tests. The cost predictive model was developed using the Artificial Neural Network (ANN) and the Deep Neural Network (DNN). The results showed that the level of awareness of the requirements for green building development was generally above average but only significant for three out of the seven professionals namely the Architects, Engineers and Land surveyors ranked respectively. The Kruskal Wallis H Test showed a significant difference of p = 0.027 at 95% confidence interval. Only 34% of the completed projects scored 40points and above required for the minimum LEED certification. There was a significant difference in the level of compliance among the TEIBs (p = 0.001). However, the Man-Whitney test showed no significant difference in the level of compliance between University and Polytechnic buildings (U= 0.483). There was also a weak correlation (+.355 at .001 significance level) between the level of compliance with and the level of awareness of the requirements for green building development. The study concluded that the built environment professionals are above average in their awareness of green building requirements with significance difference in the level of awareness across the professionals. Although the level of compliance is low at 34%, it however debunked the impression that there are no green buildings in Nigeria. The results as well showed that the ANN model can predict the cost of green building project with 99% accuracy. The study recommended the need for professional institutions to develop skills and collaborative efforts for green building development, that the Federal government through the Tertiary Education Trust Fund (TETFund) should formulate and implement policies that will enhance the level of compliance with green building principles. The ANN model should be adopted for cost prediction in making informed decision for green building development.