International Journal of Project Management 20 (2002) 593599 [Link].

com/locate/ijproman

The eects of construction delays on project delivery in Nigerian construction industry

Abstract Construction delay has become endemic in Nigeria. It is imperative to create awareness of the extent to which delays can adversely aect project delivery. This paper identies, by questionnaire evaluates and through empirical method assesses the eects of construction delays. The ndings showed that time and cost overruns were frequent eects of delay. Delay had signicant eect on completion cost and time of 61 building projects studied. Client-related delay is signicant in Nigeria. Acceleration of site activities coupled with improved clients project management procedure and inclusion of appropriate contingency allowance in precontract estimate should assuage the adverse eect of construction delays. # 2002 Elsevier Science Ltd and IPMA. All rights reserved.
Keywords: Managing projects; Cost; Time; Claims

1. Introduction The Nigerian construction industry continues to occupy an important position in the nations economy even though it contributes less than the manufacturing or other service industries. The contribution of the construction industry to national economic growth necessitates improved eciency in the industry by means of cost-eectiveness and timelines and would certainly contribute to cost savings for the country as a whole. A major criticism facing the Nigerian construction industry is the growing rate of delays in project delivery. Delay is a situation when the contractor and the project owner jointly or severally contribute to the non-completion of the project within the original or the stipulated or agreed contract period. In countries such as United State of America (USA), United Kingdom (UK) and Western Germany, Mobbs [1] found that construction time is better. In Nigeria, Ajanlekoko [2] observed that the performance of the construction industry time wise is poor. An investigation by
* Corresponding author. E-mail addresses: aaibinu2002@[Link] (A.A. Aibinu), gjagboro@[Link] (G.O. Jagboro).

Odeyinka and Yusif [3] shows that seven out of ten projects surveyed suered delays in their execution. According to Chan and Kumaraswamy [4] timely delivery of projects within budget and to the level of quality standard specied by the client is an index of successful project delivery. When projects are delayed, they are either accelerated or have their duration extended beyond the scheduled completion date. These are not without some cost consequences. The conventional approach to managing the extra cost is to include a percentage of the project cost as contingency in the pre-contract budget. According to Akinsola [5] conventional allocation of contingency is based on judgment. However construction projects are unique, as they may have a distinctive set of objectives, require the application of new technology or technical approaches to achieve the required result. This uniqueness makes the contingency allowance allocation based on assumption and intuition inadequate and unrealistic. An investigation by the authors revealed that in Nigeria 510% of precontract estimate is in most cases allowed as contingency. This allowance was found to be inadequate. Inadequate contingency implies extra nancial commitments, which in some cases are beyond the capacity of the owner. Clients are in some cases not

0263-7863/02/$20.00 # 2002 Elsevier Science Ltd and IPMA. All rights reserved. PII: S0263-7863(02)00028-5

594

A.A. Aibinu, G.O. Jagboro / International Journal of Project Management 20 (2002) 593599

prepared for such extra cost and so fund inform of loan are sought to oset the unexpected costs. Furthermore, associated delay problems can also result in dispute, arbitration, total abandonment and protracted litigation by the parties. To some extent the contract parties through claims usually agree upon the extra cost and time elongation associated with delay. Nevertheless, this has in many cases given rise to heated arguments between the owner and contractor. The question of whether a particular delay to progress of work warrants an extra cost and or extension of project duration is usually the cause of disagreement. Such situations, usually involve questioning the facts, causal factors and contract interpretation, which have been addressed by Alkass et al. [6]; and Bordoli and Baldwin [7]. In specic terms, Manseld et al. [8]; and Odeyinka and Yusif [3] have addressed the causes of delays in Nigeria building projects. Important causes of delay identied by Manseld et al [8] are: nancing of and payment for completed works, poor contract management, changes in site condition and shortages in materials. Odeyinka and Yusif [3] classied the causes of delay via project participants and extraneous factors. Client-related delays identied include variation orders, slow decision-making and cash ow problems while contractor-related delays include nancial diculties, material management problems, planning and scheduling problems, inadequate site inspection, equipment management problems and shortage of manpower. Extraneous causes of delay identied were inclement weather, Acts of God, labour disputes and strikes. However, while there seems to be a consensus as regards the causes of construction delays in Nigeria, impact of delay, on the other hand is an area in which little research has been carried out. There is need for awareness in the Nigerian construction industry on the changes delay can impose on project delivery. In this regard, a major aspect of this study is to empirically investigate the eect of delay on two key variables of project delivery namely completion cost and time. The question of how the eects of delays can be buered will also be addressed. Based on all this foregoing, the intention of this paper therefore is: 1. To identify and evaluate the eects of delays on building project execution in Nigeria. 2. To assess the eects of delay on completion cost of building projects. 3. To assess the eects of delay on completion time of building projects. 4. To investigate how the eects of delays can be minimized. The results of the study would lead to recommendations aimed at reducing the impact of delay on project deliv-

ery. It was also believed that the study would not only clarify but also create an awareness of the extent to which delay can adversely aect project delivery.

2. Analysis of questionnaire survey of the eects of construction delay in Nigeria 2.1. Methodology From existing literature on the construction industry in Nigeria and based on the preliminary investigation conducted at the outset of this study, it was possible to identify certain major eects of delay on project delivery. The six eects of delay identied were: time overrun, cost overrun, dispute, arbitration, total abandonment and litigation. A questionnaire was then drawn up and was divided into two sections. Section A sought to know the general particulars of the respondents while section B was focused on the six identied eects of construction delays. The respondents were asked to rank the individual eect of construction delays based on frequency of occurrence according to their own judgment and local working experience in the Nigerian construction industry between 1990 and 1999. The authors adopted a scale of 04 for the ranking exercise. To facilitate the analysis of the responses, the following numerical values were assigned to the respondents ratings (after Al-Hammad [9] and Assaf et al. [10]): Always4, Mostly3, Sometimes2, Seldom1 and Never0. The stratied random sampling technique was used. The questionnaires were sent to three groups of construction practitioners viz. quantity surveyors, architects/engineers and contractors. The practitioners were from rms located within southwestern Nigeria, which comprise six states including Lagos (the former capital of Nigeria). The rms were selected from a numerical list of rms in the geographical area drawn by the authors. Systematic sampling techniques were adopted in the selection. This according to Leedy [11] minimizes bias. Based on the result of tossing of a coin, the selection was commenced with the rst number in the arithmetic progression of odd numbers. Two hundred (200) rms were selected and were given the questionnaires and 102 were returned representing a response rate of 51% (Table 1). This was considered adequate for the analysis based on the assertion by Moser and Kalton [12] that the result of a survey could be considered as biased and of little value if the return rate was lower than 3040%. Eighty-three percent of the respondents are either HND or BSc holders and 86.27% of them are professionally registered. The average years of experience of the respondents stands at 22 years while on the average the respondents had handled 32 projects up to 1999.

A.A. Aibinu, G.O. Jagboro / International Journal of Project Management 20 (2002) 593599

595

2.2. Presentation and analysis of results from questionnaire survey Chan and Kumaraswamy [13] and Kometa et al. [14] used the relative importance index method. This method was also adopted to analyze the data collected from the questionnaire survey. The analysis was carried out for each group of respondents (quantity surveyors, architects/engineers and contractors). The ve-point scale 04 mentioned earlier was transformed to relative importance indices for each of the six eects of construction delays. The indices were then used to determine the rank of each item. These rankings made it possible to cross compare the relative importance of the items as perceived by the three groups of respondents. The weighted average for each item for the three groups of respondents was determined and ranks (R) were assigned to each item representing the perception of the three groups. The relative importance index (RII) was calculated for each item as follows (after Lim and Alum [15]): RII 4n1 3n2 2n3 1n4 0n5 4N

gathered from the respondents during the survey when acceleration is introduced, they are ineective as a result of reoccurrence of delay. This will be addressed later in this paper. Secondly, delay caused by a client or its agents generates additional costs to be borne by the client in the form of claims. The claims cover cost of plant and labour ineciency as a direct result of the delay, increases in cost occurring during the period of delay and increases in head oce overheads. It is therefore not surprising that cost overrun is a frequent eect of delays in Nigerian building projects. Furthermore, the low ranks assigned to dispute, total abandonment, arbitration and litigation were not unexpected. Claiming for additional expenses and or extension of time in cases of delay is usually through the process of claims presentation (Clause 26 of JCT, 1980 and Clause 24 of JCT, 1980). The presentation of claim according to Jagboro and Alli [16] may generate dispute between the parties and can lead to arbitration process. Where the arbitration process fails, litigation may result and may eventually lead to total abandonment of project. Dispute, arbitration, litigation and total abandonment of projects can be averted under good management of contractual claims. It is therefore not surprising that these eects of delay were ranked low.

where n1=number of respondents for Always; n2=number of respondents for Mostly; n3=number of respondents for Sometimes; n4=number of respondents for Seldom; n5=number of respondents for Never; N=Total number of respondents. Based on the ranking (R) of the weighted average of the relative importance indices (RII) for the three groups (see Table 2), it was observed that the most frequent eects of delay on building project delivery in Nigeria were time overrun (RII=0.786) and cost overrun (RII=0.671). They were followed by dispute (RII=0.456), total abandonment (RII=0.456), arbitration (RII=0.355) and litigation (RII=0.296). The high ranks assigned to time and cost overrun were not unexpected. Firstly, on most sites in Nigeria when an activity on the critical path of construction programme is delayed, it impacts the start and completion date of subsequent activities and this lead to overall time schedule slippage of project. According to information
Table 1 Questionnaire distribution and response Description Number distributed 90 50 60 200 Number of respondents 46 21 35 102 % of number distributed 51 42 58 51 % of number of responses 45 21 34 100

3. Empirical analysis of the eects of construction delays in Nigeria. 3.1. Methodology In order to empirically investigate the eects of construction delay, time and cost eects of delay were selected for further analysis. For the purpose of the investigation cost data were obtained in respect of Nigerian building projects, which were completed between 1990 and 1999. A data collection proforma designed for this purpose was employed. The quantity surveying and contracting rms selected in the questionnaire survey were asked to complete the proforma.
Table 2 Relative importance indices (RII) and ranks (R) of six eects of construction delays as perceived by the three groups of respondents in Nigeria Eects of delay Quantity Surveyors RII Time overrun Cost overrun Dispute Arbitration Litigation Total abandonment 0.799 0.728 0.419 0.370 0.277 0.495 Architects/ Engineers Contractors Weighted average (all groups) RII 0.786 0.671 0.456 0.355 0.296 0.456 R 1 2 3.5 5 6 3.5

R RII 1 2 4 5 6 3 0.774 0.607 0.441 0.345 0.274 0.452

R RII 1 2 4 5 6 3 0.786 0.679 0.507 0.350 0.336 0.421

R 1 2 3 5 6 4

Quantity Surveyors Architects/ Engineers Contractors Total

596

A.A. Aibinu, G.O. Jagboro / International Journal of Project Management 20 (2002) 593599

The information obtained include the following: project type, initial contract sum, estimated project duration, start date, actual completion date, number of oors, gross oor area, nal account sum, loss and expense claims arising from delays and uctuation claim during delay period. At the end of the collection exercise, information on 70 completed building projects were obtained out of which 61 were found complete and t for analysis. Most of the projects were residential buildings (33) and were followed by commercial/oce buildings (23) and educational buildings (5). The cost data obtained were adjusted to 1990 prices using the Consumer Price Index (CPI) obtained from the Federal Oce of Statistic. The adjustment was carried out using the cost adjustment formula: Adjusted cost data Index in base year=month Index in year=month of data cost data The adjustment was necessary because the projects were all tendered for during the period 19901999 and the eect of ination and competition had to be eliminated in order to ensure a homogenous cost comparison. Following the adjustment, the projects were classied based on initial contract value into those ranging from 0 to 10 million Naira and 10 million Naira and above. In order to examine the eect delay on completion cost, the following variables were identied: initial contract sum (X), nal account sum (Y), loss and expense claim (Le) and uctuation claim during delay period (Fcd). Loss and expense claim is an indication of additional expenditure by the client as a result of ascertained and approved delays (Clause 26 of JCT, 1980). The eect of loss and expense claim (Le) on project cost overrun (Co){derived from Final account sum (Y)Initial contract sum (X)} was investigated on one hand. On the other hand, the eect of uctuation claim (Clause 39 of JCT, 1980) during delay period on project cost overrun was also investigated. Simple linear regression analyses of cost overrun (Co) on loss and expense claim (Le) and of cost overrun (Co) on uctuation claim during delay period (Fcd) were performed using the statistical package for social sciences (SPSS). Regression equation and calculated values of test statistic F and the level of signicance of association between the variables were computed at 5% signicance level. The statistic R2 (coecient of determination) obtained measured how well the regression model actually ts, or its goodness of t. If all the observation falls on the regression line, R2 is 1. If there is no linear relationship between the dependent and independent variables, R2=0. R2 value is widely accepted to be an indicator of how well the regression model ts the population. The model usually does not t the

population as well as it ts the sample from which it is derived. The statistic adjusted R2 attempts to more realistically reect the goodness of t of the regression equation in the population. The value of R2 indicates the proportion of the variation in the dependent variable explained by the independent variable. The eect of delays on completion time was also investigated based on the classication mentioned earlier. A simple linear regression analysis of actual project duration was performed on the delay period. Regression equation and calculated values of test statistics F and the level of signicance of association between the variables were also computed at the 5% signicance level. 3.2. Presentation and analysis of results from the empirical study The regression equations, coecient of determination (R2) and F values obtained from regression analysis of cost overrun (Co) on loss and expense claim (Le) arising from delay are shown in Table 3 for projects of varying sizes and magnitude. The results showed that loss and expense claim arising from delays had signicant eect on cost overrun of building projects. The coecient of determination R2 were found to be 0.36 and 0.64 for projects ranging from 0 to 10 million and 10 million Naira and above respectively. These implies that 36 and 64% of the variations in cost overrun of the respective project categories can be explained by loss and expense claim. Loss and expense claim arises from approved and ascertained delays caused by the client and or his agents. The signicance of loss and expense claims suggests that client-related delays are signicant in Nigerian building projects. This corroborates the ndings of Odeyinka and Yusif [3], which showed that client-caused delay contributes signicantly to overall delay on Nigerian projects. The result is therefore not surprising. In the case of uctuation claim during delay period (Fcd) and cost overrun (Co), the relationship was found to be signicant (see Table 4). The coecient of determination R2 was found to be 0.55 for projects ranging from 0 to 10 million Naira and 0.80 for projects of 10 million Naira and above. These results imply that 55 and 80% of the variation in cost overrun of the respective project categories can be explained by uctuation claim during the delay period. Fluctuations in price have been on the increase in Nigeria. Jagboro [17] posited that the market situation associated with construction resources in Nigeria is such that unpredictable inuence of ination trend can make the initial construction cost outdated by the time projects are actually completed. The government deregulation policies aimed at liberalising the economy since 1986 are responsible for the instability in prices. It is therefore not surprising that uctuation claims during the delay period also contribute signicantly to cost overrun.

A.A. Aibinu, G.O. Jagboro / International Journal of Project Management 20 (2002) 593599

597

The regression analysis of actual project duration (APD) on delay (manifested by elongation period of projects EPP) showed that the actual project duration of building projects studied is signicantly dependent on the delay period. The regression coecient of determination (R2) and F values obtained for projects of varying size and magnitudes are shown in Table 5. The results showed a coecient of determination R2 of 0.72 and 0.62 for projects ranging from 0 to 10 million and 10 million Naira and above respectively. These imply that 72 and 62% of the variation in actual project duration can be explained by delays for the respective project categories. It corroborates the result of the questionnaire survey where time overrun was ranked highest as the most frequent eect of delays (Table 2). Analysis from the data collected for the study further revealed that the average time overrun of the building projects studied were 92.64 and 59.23% of the estimated project duration for projects ranging from 0 to 10 million and 10 million Naira and above respectively. The equations (Tables 35) generated serve as benchmarks for future research in studies relating to management of construction projects in Nigeria and also facilitate comparison with other countries.

ed were acceleration of subsequent site activities to reduce or if possible eliminate time overrun; and inclusion of contingency allowance in precontract estimate to buer cost overrun. 4.1. Acceleration of site activities It was gathered from construction practitioners surveyed in this study that in cases of delay of an activity on site subsequent site activities are often accelerated but this frequently failed to make up for the lost time. According to them at the outset of the acceleration process there seems to be progress but due to the deciency in project management procedure of the clients, the underlying problems responsible for delays remains in force and this often frustrates and renders acceleration ineective. A major source of deciency identied as signicantly responsible for this is clients continuous issue of design information/variation orders. 4.2. Contingency allowance Contingency sums are allowances of project cost included in the precontract estimate for the purpose of ensuring that the budget set aside is realistic and sucient to contain the risk of unforeseen cost increases during construction. The data obtained for this study in respect of 61 completed building projects revealed that the contingency allowances in all the cases were inadequate to oset the cost overrun. Investigation by the authors revealed that 510% contingency allowance is a common practice in Nigeria. Provision of adequate

4. Minimising the eects of construction delays in Nigeria building projects Based on the ndings of his study an investigation was further conducted on how the cost and time eects of delays could be minimised. The two methods identi-

Table 3 Summary of results of regression analysis of cost overrun (Co) on loss and expense claim (Le) for building projects based on initial contract sum classication Project category 010 million 10 million and above R2 (adjusted) 0.36 0.64 Regression equation Co=0.522 +2.614Le Co=2.57+11.665Le F value 0.0001 (signicant) 0.0001 (signicant)

Table 4 Summary of results of regression analysis of cost overrun (Co) on uctuation claim during delay period (Fcd) for building projects based initial contract sum classication Project category 010 million 10 million and above R2 (adjusted) 0.55 0.80 Regression equation Co=0.429+2.146Fcd Co=1.658+6.00Fcd F value 0.0000 (signicant) 0.0001 (signicant)

Table 5 Summary of results of regression analysis of actual project duration (APD) on delay (manifested by elongation period of projects EPP) for building projects based on initial contractsum classication Project category 010 million 10 million and above R2(adjusted) 0.72 0.62 Regression equation APD=6.468+1.075EPP APD=12.234+1.08EPP F value 0.0000 (signicant) 0.00001 (signicant)

598

A.A. Aibinu, G.O. Jagboro / International Journal of Project Management 20 (2002) 593599

contingency sum at the precontract stage could buer the extra nancial burden arising from project uncertainties. In this regard the percentage cost overrun of completed projects over the initial estimate is a strong parameter for assessing the adequacy of the contingency allowance. The authors were able to estimate from 61 sample cases studied a mean percentage cost overrun of 17.34%. This implies that 17.34% of project cost estimate should be included in the precontract estimate as contingency in Nigerian building projects against the usual practice of between 5 and 10%. The 17.34% contingency allowance estimated in this study falls within the 1520% allowance recommended by the United States Department of Energy (DOE) for budget estimates of new buildings.

dure. In this regard, continuous issue of design information/variation orders by the clients often frustrate the acceleration process. Acceleration has been found to be benecial in Western Germany [1]. There is a need for an eective clients project management procedure in Nigeria.  The contingency sums included in the precontract estimate of projects in Nigeria were not adequate to oset cost overrun. This study was able to establish an allowance of 17.34% of total cost estimate. This gure should however be improved upon by the use of a more comprehensive predictive model which is a subject for further research.

Acknowledgements 5. Conclusions, recommendations for construction industry practice and areas for further research The study investigates, rstly by questionnaire survey of construction practitioners, the eects of construction delay on project execution and, secondly, by empirical method, the eects of delay on completion cost and time. Two areas of how the eects of delays can be minimised were also investigated. The main conclusions of the study are as follows:  Cost overrun and time overrun (elongation of project duration) were the two most frequent eects of delay in the Nigerian construction industry.  Delay had signicant eects on actual project duration. The model relating delay and actual project duration provide a benchmark for future research work in the study of project management in Nigeria and also facilitate comparison with other countries.  Loss and expense claims arising from delay and uctuation claims during the delay period had signicant eect on cost overrun. The models provide a benchmark for future research work in the study of project management in Nigeria and also facilitate comparison with other countries.  Loss and expense claims arise from ascertained and approved delay caused by the client or his agent. The signicant eect of loss and expense claims on project cost overrun suggests that clients are a signicant cause of delay in Nigerian building projects. This corroborates the result of a previous study [3] where client-related delay was found to be signicant.  Acceleration of subsequent site activities in cases of delay to make up for the lost time has frequently failed in Nigeria building projects due to deciencies in clients project management proceThe authors wish to thank the anonymous reviewers of the paper for their constructive comments, valuable guidance and advice. References
[1] Mobbs GN. Speeding up construction. The Quantity Surveyor 1982;38(1):23. [2] Ajanlekoko JO. Controlling cost in the construction industry. Lagos QS Digest, Lagos 1987;1(1):812. [3] Odeyinka HA, Yusif A. The causes and eects of construction delays on completion cost of housing projects in Nigeria. Journal of Financial Management of Property and Construction 1997; 2(3):3144. [4] Chan DWM, Kumaraswamy MM. A survey of time-cost relationships in Hong Kong construction projects. Building Technology and Management Journal (Building Technology Society, School of Technology, Tunku Abdul Rahaman College, Kuala Lumpur, Malaysia) 1993;201994:5472. [5] Akinsola AO. A Neutral network model for predicting Building projects Contingency. Conference: Proceedings of Association of Researchers in Construction Management, ARCOM 96, Sheeld Hallam University, England, 1113 September 1996, pp. 507516. [6] Alkass S, Mazerolle M, Harris F. Construction delay analysis techniques. Construction Management and Economics 1996; 14(5):37594. [7] Bordoli DW, Baldwin AN. A methodology for assessing construction project delays. Construction Management and Economics 1998;16:32737. [8] Mansfeild NR, Ugwu OO, Doran T. Causes of delay and cost overruns in Nigerian construction projects. International Journal of Project Management 1994;12(4):25460. [9] Al-Hammed A. Interface problems between owners and maintenance Contractors in Saudi Arabia. Journal of Performance of Constructed Facilities 1995;9(3):194205. [10] Assaf SA, Musallami AA, Shash A. Professional architectural/ engineering consultancy practices in Saudi Arabia. Building Research and Information 1996;24(1):5962. [11] Leedy PD. Practical research. New York: Macmillan Publishing; 1980. [12] Moser CA, Kalton G. Survey methods in social investigation. UK: Heinemann Educational; 1971. [13] Chan DWM, Kumaraswamy MM. A comparative study of causes of time overrun in Hong Kong construction projects. International Journal of Project Management 1997;15(1):5563.

A.A. Aibinu, G.O. Jagboro / International Journal of Project Management 20 (2002) 593599 [14] Kometa ST, Olomolaiye PO, Harris FC. Attributes of UK construction clients inuencing project consultants performance. Construction Management and Economics 1994;12:43343. [15] Lim EC, Alum J. Construction productivity: issues encountered by contractors in Singapore. International Journal of Project Management 1995;13(1):518.

599

[16] Jagboro GO, Alli OR. A study of the monetary value of extension of time in building projects. Journal of Financial Management of Property and Construction 1999;4(3):3946. [17] Jagboro GO. The structural adjustment programme and the importation of building materials in Nigeria. Construction 1998; 5(2):1922.