A CRITICAL REVIEW OF AGRICULTURAL RELATED CONSTRUCTION PROJECT FAILURES AT THE DEPARTMENT OF AGRICULTURAL, RURAL DEVELOPMENT AND ENVIRONMENTAL AFFAIRS (DARDLEA): MPUMALANGA

The aim of the study was to determine the causes of construction project failures in the Department of Agriculture, Rural Development and Environmental Affairs (DARDLEA). A detailed literature review was carried out on construction projects that failed due to cost and time overruns. A list of construction project failure factors was identified and later categorised into contractor, client and owner-related causes. A five-point Likert scale was used to collect data from the contractors and owners and the same questionnaire was used to interview the clients or beneficiaries telephonically. The data were analysed using the Relative Importance Index (RII) and Correlation Testing. The key causes of construction project failure as per their rankings were poor communication, lack of monitoring and evaluation, lack of client or beneficiary involvement, lack of project planning, financial difficulties, poor project scheduling, incomplete project drawing, conflict, poor technical performance, and changing client requirements. The Spearman correlation concluded that relationships existed amongst the owner, client and contractor responses. From the results, the main recommendation is that all project planning, design and scheduling should be approved by the contractor, client and owner before construction commences. Secondly, all project stakeholders should be involved during the planning of projects to enhance project sustainability, and thirdly, projects should only commence when there is a guarantee that all plans, budgets and schedules are realistic, and funds are available for access. With regards to extension, the results indicate specific areas of stakeholder engagement prior and during construction to assist in managing expectations.


INTRODUCTION
The Mpumalanga Department of Agriculture Rural Development and Environmental Affairs (DARDLEA) was established in terms of section 197 of the Constitution and read with section 7 (1) and 7 (2) of the Public Services Act of 1994.The Department derives its core mandate from the provisions of schedules 4 and 5 of the Constitution of the Republic of South Africa and in accordance with section 104 (1) (b) of the Constitution (Mpumalanga Provincial Government, 2012).
The mandate of the National Department of Agriculture is the understanding of agriculture, as being inclusive of all economic activities from the provision of farming inputs, farming and value-adding (Department of Agriculture, Forestry andFisheries, 2005/2006.Department of Agricultural, Rural Development and Environmental Affairs (DARDLEA) is involved in the construction of value adding infrastructure projects for farmers in the different municipalities in the Mpumalanga province.A key observation amongst these construction projects is the fact that most of them suffer from cost and time overruns and some are ultimately abandoned before completion.The aim of this study was to investigate the causes of construction project failures at DARDLEA.

Definition of the problem
The performance of construction projects in DARDLEA is measured in terms of what was planned against the actual achievement under the constraints of schedule and budget.According to information obtained from the department's annual report for the 2013/2014 financial year, there was a gross variation between planned and actual achievement in construction projects implemented.Deviations from the targets indicated that some work was not done or was done partially, resulting in construction project failure (DARDLEA, 2014:45).In addition to these bottlenecks, Table 1 illustrates the performance of four prominent projects undertaken by DARDLEA in the province.The poor performance of the tabled projects is not unique in the province and are more the rule than the exception.This study therefore sought to answer the following research questions: • What are the main causes of construction projects failure in DARDLEA?
• What is the relative importance of the different factors causing project failures?
• What is the relationship among the responses of the different respondents?
• The following hypothesis was tested: • Ho: There is no correlation among the responses of respondents on the causes of construction project failures.• H1: There is a correlation among the responses of respondents on the causes of construction project failures.

LITERATURE REVIEW
Since availability of literature on the construction in the agricultural sector is limited, the literature reviewed covered construction projects failures in general.From the literature review, it was found that research on construction failures are generally done across industries but are often geographically specific.Despite the geographic locations, it was clear that common factors prevail towards project performance failure.In Saudi Arabia, only 30% of the construction projects were completed within schedule, while in Malaysia, 17.3% of the projects were considered "sick", meaning that the projects were delayed for more than three months or were abandoned (Sambasivan & Soon, 2006:518).It was also found that the most common cause of project failures in Saudi Arabia projects from contractor and consultant perspectives were changed orders with about 70% of the projects that experienced time overruns (Ikediashi, Ongunlana & Alotaibo, 2014:38).
According to Sweis, Sweis, Rumman, Hussein & Dahiyat (2013:114), the most dominant factors causing construction project failures in Jordan were incompetence, material price fluctuation, lack of experience of the contractor, incomplete drawings, government delays in decision making, inaccurate estimates, poor planning, and low labour productivity.
Due to rapid growth in construction activities, the causes for time and cost overruns in Dubai were attributed to tight construction schedules, unique architectural features, misalignment of international contractors and consultants, as well as unique culture and religious differences (Ren, Atout & Jones, 2008:756).
A study by Toor and Ongulana (2008:10) revealed that the causes of project delays or failures in Thailand were due to a lack of resources, poor contractor management, shortage of labour, design delay and inadequate planning.
During the literature review it was also found that some authors obtain views from, and in some cases, allocate causes of project failures to stakeholder categories.
A study was carried out in Benin by Akgobe, Feng andZhou (2012:1215) on the importance and ranking of factors causing delay for development construction projects.In their study, a questionnaire was used to collect data from contractors, owners, consultants and architects.Ten factors were found to be causing project failure, namely poor financial position of the contractor, financial difficulties by the owner, poor subcontractor performance, material procurement of contractor, changes in drawings of architect, inadequate planning and scheduling of contractor, slow inspection of completed work by the consultant, equipment availability of contractor, preparation, approval of drawings of consultant, and acceptance of inadequate drawings by consultant (Akgobe, et al., 2012(Akgobe, et al., :1218)).
During a study in Zambia, data were collected from consultants, subcontractors, contractors and clients on poor project performance.Poor cost etsimation and change orders were ranked as the most common causes of schedule overruns.It was also found that poor financial management was the most common and noticeable factor that caused quality shortfall (Muya, Kaliba, Chicombo & Shakantu, 2013:61).
In Egypt, the causes of project failure were clustered into material-related, owner-related, consultant-related, contractor or sub-contractor-related (Marzouk & El-Rasas, 2014:50).In Ghana, clients, contractors and consultants were key role players in their study of the causes of delay in the construction projects (Fugar & Agyakwah-Baah, 2010:104).
Baloyi and Bekker ( 2011) conducted a study into the cause of cost and time overruns of the 2010 Soccer World Cup.They identifiied three stakeholder categories, namely externallyrelated, client-related and consultant-related factors.In the study, the Relative Index Importance (RII) was used to analyse and rank the results (Baloyi & Bekker, 2011:59).

Research design
To effectively answer the research questions, the researcher used a descriptive case study involving a quantitative approach.The target population was purposely selected and included contractors, owners and clients who were involved in the construction, monitoring and utilisation of projects in DARDLEA.Such a target population was selected since it is believed that they have all the information related to construction projects' performance (Welman, Kruger & Mitchel., 2012:70).The sample size for the study was 44, eight being the client (included all project beneficiaries represented by chairpersons), eight contractors (referring to companies involved in the construction of projects in DARDLEA) and 28 representing the owner.Due to the fact that clients and contractors are difficult to locate, two were sampled from each of the four districts, namely Ehlanzeni South, Ehlanzeni North, Nkangala and Gert Sibande.Owners involved DARDLEA employees involved in the planning, monitoring and designing of construction projects.Seven employees from each of the four districts were involved.

Data collection
A questionnaire was developed using a 5-point Likert scale for data collection.The 5-point scale was selected to allow for neutral answers when selecting three and two ranges for negative and affirmative answering (Welman, et al., 2012:156-157).Respondents were categorised into client, owner and contractor.The questionnaire included 32 factors causing construction project failures to be ranked by the respondents.The questionnaire was distributed to the respondents via email.Telephone interviews were also carried out with respondents who did not react to the email notification.For client respondents, mostly telephone interviews were conducted since all had valid cell phone numbers with limited access to email due to remote geographical locations.The respondents were sampled from four districts, namely Gert Sibande, Nkangala, Ehlanzeni South and Ehlanzeni North.

Analysis
Descriptive statistics was used to analyse the collected data.Relative Importance Index (RII) was used to identify and rank the different causes of construction project failures as perceived by the respondents (Assaf & Al-Hejji, 2006:353;Baloyi & Bekker;2011:60;Sambasivan & Soon, 2006:523).The ranking of the causes was important for identifying and highlighting critical causes of construction project failures in DARDLEA.This would enable the department to focus their efforts in minimising such causes.
Relative Importance Index (RII) = ∑W/ (A*N) (1) Where W is the weighting given to each factor by the respondents (1 to 5), A is the highest weight (i.e. 5) and N is the total number of respondents.The RII enabled the researcher to cross compare the relative importance of the factors as perceived by the respondents (Assaf & Al-Hejji, 2006:353;Sambasivan & Soon, 2006:524).The RII was adapted to conform to the questionnaire being used.
To determine or measure the relationship about the direction and strength of the correlation among the different categories of respondents, a Spearman's rank Correlation was used (See equation (3) Where rs = Spearman's rank correlation coefficient, d= the difference in ranking of any two parties, n= number of factors (Assaf & Al-Hejji, 2006353).
If the correlation is between 1 and -1, where 1 refers to a perfect positive correlation (perfect direct relationship/agreement), -1 refers to a perfect negative correlation (perfect inverse relationship/disagreement).Figures near to zero indicate no relationship at all or little correlation while figures close to 1 indicate a good correlation (Assaf & Al-Hejji, 2006:523).

RESULTS
A total of 44 questionnaires were distributed with 32 returned.The response rates for each category is indicated in Table 3.The different factors causing construction project failure in DARDLEA were ranked according to the order of importance using the Relative Importance Index (RII) as shown in Table 4.
Table 5: Correlation among the contractor, client and owner p = 0.5 The results indicate the correlation values among respondents as follows: client-contractor (0.642), client-owner (-0.098) and owner-contractor (0.288).Client and owner disagreed on the causes of construction project failure in DARDLEA, while the contractor-client and contractor-owner agreed on the causes of construction failure in DARDLEA.In terms of the strength of the correlation, a strong positive correlation existed between the client and contractor (0.642).
It was concluded that there were differences in the responses of the respondents on the causes of construction project failure (p=0.05),therefore rejecting the null hypothesis.

CONCLUSIONS
Most construction projects in DARDLEA were never completed within budget, schedule and cost and some were abandoned before completion.The intention of the study was to investigate the causes of construction project failures in DARDLEA.The factors causing construction project failures were categorised into client-related, owner-related and contractor-related factors.In order to obtain the views of respondents on the factors causing construction project failure in DARDLEA, a questionnaire was developed and administered to different respondents working in the construction projects and those who were part of the project as clients or beneficiaries.The clients were subjected to a telephone interview, while the owners and contractors completed a similar questionnaire that was despatched to them via email.The study revealed that out of the 32 factors studied, the top ten factors which caused construction project failure in DARDLEA were poor communication, lack of monitoring and evaluation, lack of client/beneficiary involvement, financial difficulties, lack of project planning, poor project scheduling, incomplete project drawings, poor technical performance, changing client requirements, and incompetent workers.The correlation indicated that there was a weak negative correlation in responses between the client and owner.It can be concluded that there are significant differences in the responses amongst the clients, owners and contractors.
Key factors ranked highly by the respondents were lack of communication, lack of client or beneficiary involvement, lack of project monitoring and evaluation, and lack of project

Table 1 :
Project implementation challenges

Table 3 :
Response rate by respondents