Impact of the COVID-19 pandemic on risk factors in residential projects

ABSTRACT

The global economy was affected by the COVID-19 pandemic. This paper estimates the impact of the COVID-19 pandemic on the overall risk of construction projects in Egypt and identifies risk factors affected by the pandemic. A structured questionnaire was designed to collect data. One hundred thirty-two cases before the COVID-19 pandemic and 68 cases after the pandemic were managed. The outliers were excluded using Mahalanobis distance. The distribution of the samples before and after the COVID-19 pandemic was studied using the Kolmogorov-Smirnov test and the difference in medians using the Mann-Whitney U test. The delay in payment and the fluctuation in material prices are the most critical risk factors after the epidemic spread in residential projects in Egypt. Suspension of projects has been the most significant consequence of the COVID-19 pandemic, and the pandemic increased the overall risk by 17%. This research helps to identify the main risk factors after the pandemic, and then the project manager can prepare an appropriate risk response plan.

GRAPHICAL ABSTRACT

KEYWORDS:

• Residential projects
• Risk-factors
• COVID-19 pandemic
• suspension
• delays of payments

1. Introduction

The World Health Organization considered the COVID-19 the third pandemic in the twenty-first century (World Health Organization 2020). The Covid-19 pandemic has led to major lockdowns in most countries, resulting in a significant reduction in economic activities. Unfortunately, COVID-19 has an enormous negative impact on gross domestic product (GDP).

A risk is an event or condition that, if it occurs, has a positive or negative effect on one of the anticipated objectives (Malik et al. 2019). The construction activities have unique features, such as long period projects, including complicated processes, financial intensity, abominable environment, and dynamic organizational structures (Taylan et al. 2014). Therefore, construction projects are riskier than other projects (Luo et al. 2015). It is necessary to reduce the negative impact of risks on projects, and this is done by investigating and managing risk factors before they occur (Park et al. 2016). There is a need to focus on the main risks, as identifying and assessing all risks takes a lot of effort and money without much impact on the project’s overall risk and may lead to adverse results (Zayed, Amer, and Pan 2008). Risk assessment uses analytical tools to forecast their probabilities and severities in advance and place risks in an appropriate category to develop strategies to deal with risk factors (Taylan et al. 2014). If risks are not analyzed well and effectively or strategies are not trained to handle them, the project may fail. The risk assessment process can be done through quantitative or qualitative methods (Mahendra, Pitroda, and Bhavsar 2013). Familiarity with risk management processes leads to actions to reduce their adverse effects.

The average of delayed projects due to the COVID-19 pandemic was 21% (Fairlie 2020). The Covid-19 pandemic increased the cost of materials by 20% in Oman construction projects and lowered oil prices, thus cutting the supply chain in construction projects (Al Amri 2021). 30% of construction workers have stopped working on construction sites due to fear of the spread of the Corona epidemic in India. The pandemic also delayed the schedule due to a drop in production rate by 20% from pre-COVID-19 levels (Laing 2020). The coronavirus pandemic is costing India’s construction industry up to 30,000 Rs per day; hence the investment in the construction industry was reduced by 13 to 30 %. Whereas China, as a result of the lockdown resulting from the outbreak of the Coronavirus, all construction works were stopped leading to losses of 10 million yuan in the construction industry. Whereas in the UK the construction industry is losing £ 301.5 m a day due to the pandemic (Biswas et al. 2021).

Investigation of the influence of the COVID-19 pandemic on risk factors in building projects, mainly residential constructions, is lacking. There is a gap in studies on the changing risk factors resulting from the Corona epidemic, particularly in developing nations like Egypt. This research aims to re-analyze the risk factors before and after the coronavirus to estimate the outbreak’s impact on risk factors in the construction of residential projects in Egypt. The research also aims to forecast the epidemic’s effect on the reduction in productivity, project suspension, and overall project risk.

2. Literature review

2.1. Risk factors before COVID-19

Numerous studies identified and classified the risk factors in groups. For example, Shen, Wu, and Ng (2001) performed a risk analysis based on 54 responses in China. The risk factors were classified into financial, management, legal, political, market, and technical groups. Another study was performed in Saudi Arabia, where the risk factors were divided into risks related to; time, cost, safety, quality, and environment. The key risk factors were excessive approval procedures, absence of advanced planning studies, tight project schedules, delays in payments, and design changes (Taylan et al. 2014). Odimabo, Oduoza, and Suresh (2017) used a risk matrix to prioritize the risk factors, then developed a Bayesian belief network to estimate the strength of the relationship between the risk factors. The main risks were improper construction methods, late payment, changes in the prices of materials, and poor communication between the parties.

In Egypt, 65 risk factors for construction projects have been identified and categorized into internal and external groups. The main risk factors were the changes in currency rates, changes in tax rates, fuel shortages, and fire risk. (Khodeir and Mohamed 2015). Another study for identifying and assessing the risk factors in construction projects in Egypt depended on data obtained from sixteen companies. The risk factors were grouped into four main groups; site conditions, project parties, resources, and project features (Abd El-Karim, Mosa El Nawawy, and Abdel-Alim 2017). Osman, Issa, and Zakaria Eraqi (2020) identified eighty-one risk factors that cause cost overruns or delay the schedule in non-residential projects in Egypt. The analysis was performed using the probability and impact indices. The key risk factors were the changes in material prices, a direct attribution method, and a lack of suitable funds for the project. Badawy (2021) identified the causes of the risk of change orders in road projects using the Delphi technique. The data were analyzed using a structural equation model. The critical factor of change orders was the change in design.

2.2. Impact of COVID-19 on risk factors

An online questionnaire was performed to determine the extent to which participants in the Chinese construction industry are aware of the coronavirus pandemic and the appropriateness of current measures to counter the risk of the spread of the coronavirus among workers (Zheng, Chen, and Ma 2021). The effect of COVID-19 on the supply chains of the construction projects was studied depending on 71 questionnaires. They found that the COVID-19 affected the workflow and led to disruption in the supply chain (Ogunnusi et al. 2020). Laing (2020) declared that the primary risks after the COVID-19 pandemic are the closure of the project, shortage of materials and labor, and the change in construction methods to accommodate social distancing guidelines. The impacts of COVID-19 on the Malaysian construction industry included project suspensions, time and cost overruns, and job losses (Gamil and Alhagar 2020). The pandemic has adverse effects on supply chains in construction projects (Fernandes 2020). The significant negative impacts of the COVID-19 pandemic on the construction industry in the United States were project delays, supply chain disruption, drop-in productivity, and increased material prices (Alsharef et al. 2021).

After COVID-19 became a part of our daily lives, Amoah and Simpeh (2021) declared several preventive procedures for construction stakeholders to help reduce COVID-19 safety implementation obstacles to minimize the diffusion of disease among laborers. To reduce the impact of COVID-19 on construction sites, some construction companies have adopted basic measures, including restricting site access, activating the social distancing protocol, and emphasizing the importance of hygiene (Salami, Ajayi, and Oyegoke 2022). Authors from 27 countries participated in research related to the Coronavirus and its impact on the construction sector. The most significant number of authors was from the United States of America, followed by China and then South Africa. Unfortunately, Egypt was not one of these countries, indicating a gap in research on the effect of COVID-19 on the construction sector in Egypt (Ayat, Malikah, and Kang 2021). Egypt has a significant movement in the construction industry, where many projects have been implemented. The major construction boom began in 2014 and reached its peak in 2018. Cairo’s new capital was established, and thousands of roads were built. Construction projects faced risk factors, which led to cost overruns due to the lack of risk management processes in these projects and the impact of the pandemic, where the construction industry has been affected by the occurrence of the COVID-19 pandemic. Therefore risk assessment models should include a careful analysis of the consequences of the epidemic for future projects.

There is a gap in research studies on the impact of the COVID-19 pandemic on risk factors in construction projects, especially in residential projects. In other words, what are the risk factors that have increased or decreased as a result of the Corona pandemic, especially in developing countries such as Egypt? This study aims to rank the risk factors before and after the Covid-19 pandemic to determine the epidemic’s impact on risk factors in the construction of residential projects in Egypt. The study also includes the impact of COVID-19 on three measurements of project performance, namely, the reduction in productivity, project suspension, and overall project risk.

3. Methodology

The methodology in this study consists of three parts. The first part identified the risk factors affecting the project cost and schedule from an extensive literature review. The second part concerns data collection, where a structured questionnaire was designed to collect data on the impact of each risk factor on the project’s cost and schedule if the risk occurs. The Delphi technique was performed to reach a consensus among experts about the final list of the risk factors.

The third part of the methodology relates to data analysis, where data have been classified into two groups; pre-COVID-19 and post-COVID-19. On 11 March 2020, the World Health Organization (WHO) declared COVID-19 a global pandemic. Hence, any residential project that was started and finished before that date was considered a sample in the pre-COVID-19 set. At the same time, the post-COVID-19 set includes data collected from residential projects that were started and completed after that date. The projects that started before the pandemic and finished during the pandemic were excluded from this study. The abnormal cases were excluded before data analysis using the Mahalanobis distance; then, it was investigated whether the two groups followed the normal distribution. Then, it was checked whether the two groups were drawn from the same population or not using the Kolmogorov-Smirnov test. Finally, the Mann-Whitney U test was used to validate whether there is a difference in the median of each risk factor between the two groups or not. The difference between the medians between the two sets indicates the impact of the COVID 19 pandemic on the risk factor. The methodology of the research is illustrated in Figure 1.

Figure 1. The methodology of research.

4. Identifying the risk factors

Based on the previous literature, 33 risk factors affecting residential projects’ costs or schedules have been identified. Five experts were selected with at least 15 years of experience in project management. After three rounds, the Delphi technique was used to reach a consensus among the experts; experts were asked to add any significant missing risk factors in the first round. The responses were that no risk factor was missing. Responses were sent back to experts to determine the probability and severity of each factor on a three-point Likert scale. The risk score for each risk factor was computed by multiplying the likelihood and severity, and then the average risk score for each risk factor was estimated. Then any factor with an average low-risk score was excluded from the list. Therefore, seven risk factors were excluded; hence, the second list included only 26 factors. The seven excluded risk factors were: poor material purchases, pre-qualification of contractors, number of subcontractors, poor stockpiling of materials, changes in rules and regulations, poorly designed contract forms, and war and revolutions. The experts received the second list and were asked if they agreed that the second list constituted the most important risk factors or not. The responses were that the second list contains the most critical risk factors in residential construction projects in Egypt, which means that the second list can be considered the final list of risk factors. The risk factors in the final list were sorted alphabetically with their references, as shown in Table 1.

Table 1. Critical risk factors.

I.D.	Risk factor	References
R01	Adverse weather conditions	(Alarcón et al. 2011), (Mahamid 2011), (El-Sayegh 2008)
R02	The fluctuation in material prices	(Osman, Issa, and Zakaria Eraqi 2020), (Marzouk and Aboushady 2018), (Odimabo, Oduoza, and Suresh 2017), (Aboshady, Elbarkouky, and Marzouk 2013)
R03	Bribes and corruption	(Marzouk and Aboushady 2018), (Hashemi et al. 2013), (Choudhry and Iqbal 2013), (Baghdadi and Kishk 2015)
R04	Change in construction methods	(Abd El-Karim, Mosa El Nawawy, and Abdel-Alim 2017), (Zayed, Amer, and Pan 2008), (Tang et al. 2007)
R05	Change in tax rates	(Khodeir and Mohamed 2015), (Abd El-Karim, Mosa El Nawawy, and Abdel-Alim 2017)
R06	Conflict in contract documents	(Banik and May 2006), (Odeh and Battaineh 2002), (Tran and Molenaar 2014)
R07	Delay in obtaining licenses from the authorities	(Taylan et al. 2014), (Park et al. 2016), ((El-Sayegh and Mansour 2015)
R08	Delay in payments	(Eskander 2018), (Taylan et al. 2014), (Mahamid 2011), (Kartam and Kartam 2001)
R09	Delays in the approvals of submittals	(Eskander 2018), (Marques and Berg 2011), (El-Sayegh and Mansour 2015)
R10	Difficulty in site access	(Abd El-Karim, Mosa El Nawawy, and Abdel-Alim 2017)
R11	Exchange rate fluctuations	(Eskander 2018), (Marzouk and Aboushady 2018), (Xu et al. 2018), (Osman, Issa, and Zakaria Eraqi 2020)
R12	Fire risk	(Khodeir and Mohamed 2015)
R13	Inadequate claim administration	(Kartam and Kartam 2001), (Alarcón et al. 2011)
R14	Inadequate safety procedures	(Elsebaei et al. 2020), (Venkataraman 2008), (Tang et al. 2007), (Amoah and Simpeh 2021), (Salami, Ajayi, and Oyegoke 2022)
R15	Increase in labor salaries	(Marzouk and Aboushady 2018), (Aboshady, Elbarkouky, and Marzouk 2013)
R16	Inefficient planning	(Choudhry and Iqbal 2013), (El-Sayegh and Mansour 2015), (Hashemi et al. 2013)
R17	Insecurity of the site	(Hashemi et al. 2013), (El-Sayegh 2008)
R18	Insufficient technology, skills, or techniques	(Tang et al. 2007), (Hashemi et al. 2013), (Chan et al. 2011)
R19	Oil prices spike	(Xu et al. 2018), (Khodeir and Mohamed 2015)
R20	Owner change orders	(Badawy 2021), (Marques and Berg 2011), (Choudhry and Iqbal 2013), (Tran and Molenaar 2014)
R21	Poor coordination	(Abd El-Karim, Mosa El Nawawy, and Abdel-Alim 2017), (Choudhry and Iqbal 2013), (Odimabo, Oduoza, and Suresh 2017), (Osama, Sherif, and Badawy 2021)
R22	Quality and integrity of the design	(Eskander 2018), (Creedy, Skitmore, and Wong 2010), (El-Sayegh and Mansour 2015)
R23	Supply chain disruption	(Mahamid 2011), (Kartam and Kartam 2001), (Zayed, Amer, and Pan 2008), ((Osei-Kyei and Chan 2017), (Salami, Ajayi, and Oyegoke 2022)
R24	The financial failure of the contractor	(Park et al. 2016), (Eskander 2018), (Osama, Sherif, and Badawy 2021)
R25	Third-party liability risk	(Tran and Molenaar 2014), (Chan et al. 2011), (Banik and May 2006)
R26	Unexpected underground utilities	(El-Sayegh and Mansour 2015), (Creedy, Skitmore, and Wong 2010), (Zayed, Amer, and Pan 2008)

5. Data gathering

In this paper, the use of interviews and questionnaires to collect data was relied on due to the suitability of these methods in investigating a new topic that was not well identified in the previous literature (Ayat, Malikah, and Kang 2021). A structured questionnaire was developed to collect residential construction project data through interviews. The questionnaire contains four parts. The first part includes the general personal information of the respondent: name, job title, academic certificate, and years of experience. The second part provides available information about the project’s name and location. Then, the third part includes the effect of risk factors that have already occurred in the project on the cost and schedule. Finally, the fourth part concerns the impact of COVID-19 on projects in three dimensions: reduction in productivity, suspension of the project, and overall risk. A three-point Likert scale was used to facilitate respondents’ answers. The classifications of the impact of each risk factor are shown in Table 2.

Table 2. The classifications of risk factor.

Rank	Classification	Impact on cost	Impact on time
1	low	< 10%	<1 month
2	Moderate	10% – 20%	1–3 months
3	high	> 20%	> 3 months

While the population in this research is the entire residential projects established during the year 2018 until March 2020 in Egypt and the residential buildings constructed between March 2020 and March 2021, which is an unknown number; therefore, the population was considered unlimited. The sample size can be estimated using Equation (1)(1) $n=\frac{Z^2\ast P\ast \left(1-P\right)}{C^2}$(1) . Where “n” represents the sample size, “Z” represents the critical value at 95% confidence level, which is 1.96, “P” represents the sample proportion which was 0.5, and “C” represents the margin of error (Dhar, Binu, and Mayya 2014). In this paper, data were collected from 200 residential projects. By substitution in Equation (1)(1) $n=\frac{Z^2\ast P\ast \left(1-P\right)}{C^2}$(1) , the margin of error was 0.07.

(1) $n=\frac{Z^2\ast P\ast \left(1-P\right)}{C^2}$(1)

The questionnaire was conducted through personal interviews with experts working on various residential construction projects in Egypt. 200 questionnaires were collected. One hundred thirty-two questionnaires were received from residential construction projects that started and finished before the Corona pandemic. The remaining 68 questionnaires were received from residential construction projects that started and were completed after the Corona pandemic. The characteristics of the respondents of the questionnaires are shown in Table 3.

Table 3. The classification of respondents.

Experience	No.	%	Belong to	No.	%
5 to 10	29	14.5%	Contractor	88	44%
10 to 15	55	27.5%	Sub-Contractor	54	27%
> 15	116	58%	Consultant	36	18%
			Client	22	11%

The general description of the residential buildings was classified into four categories; floor area, number of floors, number of basements, and total project duration. The general description of the residential buildings was classified into four categories; floor area, number of floors, number of basements, and total project duration. The floor area was classified into less than 200 m², from 200 to 400 m², from 400 to 600 m², and more than 600 m². The number of floors category was classified into four groups one or two floors, three to five floors, six to eight floors, and more than eight floors. The basement category was divided into no basement and one basement. The total project duration category was divided into four groups up to six months, from 6 to 12 months, 12 to 24 months, and more than 24 months. The statistical data on residential construction projects in Egypt that were used in this research was illustrated in Table 4.

Table 4. Statistical data of residential projects.

floor area	No.	number of floors	No.	total project duration	No.	number of basements	No.
less than 200 m^2	31	one or two floors	37	up to six months	33	no basement	53
from 200 to 400 m^2	56	three to five floors	53	from 6 to 12 months	46	one basement	147
from 400 to 600 m^2	66	six to eight floors	55	from 12 to 24 months	62
more than 600 m^2	47	more than eight floors	55	more than 24 months	59

6. Data analysis

Four data analysis processes were performed. The first process concerns the exclusion of outliers using the Mahalanobis distance. The second process relates to examining whether the two groups’ distribution is the pre-COVID-19 and the post-COVID-19 following the normal distribution or not using the Kolmogorov-Smirnov test. The third process is to verify whether the two samples were drawn from the same population using the Kolmogorov-Smirnov test. The fourth process is to check whether there are significant differences in the medians between the two groups or not, using the Mann-Whitney test.

Outliers must be removed before beginning the data analysis process to deduce a good analysis. A Mahalanobis distance was used to determine abnormal cases. Mahalanobis distance deduces the distance between a series of data and a specific point and can be estimated from Equation (2)(2) $D=\sqrt{{\left(x-m\right)}^T.C^{-1}.\left(x-m\right)}$(2) based on the number of independent variables in the model. The maximum Mahalanobis distance can be calculated at a critical alpha value of 0.001. Hence, any data with a distance more than the Mahalanobis distance is considered an outlier and should be removed from the data set (Elfadaly, Garthwaite, and Crawford 2016). Data were collected for 200 residential construction projects. Using a Mahalanobis distance to remove outliers, the probabilities of 11 projects were less than 0.001; therefore, these 11 projects were excluded from the analysis. Hence, the final number of residential construction projects was 189, divided into 121 cases that were pre-COVID-19 and 68 cases the post-COVID-19.

(2) $D=\sqrt{{\left(x-m\right)}^T.C^{-1}.\left(x-m\right)}$(2)

“D” denotes the Mahalanobis distance, “x” denotes the specific case, “m” denotes the average of independent variables, and “C⁻¹ “denotes the inverse covariance matrix of independent variables.

The one set Kolmogorov-Smirnov test was performed separately on pre-COVID-19 cases and post-COVID-19 cases to test if the sample before and after the pandemic followed the normal distribution or not. This test depends on the maximum absolute difference between the cumulative sample and the cumulative normal function distribution “D,” which can be calculated from Equation (3)(3) $D=Max{.}_X\left|F_n\left(x\right)-N_n\left(x\right)\right|$(3) (Fasano and Franceschini 1987). All factors have P-values less than 0.05, which means they do not follow the normal distribution; hence, non-parametric tests should be applied.

(3) $D=Max{.}_X\left|F_n\left(x\right)-N_n\left(x\right)\right|$(3)

To test if the sample before the COVID-19 pandemic and the sample after the COVID-19 pandemic were drawn from a population with the same distribution or not, A non-parametric test for two independent samples Kolmogorov-Smirnov test was performed. This test depends on the maximum absolute difference between the cumulative of the two samples, D, which can be calculated from Equation (4)(4) $D=Max{.}_X\left|F_n\left(x\right)-G_m\left(x\right)\right|$(4) . The critical value of “D” can be estimated from Equation (5)(5) $D_{cr}=c\left(\alpha \right).\sqrt{\frac{\left(n+m\right)}{n.m}}$(5) (Fasano and Franceschini 1987). The fluctuation in material prices, delay in obtaining licenses from the authorities, inadequate safety procedures, oil prices spike, and the supply chain disruption did not belong to the same population. The overall risk, reduction in productivity, and the project’s suspension were also not extracted from the same population, as shown in Table 5.

(4) $D=Max{.}_X\left|F_n\left(x\right)-G_m\left(x\right)\right|$(4)

(5) $D_{cr}=c\left(\alpha \right).\sqrt{\frac{\left(n+m\right)}{n.m}}$(5)

Table 5. The risk factors were not drawn from the same population.

ID	Description	Most Absolute Extreme Differences	Kolmogorov-Smirnov Z	Sig. (2-tailed)
R02	The fluctuation in material prices	0.257	1.696	0.006
R07	Delay in obtaining licenses from the authorities	0.242	1.597	0.012
R14	Inadequate safety procedures	0.261	1.719	0.005
R19	Oil prices spike	0.210	1.386	0.043
R23	Supply chain disruption	0.330	2.175	0.000
I01	Reduction in productivity	0.263	1.732	0.005
I02	Suspension of the project	0.476	3.138	0.000
I03	Overall Risk	0.224	1.478	0.025

Where “n” and “m” are the two-samples sizes 68 and 121 in this research, respectively, at a 95% level of confidence, the corresponding value for α is 0.05, and the value of c(α) is 1.36.

6.1. Median comparison

A Mann – Whitney U test was performed to test whether there is a significant difference between medians of the data before and after the COVID-19 pandemic for overall risk and different risk factors. The differences between the two independent groups could be compared using the Mann-Whitney U test if the data met the following four assumptions: the dependent variable should be ordinal or continuous. In this study, the dependent variable “ risk factor” was measured on a three-point Likert scale, so this assumption is valid; the independent variables should consist of two categories. In this study, the independent variables are classified into two groups; the pre-COVID-19 or the post-COVID-19, which means that the independent variable meets the second criterion. Observations should be independent, which was achieved in this study. Finally, the two variables do not follow normally distributed, which was achieved according to this research’s second data analysis process.

Suppose the two-set drawn from the same population. In that case, the null hypothesis is that the distributions of the two groups are equal (Ho), and the alternative hypothesis is that the medians of the two groups are not equal (H1). If the two-set were not drawn from the same population, then the null hypothesis is that the distributions of scores of the two groups are equal (Ho), and the alternative hypothesis is that the mean ranks of the two groups are not equal (H1). The Mann-Whitney value can be calculated using Equation (6)(6) $\mathrm{U}=N_1\ast N_2+0.5\ast N_1\ast \left(N_1+1\right)-{\sum }_{i=1}^{i=N_1+N_2}{R}_{\left(i\right)}$(6) .

(6) $\mathrm{U}=N_1\ast N_2+0.5\ast N_1\ast \left(N_1+1\right)-{\sum }_{i=1}^{i=N_1+N_2}{R}_{\left(i\right)}$(6)

Where “U” represents the Mann-Whitney value of the first set, “N₁” & “N₂” represent the sample size of the first and second set, respectively, and R_(i) represents the rank of factor (i).

The P-values of the overall risk, the suspension of the project, and four risk factors were less than 0.05; hence, they were not drawn from the same population; therefore, the alternative hypothesis “the median ranks of the two groups are not equal” was accepted. These four risk factors are the fluctuation in material prices, delay in obtaining licenses from the authorities, inadequate safety procedures, and supply chain disruption. The risk factor “Increase in labor salaries “was drawn from the same population and had p-values less than 0.05. Thus, the alternative hypothesis that “the medians of the two groups are not equal” was accepted.

The oil price spike and the reduction in productivity were not extracted from the same population. They had p-values more than 0.05; hence the null hypothesis “the distributions of scores of the two groups are equal” was accepted. The remaining twenty risk factors were drawn from the same population and had P-values more than 0.05 hence the null hypothesis that “the distribution of the two groups is equal.” Risk factors for which the alternative hypothesis to the Mann-Whitney test has been accepted are shown in Table 6.

H₁: The mean ranks of the two groups are not equal.

Table 6. The results of the Mann-Whitney test.

ID	Description	Mann- Whitney U	Sig. (2-tailed)	Decision
R02	The fluctuation in material prices	3337	0.022	Accept H1
R07	Delay in obtaining licenses from the authorities	3279	0.015	Accept H1
R14	Inadequate safety procedures	3249	0.013	Accept H1
R15	Increase in labour salaries	2889	0.000	Accept H1*
R23	Supply chain disruption	3401	0.035	Accept H1
I02	Suspension of the project	2602	0.000	Accept H1
I03	Overall risk	3100	0.001	Accept H1

H₁*: The medians of the two groups are not equal.

Table 7 represents the significant difference between the risk factors before and after the pandemic; the risk factor “Suspension of the project” had the highest difference, which means this factor needs strong strategic plans to overcome its negative impact, followed by the “Increase in the labor salaries” with difference factor of 0.55. The “Inadequate safety procedures” and delay in obtaining licenses from the authorities “come next with a risk difference of 0.49 and 0.42, respectively. Finally comes the risk factors of “The fluctuation in material prices” and “Supply Chain Disruption” with risk differences around the range of 0.3. The authors aim this risk distribution approach to provide a feasible reference for preparing the different mitigation plans within the different construction projects.

Table 7. Impact of the COVID-19 pandemic on risk factors.

ID	Description	Before COVID-19	After COVID-19	Difference	Ratio
R02	The fluctuation in material prices	1.99	2.32	0.33	17%
R07	Delay in obtaining licenses from the authorities	1.11	1.53	0.42	38%
R14	Inadequate safety procedures	1.50	1.99	0.49	33%
R15	Increase in labour salaries	1.30	1.85	0.55	42%
R23	Supply chain disruption	1.89	2.16	0.27	14%
I02	Suspension of the project	1.65	2.37	0.72	44%
I03	Overall Risk	1.87	2.18	0.31	17%

The ranking and the mean of the top risk factors before and after the pandemic were presented in Table 8. “Delay in payments” risk factor is the top-ranked risk factor before and after the pandemic due to its strong impact on the construction projects. Still, its mean increased after the pandemic, which shows the effect of the situation. Due to the pandemic, the availability of new construction projects decreased, so the financial capability of the contractors became more stable, which justifies why the “The financial failure of the contractor” risk factor moved to the third place after the pandemic. The second factor was replaced by “The fluctuation in material prices” because this pandemic affected the whole supply market, including manufacturing progress, trading, and exchange; it caused a freezing condition, increasing the scarcity of raw materials. The “Owner change orders” risk factor moved from the third place to the fourth place due to the slow down situation caused by the pandemic, so the change orders decreased as the projects’ progress decreased; the same situation happened to the fourth risk factor before the pandemic, the “Delays in the approvals of submittals”, as the projects’ progress decreases, the significance of this factor also decreased. Finally, the “Exchange rate fluctuations” factor disappeared from the list after the pandemic due to the extreme slowdown movement in the financial and trading market.

Table 8. The top risk factors before and after the COVID-19 pandemic.

	Before		After
Rank	Factor	Mean	Factor	Mean
1	Delay in payments	2.37	Delay in payments	2.34
2	The financial failure of the contractor	2.29	The fluctuation in material prices	2.32
3	Owner change orders	2.23	The financial failure of the contractor	2.26
4	Delays in the approvals of submittals	2.21	Owner change orders	2.22
5	Exchange rate fluctuations	2.02	Delays in the approvals of submittals	2.18

Before the pandemic, the first risk factor was the delay in payments, which had a mean value of 2.37, followed by the contractor’s financial failure, which had a mean value of 2.29. The third risk factor was the owner change orders which had a mean value of 2.23, followed by the delays in the approvals of submittals which had a mean value of 2.21. The “exchange rate fluctuations” factor was in fourth place with a mean value of 2.02. After the pandemic, the top risk factor was the delay in payments with a mean value of 2.34; the second risk factor was the fluctuation in material prices with a 2.32 mean value. The third risk factor was the financial failure of the contractor, which had a mean value of 2.26, followed by owner change orders which had a mean value of 2.22.

6.2. Discussion

The COVID-19 pandemic has led to significant changes in the average of five risk factors out of the 26 risk factors studied. The most affected risk factor by the COVID-19 pandemic was “the increase in labor salaries,” with a difference in means of 0.55. The COVID-19 pandemic has increased the average risk factor “the fluctuation in material prices” from 1.99 before the pandemic to 2.32 after the pandemic. The “delay obtaining licenses” risk factor increased from 1.11 before the pandemic to 1.53. The COVID-19 pandemic increased the average risk factor “Inadequate safety procedures” from 1.50 to 1.99, while the factor “supply chain disruption” was changed from 1.89 before the epidemic to 2.16 after the pandemic.

Risk analysis to evaluate the Arab construction projects was conducted utilizing the analytical hierarchy process. The factor of insufficient client funds, which will lead to a delay in payment, was the most critical risk factor among 37 factors (Eskander 2018). The schedule delay due to the late payment factor ranked fourth in research to assess risks in construction projects in Saudi Arabia using fuzzy logic (Taylan et al. 2014). In a study to classify the risk factors causing time delays in the construction of road projects in the West Bank in Palestine, 43 risk factors were classified into three groups; high-risk, medium-risk, and low-risk. The delay in payments was one of the factors in the high-risk group (Mahamid 2011). In this study, the delay in payments was the top risk factor.

The increase in the prices of raw materials came in third place in research to model risks using expected monetary value and fuzzy logic in real estate development projects in Egypt (Marzouk and Aboushady 2018). A study to determine the impact of risk on cost and schedule for construction of non-residential projects in Egypt ranked the increase in the cost of raw materials as the most important risk factor (Osman, Issa, and Zakaria Eraqi 2020). Another study for assessing the risk factors using Bayesian Belief Networks declared that the fluctuation in material prices is the second top factor out of 27 risk factors that affect the cost overrun construction projects in Nigeria (Odimabo, Oduoza, and Suresh 2017). In this study, the fluctuation in material prices was ranked the second top risk factor.

Delayed payments, the financial failure of the contractor, owner-change orders, and delays in the approvals of submittals were ranked among the top five risk factors both before and after the COVID-19 pandemic. While the factor of the fluctuation in material prices was ranked sixth before the COVID-19 pandemic, it has become the second most important risk factor after the COVID-19 pandemic.

The suspension of project activities is the factor most affected by road and transport projects in Australia and Asia due to the COVID-19 pandemic (Isa et al. 2021). The primary risks in projects after the COVID-19 pandemic are the closure of the project and shortage of materials and labor. Another research to investigate the impact of COVID-19 on construction projects illustrated that the suspension of the project was the top effect of COVID-19 on construction projects (Gamil and Alhagar 2020). This research shows that the emergence of the COVID-19 pandemic led to an increase in construction project suspension by 44%, with a difference in averages of 0.71, and the impact on the project’s overall risk was increased by 17%.

The research was executed to analyze and rank the risk factors after COVID-19 on building residential projects in Egypt. This research gives a better insight into the current status of risk factors after COVID-19 and highlights the significant change in some of the risk factors that require attention from project managers. The research also analyzes the impact of COVID-19 on project performance through three dimensions to better understand the project performance after the COVID-19 pandemic and assist stakeholders in decision-making in future crises.

7. Conclusion

Residential projects are affected by many risks. The emergence of the Covid-19 virus has significantly impacted some risk factors in housing projects, especially in developing countries such as Egypt. This research aims to study the impact of the COVID-19 pandemic on risk factors. In addition to investigating the epidemic’s effect on the overall risk factor, productivity reduction, and project suspension. Thirty-three risk factors affecting residential construction projects were identified in the initial list of risk factors. The experts reached the consensus after three rounds using the Delphi technique, and accordingly, the final risk factors list consisted of 26 risk factors. A questionnaire was designed to collect data from real projects. Data were collected from 200 residential construction projects: 132 projects before the COVID-19 and 68 after the pandemic. Three tests were performed to check the normality of the distributions, verify whether the two samples were drawn from the same population, and check whether there are significant differences in the medians. The Mahalanobis distance was used to exclude the abnormal cases; hence 11 projects were excluded for more accurate results.

Five risk factors have significant changes in the medians due to the COVID-19 pandemic. These five factors were the fluctuation in material prices, delay in obtaining licenses from the authorities, inadequate safety procedures, Increase in labor salaries, and supply chain disruption. After the pandemic, there was a significant change in the risk factor “material price volatility,” as it was classified as the second risk factor. COVID-19 pandemic increased the overall risk in residential projects and increased Egypt’s project suspension.

8. Limitation of research

A risk is an event or condition that, if it occurs, will have a positive or negative effect on one of the anticipated objectives (Malik et al. 2019). Project objectives include cost, schedule, quality, and scope. This research is limited to risk factors that, if they occur, would harm the price or schedule of the construction project. Data were collected from residential projects with fixed-price contracts. These projects were located in Egypt only. This study is limited to investigating the risk factors identified in this research. Since not all risk factors can be analyzed, there is always a management reserve. Hence, some risk factors have not been taken into account, such as the risk of infection. This research is based on a qualitative risk analysis. Therefore, the comparison of risk factors was based on average responses using a three-point Likert scale to determine the probability and severity of each factor, which depends on the subjectivity of the respondents.

Availability of data and material

Some or all data, models, or codes that support the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgments

The authors extend their appreciation to the Research Supporting Project number (RSP-2021/264), King Saud University, Riyadh, Saudi Arabia, for funding this work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Research Supporting, King Saud University, Riyadh, Saudi Arabia [RSP-2021/264].

Notes on contributors

Mohamed Badawy

Dr. Mohamed Badawy finished his B.Sc. in the Civil engineering department, Cairo University, Cairo Egypt in 1995. He completed his M.Sc. in the civil engineering department, Ain Shams University, Cairo Egypt in 2005. His Ph.D. in the civil engineering department, Ain Shams University, Cairo Egypt in 2010 He is currently an associate professor at Ain Shams University in Construction Engineering and Project Management. He participated in 20 published papers until 2022.

Fahad K. Alqahtani

Dr. Fahad K. Alqahtani obtained his BS in Civil Engineering in 2007 from King Saud University in Riyadh, Kingdom of Saudi Arabia. In 2009 he joined the University of Birmingham where he obtained his MSc in Construction Management in 2010. He then continued at the same university, where he obtained his Ph.D. in 2017. He is currently an assistant professor at King Saud University in the field of Engineering Sustainability, Construction Engineering, and Project Management

Mohamed A. Sherif

Mohamed Sherif is a research and teaching assistant in the civil and environmental engineering department, the University of Hawaii at Manoa, USA.