CHAPTER 1

INTRODUCTION

1.1 Background to the study

The importance of ensuring accuracy and reliability of construction time estimates at the tendering stage cannot be overemphasized. Accurate early estimates of construction time typically provides clients and contractors with a basis for evaluating the success of a project and the efficiency of the project organisation (Nkado, 1995). They also provide them with a basis for ascertaining logistical and cash flow implications for feasibility, budgeting, planning, monitoring and even litigation purposes (Skitmore and Ng, 2003). Furthermore, they serve as a criteria in determining the best combination when performing time-cost optimization (Que, 2002). It is therefore clear that construction time has become a vital tool used by clients and contractors to ensure the success of construction projects. This success will however, only be achieved when construction time is accurately predicted.

Construction time/periods are often calculated on the basis of the planner’s own previous experience on similar projects (Choudhury and Phatak, 2004). However, as pointed out by Skitmore and Ng (2003), the need to reduce the problem of subjectivity associated with the planner’s experience and judgement to correctly interpret project and site information and make the best possible decisions, has long resulted in the development of construction time predicting models.

The development of construction time predicting models commenced with the use of project scope factors. Project scope is a measure of project size, which can be described as construction cost, project duration, gross floor area, number of storey, building type and procurement method (Walker, 1995).

1

The fact that the development of construction time predicting models commenced with the use of project scope factors, can clearly be seen in the time-cost model originally developed by Bromilow (1969). It is also evident in the validation of Bromilow (1969) model by other researchers e.g. in Australia (Ireland, 1983; Yeong, 1994; Ng et al., 2001), in the UK (Kaka and Price, 1991), in Hong Kong (Chan, 1999), in Nigeria (Ojo, 2001; Ogunsemi and Jagboro, 2006), in Korea (Long and Young, 2009), as well as in Ghana (Ameyaw et al., 2012). Efforts to improve the Bromilow (1969) model are also clear indications that the development of construction time predicting models commenced with the use of project scope factors, e.g. Chan and Kumaraswamy (1995) combined cost and gross floor area (GFA) to model construction time; Love et al. (2005) on the basis that cost is a poor predictor of construction time since the final cost is not known at the initial stages of a project, modelled construction time using GFA and number of storey; Chen and Huang (2006) in a bid to finding a way to quickly estimate the duration and cost of reconstruction projects in the early planning phase where detailed information is not available, modelled construction time using cost and number of floors.

Project scope based models were found good for predicting construction time, however, as researchers (e.g.; Walker, 1995; Chan, 1999; Ng et al., 2001) rightly observed, they lacked the inclusion of several other non-project scope (qualitative) factors which influence construction time. Walker (1995) specifically noted that the principal criticism of project scope based models is that the management process is too complex to be relegated to the value of a constant conveniently described in a formula, regardless of the correctness of the statistical methodology used. It is on this basis that a recommendation for further research was generally been made by the aforementioned research works, for the identification of those other non-project scope (qualitative) factors which influence

2

construction time, and the full incorporation of their coefficients and weightings into a model for predicting construction time.

Several researches have been carried out to identify other non-project scope (qualitative) factors which influence construction time, some of which include the works of Ireland (1985), Walker (1995), Nkado (1995), Koushiki et al. (2005) as well as Olupolola et al.

(2010). Research works have also been conducted to develop multivariate construction time predicting models combining project scope and other qualitative construction time influencing factors together (e.g., Chan and Chan (2002) developed a multivariate model for public housing projects in Hong Kong; Hoffman et al. (2007) also developed a multivariate model for facility projects funded by the US air force).

However, because there is little consensus as to which combination of the qualitative construction time influencing factors provide an accurate predictor of construction duration (Nkado, 1999) and inaccuracies may result when regression models are used to extrapolate or predict values of the dependent variable for independent variables which are outside the population for which the original model was developed (McClave et al., 2000), it is obvious that existing multivariate models are only applicable in the regions/countries where they were developed.

1.2 Statement of the problem

Project scope based construction time predicting model has been developed for Nigeria by Ogunsemi and Jagboro (2006). However, the research work of Olupolola et al. (2010) which revealed 23 key factors as having stronger effect on the time performance of building projects in Nigeria, is an indication that just like other project scope based construction time predicting models developed across the world, the time-cost model

3

developed by Ogunsemi and Jagboro (2006) falls short of other factors that influence construction time (in Nigeria).

Nigeria is therefore faced with the question of how these other factors can be combined with project scope factors to predict construction time, since despite the fact that multivariate models have been developed in some parts of the world they cannot be used in Nigeria because as Nkado (1995) rightly observed, there is little consensus as to which combination of these more qualitative factors provides an accurate predictor of construction duration. In addition, McClave et al. (2000) has noted that inaccuracies may result when regression models are used to extrapolate or predict values of the dependent variable for independent variables which are outside the population for which the original model was developed.1.3 Aim and Objectives

1.3.1 Aim of the research

To develop a multivariate construction time predicting model for Nigeria.

1.3.2 Objectives

The objectives through which the above stated aim was achieved were as follows:

i.            To evaluate the project scope factors influencing construction time of building projects.

ii.            To assess the qualitative factors influencing durations of building projects.

iii.             To develop the model.

iv.            To test the model.

4

1.4 Justification for the Study

Modelling building construction durations in Nigeria will provide Nigeria with a construction time predicting model which incorporates both project scope factors and managerial/qualitative construction time influencing factors together. This will serve as a means to ensure more accurate and realistic predictions of construction time as all construction time influencing factors will be taken into consideration any time the model is used to make predictions.

Consultancy firms and contractor organizations will therefore have the opportunity to make more realistic schedules and benchmark measures of construction period, thereby facilitating effective planning.

1.5 Scope and Limitations

1.5.1 Scope of the study

Building construction projects of a contract sum not less than ten million naira, completed via the traditional procurement route, between the years 2007-2012 were considered for this research. The range of years (2007-2012) and the choice of projects having a contract sum not less than ten million naira, was to ensure that the most recent trend within the construction industry is captured. As it was not possible to cover the whole Nigeria in the study, only Abuja (FCT), Kaduna and Kano states of Nigeria were covered by the study. These states were selected because they were considered to be among some of Nigeria’s most developed states, with a relatively high concentration of construction firms in them.

5