Analysis and Evaluation of the Influence of Selected Factors on the Occurrence of Defects in Polish Housing Construction Using the Example of the Lower Silesia Region

Abstract

In literature relevant to this topic, attention is mainly paid to the qualitative and quantitative identification of defects in housing construction, and the factors that cause these defects. There is a research gap regarding the quantitative relationships between factors and defects, and the identification of factors that have a decisive impact on the occurrence of defects. The authors’ contribution to research regarding quality management in construction investments involves the identification of defects in residential buildings, identification of factors that generate construction defects occurring at various stages of the investment process, and also the assessment of their discriminatory power. This analysis used the results of technical inspections of buildings carried out in Poland in 2017–2020 in the Lower Silesia region. The study of the factors that influence quality in housing construction was carried out using the diagnostic survey method and the survey technique. Discriminant analysis was used for the calculations, with a number of influence factors being found. The following factors have the greatest discriminatory power: C1—a lack of internal control of design documentation before the start of the construction of the facility; C15—a lack of stability of the team (high staff turnover) that conducts contract tenders; C30—a lack of executive potential for preparing the facility for technical acceptance. Identifying the relationships between factors and quality, measured by the number and type of defects, will constitute the basis for developing procedures for conducting and controlling construction works and taking appropriate preventive actions in the form of employee training.

1. Introduction

The investment process in the construction industry is multi-stage and complicated. At its individual stages, disruptions may occur, which in turn translates into there being a high investment risk and a low quality of completed building objects. Achieving a high level of quality in a residential building, characterized by the highest level of performance features, is crucial for the client and all participants in the investment process. Therefore, it is reasonable to find the answer to the question of which factors generated during the individual stages of the construction process affect the final quality of the building, and also what impact these factors have on the formation of defects in residential housing. Researchers, dealing with the issue of housing quality in various countries of the world, have identified a number of recurring defects in residential building elements [1,2,3]. The subject of analysis has also been the factors influencing the formation of defects identified at various stages of the investment process [4,5]. However, to date, the extent to which individual factors influence the formation of defects has not been defined. This research area represents a research gap, the filling of which will contribute significantly to improving quality in residential construction.

The identification of negative factors that occur in the investment process, and the assessment of their impact on the quality of the facility, provide important knowledge that gives the basis for controlling the course of the process. As a result of taking appropriate corrective actions, these factors can be eliminated in the future, which in turn will allow a facility of the highest quality to be obtained.

The aim of the research was to identify the defects in residential buildings, the factors that affect the quality in residential buildings which occur at various stages of the construction process, and also to assess the discriminatory power of construction defects using the identified factors. The undertaken research will constitute the basis for developing procedures for conducting and controlling construction works. They will be beneficial for investors, construction entrepreneurs, site managers, designers, and supervisory inspectors, but above all for customers who will receive a product of the highest quality.

2. Literature Review

Quality means a set of product and service features that subjectively create the value of a product, which in turn affects the fulfilment of the customer’s expectations. It can be unequivocally stated that quality is currently one of the main factors necessary for an enterprise to achieve market success [6]. It is the quality of workmanship that determines whether a construction project will be positively perceived by customers [7]. Without taking into account quality or a quality management system, a company cannot survive in a volatile market in the long term.

2.1. Defects Found in Residential Buildings

The measure of quality is the number and type of defects that can be found during the technical acceptance of building structures. Physical defects involve the inconsistency of the sold item with the contract (Announcement of the Marshal of the Sejm of the Republic of Poland from 9 June 2022 on the publication of the consolidated text of the Act—Civil Code [8]). The sold item is inconsistent with the contract if it does not have the properties that it should have regarding the purpose specified in the contract, if it does not have these properties as a result of circumstances or intended use, if it does not have the properties that the seller assured the buyer about, if it is not suitable for the purpose that the buyer informed the seller about at the conclusion of the contract and the seller did not object to such a purpose, or if it was delivered to the buyer incomplete.

Defects are a common phenomenon in the construction industry. They are the result of design errors, the use of poor quality building materials and poor workmanship. There are many publications regarding the quality and the frequency of defects in residential buildings [1,2,3,9]. For example, Ojo and Ijatuyi [10] conducted a study of defects using the Sunshine Gardens estate in Akure, Nigeria. The authors found defects in all the elements of both the building’s construction and finishing. The most common defects with regard to the roof structure and its covering included the use of low-quality materials, improper wood processing, poor workmanship, and the inaccurate supervision of construction workers. Low-quality materials were used to build the walls, and the window and door lintels were too short. The floors were made of low-quality materials.

In turn, based on research conducted in Spain [11], 3647 defects were localized and classified in a set of 68 completed residential buildings. The defects were divided into three groups, namely:

• The type of defect (lack of functionality, detachment of an element, lack of levelness, incorrect assembly, curved surfaces, dirt on elements, incorrect external appearance, incorrect dimensions, and problems with water);
• The element of the building in which the defect was found (ceiling, concrete walls, doors, external joinery, elevation, floor, wooden elements, plasters and internal walls, various elements, electrical installations, sanitary installations, pillars, roof, paving slabs, stairs, and windows);
• The location of the defect in the building (common areas, external facades, floors, garages, kitchens, bedrooms, restrooms, rooms, terrace, and washrooms).

Plebankiewicz and Malara [12] classified defects in residential buildings into three groups:

• Defects of great importance—causing a defect that prevents the proper functioning of the premises, which may pose a threat to the health or life of people. Removing very important faults usually requires removing the causes and their effects. Examples of very important defects include flooding of the property, cracking of a structural element, short circuiting the electrical installation, causing a fire, or having a leak in the roof covering;
• Significant defects—causing limitations in the proper functioning of the facility. This definition defines a number of defects that limit the possibility of unlimited use, but do not force the cessation of use, as they do not pose a direct threat to the health and life of users. Examples include intercom failure, deteriorating tiles on the balcony, leaking window and balcony joinery, or slow-acting locks on the main doors;
• Defects of minor importance—these are minor defects that do not impede the functioning of the premises. These are defects of a visual or cosmetic nature. Removing these defects does not require the use of any advanced technology, and the time to remove them may vary. They can be illustrated with the following examples: scratches on walls, paint chips, spontaneous scratches on glass, or silicone detachment on balcony tiles.

Forcada et al. [13] pointed out the great importance of the inspection of construction processes by future owners. Although inspections take place during construction or at the time of handover, clients usually do not participate in them. This situation creates a gap between the quality perceived by both contractors and customers. The article presents an analysis of 52,552 acceptance defects in 2179 apartments in Spain, identifying their nature, construction element, and the industry in which these defects are found. These results were compared with previous studies that analyzed defects detected at the construction stage and those that remained after the building was handed over to the client. Studies have shown that construction defects are removed during construction thanks to existing quality standards. However, aesthetic and functional defects remain at the time of delivery. The authors conclude that many functional defects result from the lack of end-user involvement in the early stages of the project.

Similar studies were conducted by Shirkavand et al. [14]. The article presents research on the most common building defects at the time of handing over the buildings for acceptance, the causes of defects, the consequences for the main contractor, subcontractors, users and customers, and the possibilities of improvements (repairs). The results show that the most common defects recorded at the time of handing over the building are defects related to surface damage. Repairing them is neither difficult nor expensive. Another area where the most common negative deviations occur are defects related to technical installations. Unfortunately, this is often a serious problem because it prevents the entire building from fully functioning. The main cause of defects is incompleteness and poor design, and the main consequences of defects are economic losses and erosion of trust between the various stakeholders in the construction industry.

Rotimi et al. [15], using 216 residential buildings commissioned in 2008–2011 in New Zealand, determined the level of detection of defects by independent building inspectors. The most common defects included uneven painting, nail marks, poor quality finishing of rooms and floors, improperly fitted door and window handles, cracks in buildings, and the incorrect installation of toilets.

Ismail et al. [16], point out that construction defects are a common problem in housing provisions in Malaysia. Theoretically, new homes should be free from defects. To verify the quality of the apartments being built, buildings should be inspected from the early stages of construction until the handover stage. Defects are identified and classified by defect type, and the overall results indicate that most common construction defects in new homes are cosmetic. These defects are mainly due to poor workmanship. To mitigate this problem, developers must ensure that contractors are qualified, and that a thorough building inspection has been performed before handover of the home.

Gurmu and Paton-Cole [1] studied defects in residential buildings in the Australian state of Victoria. Most defects in these buildings were often hidden and appeared during the building’s use. Therefore, repairing these defects becomes difficult and expensive because access to the source of the defective components in the completed building may be difficult.

The analysis of selected articles indicates that defects in housing construction are common. They are located in various elements of buildings [11,14], in regards to the structure and finish, and may affect the safety of use of apartments to varying degrees [12]. They are generated at various stages of the investment process, both during the design, construction, and operation of the facility [13,17]. Defects in the building’s structural elements should be identified in the earlier stages of construction because they are covered during the finishing works [1,13]. By far the largest number of defects are cosmetic. Removing defects in residential buildings involves additional financial outlays. In order to avoid the above problems, research should be conducted to identify the factors that influence the generation of faults in the investment process and to determine their impact on the quality of housing construction.

Table 1. Synthesis of research results on the defects in residential buildings based on a literature review.

Authors	Statements
Ojo and Ijatuyi [10]	The most common defects in roof construction and roofing were the use of low-quality materials, improper woodworking, poor workmanship, and inaccurate supervision of construction workers. Low-quality materials were used to build the walls, and the window and door lintels were too short. The floors were made of low-quality materials.
Forcada et al. [11]	Defects were divided into three groups: the type of defect (ack of functionality, detachment of an element, lack of levelness, incorrect assembly, curved surfaces, dirt on elements, incorrect external appearance, incorrect dimensions, and problems with water), the elements of the building (ceiling, concrete walls, doors, external joinery, elevation, floor, wooden elements, plasters and internal walls, various elements, electrical installations, sanitary installations, pillars, roof, paving slabs, stairs, and windows), and the location of the building (common areas, external facades, floors, garages, kitchens, bedrooms, restrooms, rooms, terraces, and washrooms).
Forcada et al. [13]	The research shows that construction defects are removed during construction thanks to existing quality standards. However, aesthetic and functional defects remain at the time of delivery. The authors conclude that many functional defects result from the lack of end-user involvement in the early stages of the project
Plebankiewicz and Malara [12]	Defects were classified into three groups: Defects of great importance. Examples: flooding of the property, cracking of a structural element, short circuit of the electrical installation, causing a fire or leaking roof covering. Significant defects. Examples: intercom failure, deteriorating tiles on the balcony, leaking window and balcony joinery, or slow-acting locks on the main doors. Defects of minor importance. Examples: scratches on walls, paint chips, spontaneous scratches on glass or silicone detachment on balcony tiles.
Shirkavand et al. [14]	The most common defects recorded at the time of handing over the building are defects related to surface damage. Repairing them is neither difficult nor expensive, as opposed to defects related to technical installations.
Rotimi et al. [15]	The most common defects included uneven painting, nail marks, poor quality finishing of rooms and floors, improperly fitted door and window handles, cracks in buildings, and the incorrect installation of toilets.
Ismail et al. [16]	The most common construction defects in new homes are cosmetic.
Gurmu and Paton-Cole [1]	Most defects in these buildings were often hidden and appeared during the building’s use. Therefore, repairing these defects becomes difficult and expensive because access to the source of the defective components in the completed building may be difficult.

contains a synthesis of defects occurring in construction identified based on the literature.

2.2. Factors Affecting Quality

The identification of factors that influence the quality of housing construction should be the basis for the activities of every construction company. This is confirmed by Crosby’s vision of quality [18] in which the author states that quality applies to all the areas of a company’s activity, and can be measured with the use of costs.

The final quality of a building is the sum of all the factors related to the two basic creative aspects of the building, namely [19]:

• Material aspects—resulting from the proper selection and use of building materials during the implementation of investments;
• Personal aspects—all the aspects related to the work of qualified people responsible for managing the design and implementation of works, supervision, and production by persons with appropriate qualifications, preparation, experience, and technical knowledge.

The research presented by Ahzahar et al. in [20], which was conducted in Penang (Malaysia) among participants of the construction process, indicated the following material factors that are important for the quality of residential buildings: the improper use of building materials, and damage to materials during construction. Inaccurate material specifications or the delivery of low-quality materials to the construction site can also be a source of defects in residential buildings [21,22,23].

In turn, with regards to personal factors, the most commonly mentioned are the insufficient skills of employees and mistakes made by them, poor experience and knowledge of the manager, and a shortage of qualified employees [22,24,25].

Josephson and Hammarlund [26] analyzed 2879 faults collected in 7 construction projects and identified 5 different types of fault causes, namely knowledge, information, motivation, stress, and risk. Jha and Iyer [27], by analyzing the responses of specialists in about 50 large- and medium-sized organizations in the Indian construction industry contained in research questionnaires, identified the reasons that negatively affect the quality of construction: difficult climatic conditions in the construction site, negative attitude of project participants, ignorance and lack of knowledge of the project manager, etc.

Ojo and Ijatuyi [10] indicated that the main factors contributing to defective construction are the use of substandard building materials, poor workmanship, inadequate supervision, and design deficiencies. As countermeasures, they pointed to strict supervision, proper construction management and quality control, thorough training and education of craftsmen, and the use of high-quality materials. The paper’s authors also recommended the use of appropriate construction management and quality control measures to reduce the occurrence of defective construction and promote a pleasant living environment.

In research presented by Atkinson [25] conducted among people holding managerial positions during the implementation of investments, it was emphasized that errors in managerial decision making have a more significant impact on the product than omissions made by employees. The following factors, which depend on the managerial staff and which influence quality, were indicated: the manager’s experience and knowledge, manager qualifications, the level of informal communication, the quality of formal communication, organizational culture, poor planning and programming, a lack of strategic planning, the poor quality of project management, delays in decision-making, a lack of control in the investment process, cost and time pressure, and also poor communication between the designer and the contractors. The important role of training and education in the field of management in the context of identifying the correct technical solutions was also indicated. In addition, the phenomenon of corruption can also be seen to be of great importance for the quality of buildings. This is particularly visible in developing countries [21].

The design phase and the proper preparation of design documentation is of key importance for the quality of buildings under construction. Tayeh et al. [28] identify the main factors that affect the occurrence of defects at the design stage of residential buildings in the Gaza Strip. For the purposes of the analysis, a survey was carried out which identified three main design errors: ignored or incorrectly performed soil tests, a lack of qualified supervision of drawings, and inconsistency between architectural and construction drawings. In addition, the authors of studies [20,22,24] point to the quality of the design, namely poorly made project documentation [20], the ambiguity of design details, and the introduction of changes to the structure of the facility [22,24].

During the construction and operation phase, attention was paid to the type and location of the building, climatic conditions, occupational safety challenges during project implementation, the lack of inspection and inadequate supervision on the construction site, the lack of required maintenance, as well as changes with regards to using the facility [17,20,22].

Based on surveys conducted among employees of design offices, consultants, contractors and building owners who are responsible for the design, execution, and maintenance of residential buildings, it was shown that the assessment of the impact of factors on quality depends on the evaluating group [23]. Contractors considered architectural design flaws to be the most significant, consultants considered defects resulting from the technical specification as the number one factor, and clients considered defects in the architectural design as the most important.

The authors of many studies emphasize the importance of controlling the construction process to obtain a high-quality product [4,21,22,24,25]. Fernández et al. [4] analyzed various quality control factors in housing projects in Spain. Based on their research, they proposed changes regarding control, which involved paying attention not only to the management process but also to the quality and control of the supply of materials. Zalejska et al. proved that the size of the development company and the location of the building also have a significant impact on construction defects [29].

The occurrence of defects in residential buildings is associated with the need to remove them, which in turn generates significant costs. Based on research conducted by Mills et al. [30] in Australia between 1982–1997, it was found that the cost of repairing defects is about 4% of the contract value. In turn, Josephson [26] found that the cost of repairing defects corresponds to 4.4% of the construction costs of buildings, and the time to repair them is 7% of the total working time.

Defects result in the budget for a construction investment being exceeded. Therefore, an important element of research in this area is the search for a model approach for managing defects. This approach was proposed by Park et al. [31] in the form of a building defect management system with the use of augmented reality (AR) and Building Information Modeling (BIM). Chatterjee et al. proposed a model for predicting defects in multi-story reinforced concrete buildings using neural networks (NN-PSO classifier) [32], while Aljassmi and Han proposed a model for analyzing the causes of defects and faults using trees and measures of risk [5].

Defects in residential buildings can occur in all of their elements. Verification of the nature of the detected defects and the identification of factors that influence their creation may lead to valuable conclusions regarding company management, which in turn will affect the achievement of high-quality residential buildings.

Based on the conducted study of the literature, it was found that the main factors that influence the formation of construction defects are material factors and personal factors. Material factors include the improper use of building materials, the delivery of low-quality materials to the construction site, damage to materials during construction, incorrect and poorly executed technical specifications, and non-compliance with the specifications of the performed works. However, personal factors also have a decisive impact on the phenomenon of defects and faults in residential buildings. Studies have shown that the main factors that affect quality include human factors at the managerial level. Based on the literature review regarding the identification of defects and the factors that affect their generation, no studies were found that would indicate the strength of the impact of individual factors on the formation of defects. According to the authors, this is an area of research that is a blank spot in the field of research regarding quality in housing construction. This issue is the subject of the research presented in this article.

Table 2. Summary of research results on factors influencing quality in construction.

No. of Factor	Literature Item	Type of Factor	Description
1	[10,14,20,22,24,28]	Errors in the design documentation	design errors, faulty structural and architectural design, ambiguity of design details, omission to take into account the conditions of use (maintenance) at the design stage, defects in construction drawings
2	[4]	No legal basis for the design	lack of design standards, incompatibility of the project with existing technology
3	[22,24]	Making changes to the structure	changes in structure during construction and use
4	[10,16,22,23,25]	Qualifications and skills of workers	unskilled contractors, insufficient skills of workers, contractor errors
5	[10,22,23,25,27]	Qualifications and skills of the management staff	qualifications, experience and knowledge of the manager, skills related to construction
6	[10,24,25,29]	Quality management	lack of quality management during construction, lack of control during construction, poor supervision on site
7	[27,33]	Engagement	involvement of the supervisory team, involvement of other employees
8	[25,29,30]	Construction management skills	poor planning and scheduling, delays in decision making, planning/programming
9	[24,25,30]	Team communication	lack of communication, level and quality of formal and informal communication, organizational culture, lack of communication between the designer and contractors
10	[10,20]	Lack of work teams	lack or deficiency of resources
11	[10,20,21,22,23]	Delivery of materials not in accordance with specifications	non-compliance with specifications, incorrect specifications, defects resulting from material specifications, use of unqualified trade subcontractors
12	[10,20,21,22,23]	Poor materials management	poor material management, improper use of building materials, damage to materials during construction
13	[27]	Climatic influences	difficult climatic conditions, bad weather conditions
14	[14,25]	Financial difficulties, time	client’s financial uncertainty, cost pressure, time pressure
15	[13]	End customer engagement	participation of the end customer in technical acceptance

lists the factors influencing the generation of defects identified based on the literature.

3. Methodology and Materials

A methodology for identifying and assessing the impact of selected factors on the generation of defects during the construction of residential buildings was developed. This methodology consists of three stages:

• The first stage included in situ tests carried out on construction sites, which consisted of:

  ○

  identifying defects in residential buildings during technical acceptance;

  ○

  identifying factors occurring at various stages of the construction process that affect the quality of a residential building.

• In the second stage, in order to determine the strength of the influence of factors on the generation of defects, statistical analyses of the survey results and the results of analyzing defects were performed. An assessment of the impact of factors on quality was made in relation to the examined investments. Discriminant analysis was used in the calculations.
• In the third stage, conclusions from the research were formulated.

3.1. Identification of Defects in Residential Buildings

Technical acceptance procedures were carried out in eight residential buildings that were implemented in Poland in the Lower Silesia region in the years from 2017 to 2020. During the acceptance of the residential premises, over 9300 construction defects were recorded in 669 residential dwellings that had a total usable area of 36,920 m². The seven- and eight-story buildings were reinforced concrete structures, while the lower buildings had a mixed reinforced concrete and masonry structure [9].

Table 3. List of buildings subjected to the testing.

No.	Investment	Usable Area [m2]	Number of Dwellings	Main Construction	Building Volume [m3]	Number of Stories	Year of Commissioning
1	A	3500	34	reinforced concrete and masonry	15,425.21	2	2018
2	B	6682	141	reinforced concrete	54,336.16	8	2017
3	C	6370	135	reinforced concrete	52,447.02	7	2019
4	D	4031	82	reinforced concrete and masonry	19,030.95	4	2019
5	E	1907	24	reinforced concrete and masonry	9,001.04	4	2019
6	F	3579	67	reinforced concrete and masonry	16,892.88	4	2019
7	G	4804	78	reinforced concrete and masonry	22,674.88	4	2020
8	H	6047	108	reinforced concrete	42,455.60	8	2019
	Total	36,920	669

contains basic information about the buildings.

In order to identify defects of the greatest importance for the quality of residential buildings, the Pareto-Lorenz analysis and the ABC classification were used [34,35]. The proposed calculation methodology is very simple and universal, and is often used in research on the influence of causes on effects in production processes. The Pareto principle, also known as the 20/80 rule, states that usually 20% of the causes generate 80% of the effects. Therefore, eliminating 20% of the causes in residential construction may result in avoiding 80% of the most common defects [36].

The analysis procedure is as follows:

• Defining the types of defects and their number.
• Sorting the identified defects in terms of number from maximum up to minimum.
• Calculation of the percentage of individual defects in the set of all identified defects.
• Calculation of the cumulative percentage of consecutive sorted defects.

In order to determine the importance of individual types of defects on the quality of buildings, ABC analysis, known in economics and easy to apply, was used [35]. According to the ABC method, it is necessary to divide the defects into three groups. It was assumed that:

• the set of very significant defects, marked as A, consists of defects that constitute 80% of all defects found during technical acceptance;
• the set of significant defects, marked as B, constitutes 15% of all defects identified during technical acceptance;
• the set of minor defects, marked as C, constitutes 5% of all identified defects.

3.2. Evaluation of Influence Factors

The study of the factors that influence the quality of housing construction was carried out using the diagnostic survey method and the survey technique. The scope of the research included the following elements of the construction process: preparation of the investment including analysis of design documentation, preparation of an offer for the implementation of works, implementation of construction works, acceptance of the building, and completion of the handover of the facility along with the removal of defects found during acceptance. The research on impact factors took place in the following stages:

• Based on a literature review, a list of factors influencing the formation of defects was determined. 15 factors were identified and are listed in Table 2.
• This list has been supplemented with additional factors resulting from the engineering experience of the authors of the article and obtained based on interviews conducted among construction management staff. 33 factors were obtained.
• A survey was prepared containing 33 identified factors that were assigned to individual phases of the investment process. The survey consisted of two parts:

  ○

  Part one included questions of a general nature, regarding the type of investment, the function performed during the investment implementation, and the respondent’s professional experience.

  ○

  Part two was divided into fragments corresponding to the stages of the investment process in construction. Each task included a set of questions regarding previously defined factors based on the literature and our own observations.

• The survey was addressed to respondents who were construction and company management staff, construction managers, works managers, and people working in the production preparation department.
• Respondents assessed the factors that influence the occurrence of defects in housing construction. A 5-point Likert scale was used for the assessment, where 1 meant “I strongly disagree,” and 5 meant “I strongly agree.” [37]. The survey results were recorded in the form of a table.

3.3. Discriminant Analysis

The subject of the analysis was the assessment of the impact of individual factors on the total number of defects found in the examined dwellings. To determine the significance of the discriminatory power of the number of defects with regard to individual factors, discriminant analysis was performed. During this analysis, the a posteriori probability and the statistics of the multivariate Lambda Wilks test were determined [38,39,40].

The discriminatory variables were: 33 defined influence factors (C1÷C33), the number of dwellings in the building (C34), and the area of the dwellings (C35). The set of analyzed residential dwellings was divided into three groups. The grouping variable was the identified number of defects. The analysis was carried out using the stepwise method and the SPSS program for statistical calculations. The stepwise method enabled the subset of factors that best discriminate the number of defects to be found.

In order to estimate the discriminant function, the a posteriori probability $P\left(G_p/d\right)$ of the object (a dwelling) that belongs to the group described by the discriminant function $G_p$ was calculated:

$$P\left(G_p/d\right)=\frac{P\left({d/G}_p\right)P\left(G_p\right)}{{\sum }_{i=1}^{n}P\left({d/G}_p\right)P\left(G_p\right)},$$

(1)

where:

• $P\left(G_p\right)$—the a priori probability that object p (a dwelling) belongs to group $G_p$ ($p=1,\dots ,P)$
• $P\left({d/G}_p\right)$—the conditional probability that the discriminant function $G_p$ for a given object will take a certain value d if the group to which our object belongs to is known.
• $P\left(G_p/d\right)$—the a posteriori probability of the object (a dwelling) that belongs to the ${\mathrm{G}}_{\mathrm{p}}$ group if the value of the discriminant function for it is d.

Probability $P\left(G_p/d\right)$ indicates how accurate the prediction of the number of defects is.

The impact of factors on the quality of the residential housing was assessed using the following statistics [41]:

• Wilks’ partial lambda determines the variable’s contribution to the discrimination of groups. The lower the value of this statistic, the greater the discriminatory power of a given variable.
• The F statistic value measures the discriminatory power of a variable and indicates the order in which the input variables are entered into the model.
• The critical p-level indicates whether the variable contributes significantly to the model (discriminant variable).
• Variable tolerance and tolerance index. The value of the tolerance index is calculated as 1-tolerance of the variable when other variables are included in the model. Tolerance is a measure of the model’s fitting, i.e., the quality of the contribution of the independent variables (in our example, the analyzed influence factors) in the explaining of the dependent variable—the number of defects. If the tolerance value is close to 0, there is a poor fitting of the model.

4. Results

4.1. Results of the Analysis of Defects

Table 4. Quantitative summary of the defects in the buildings subjected to the tests.

No.	Defect Type	Investment								Total	Percentage Share
		A	B	C	D	E	F	G	H
1	Plasters	366	100	389	382	108	404	76	275	2100	22.43%
2	Windows	218	63	176	248	196	273	52	183	1409	15.05%
3	Flooring	130	6	139	111	53	125	16	129	709	7.57%
4	Electric installations	74	30	70	124	41	169	6	153	667	7.12%
5	Common parts	0	141	44	38	47	58	1	232	561	5.99%
6	Water and sewage installations	131	9	34	58	49	151	8	78	518	5.53%
7	Shortcomings in the cleaning of elements	79	15	102	59	61	129	11	49	505	5.39%
8	Doors	26	15	138	110	27	60	14	100	490	5.23%
9	Glazing	16	0	96	66	33	52	22	137	422	4.51%
10	Other elements	113	5	69	49	32	60	24	31	383	4.09%
11	Tiles	5	38	31	37	28	79	2	85	305	3.26%
12	Traces of moisture	66	13	38	88	18	39	20	12	294	3.14%
13	Balustrades	1	13	48	35	38	99	2	34	270	2.88%
14	Facades	25	31	12	1	41	135	14	0	259	2.77%
15	Windowsills	37	0	25	11	19	18	1	28	139	1.48%
16	Ventilation	54	10	21	17	4	19	0	12	137	1.46%
17	Insulations	47	1	32	5	6	6	21	15	133	1.42%
18	Roofs	26	4	1	16	5	5	4	0	61	0.65%
	Total	1414	494	1465	1455	806	1881	294	1553	9362	100.00%

contains data concerning the number of defects that were identified in individual buildings. Among the identified types of elements with construction defects, the following were distinguished: plasters, windows, flooring, electrical installations, common areas (in total), water and sewage installations, doors, glazing, tiles, balustrades, facades, windowsills, ventilation, insulation, roofs, and shortcomings in the cleaning of elements.

Table 5. Datasheet for constructing the Pareto-Lorenz diagram.

No.	Defect Type	Number of Defects px	Cumulative Number of Defects Sx	Cumulative Percentage Share of Defects ux	Defect Group
1	Plasters	2100	2100	22.43	A
2	Windows	1409	3509	37.48
3	Flooring	709	4218	45.05
4	Electric installations	667	4885	52.18
5	Common areas	561	5446	58.17
6	Water and sewage installations	518	5964	63.70
7	Shortcomings in the cleaning of elements	505	6469	69.10
8	Doors	490	6959	74.33
9	Glazing	422	7381	78.84
10	Other elements	383	7764	82.93	B
11	Tiles	305	8069	86.19
12	Traces of moisture	294	8363	89.33
13	Balustrades	270	8633	92.21
14	Facades	259	8892	94.98
15	Windowsills	139	9031	96.46	C
16	Ventilation	137	9168	97.93
17	Insulations	133	9301	99.35
18	Roofs	61	9362	100.00
	TOTAL	9362

presents data regarding the types of construction defects and their number, the cumulative number of defects, the cumulative percentage of defects, and the division of defects into three groups: A, B and C [35].

Based on the conducted calculations, a diagram of the Pareto distribution with the Lorenz curve was prepared [36,42,43] and is shown in Figure 1.

• Group A, marked in red, includes plasters, windows, flooring, electrical installations, common areas, water and sewage installations, doors, glazing, and the cleaning of elements. This group accounts for 80% of all the defects.
• Group B, marked in orange, includes other elements, tiles, traces of moisture, balustrades, and facades. This group accounts for 15% of all the defects.
• Group C, marked in green, includes windowsills, ventilations, insulations, and roofs. This group accounts for 5% of all the defects.

It should be noted that it is only the first two types of construction defects, i.e., plasters and windows, that account for as much as 37.48% of all the defects found during the acceptance of the residential premises. It was also found that two types of defects, i.e., plasters and windows, are responsible for nearly half (as much as 47.54%) of all the defects in group A, which shows how important these areas of construction work are for the investment construction process.

4.2. Results of the Analysis of Influence Factors

The respondents assessed, on a five-point Likert scale, the impact of 33 factors on the quality of housing construction. Individual identified factors were assigned to individual stages of the construction process. Quantitative and percentage results are presented in

Table 6. Results of assessing the factors by respondents.

No.	Factor	Factor Evaluation
		1	2	3	4	5
I	STAGE I—making a decision by the contractor to start the process of preparing the documentation analysis
C1	Lack of internal control of the design documentation before the start of the construction of the facility	0 (0.0%)	1 (1.8%)	15 (26.8%)	20 (35.7%)	18 (32.1%)
II	STAGE II—preparation by the contractor of an offer for the implementation of the investment
C2	Errors in bills of quantities (poorly made bills of quantities that do not take into account a number of contract items)	0 (0.0%)	8 14.3%)	8 (14.3%)	28 (50.0%)	12 (21.4%)
C3	Design errors in the documentation	0 (0.0%)	4 (7.1%)	5 (8.9%)	21 (37.5%)	25 (44.6%)
C4	There is no database regarding companies that perform construction works at a high level	0 (0.0%)	0 (0.0%)	3 (5.4%)	17 (30.4%)	33 (58.9%)
C5	Failure to analyze offers for the execution of works and also failure to confront them with the investor’s cost estimate	2 (3.6%)	5 (8.9%)	13 (23.2%)	25 (44.6%)	11 (19.6%)
C6	No production preparation department	4 (7.1%)	14 (25.0%)	22 (39.3%)	5 (8.9%)	0 (0.0%)
III	STAGE III—implementation of the building shell
C7	Changes in the design of the facility’s elements during its construction which generate additional technical problems to be solved (e.g., optimization of the structure, installations. etc.)	0 (0.0%)	0 (0.0%)	3 (5.4%)	32 (57.1%)	21 (37.5%)
C8	Delays in the implementation of reinforced concrete works. i.e., a key element of each investment, which translates into delays in subsequent scopes of works	1 (1.8%)	8 (14.3%)	15 (26.8%)	26 (46.4%)	5 (8.9%)
C9	Lack of experience of the site manager in the organization of large construction sites	3 (5.4%)	6 (10.7%)	1 (1.8%)	6 (10.7%)	31 (55.4%)
IV	STAGE IV—execution of finishing works
C10	Lack of interior design for common areas—works are carried out based on current arrangements during the construction	0 (0.0%)	2 (3.6%)	3 (5.4%)	0 (0.0%)	20 (35.7%)
C11	Lack of participation of the construction manager in the process of contracting contractors for finishing works	0 (0.0%)	0 (0.0%)	3 (5.4%)	24 (42.9%)	29 (51.8%)
C12	Contracting companies without experience and the verification of their competence and references	0 (0.0%)	0 (0.0%)	0 (0.0%)	16 (28.6%)	40 (71.4%)
C13	Lack of a prepared contract team that is capable of carrying out dozens of tenders in a short time interval	0 (0.0%)	2 (3.6%)	10 (17.9%)	29 (51.8%)	14 (25.0%)
C14	Lack of a designated person from the contracting department who cooperates with the site manager on an ongoing basis.	0 (0.0%)	0 (0.0%)	0 (0.0%)	34 (60.7%)	22 (39.3%)
C15	Lack of stability of the team (high staff turnover) to conduct contract tenders	0 (0.0%)	0 (0.0%)	4 (7.1%)	39 (69.6%)	13 (23.2%)
C16	The unpredictability of the skills of contractors. especially unqualified workers from across the eastern border	0 (0.0%)	0 (0.0%)	1 (1.8%)	13 (23.2%)	42 (75.0%)
C17	Pursuing a “one company” policy (i.e., the same contractor carrying out several investments for the same investor)	0 (0.0%)	0 (0.0%)	19 (33.9%)	25 (44.6%)	12 (21.4%)
C18	Lack of experience of the site manager in checking whether construction site engineers have completed construction site organization tasks	1 (1.8%)	2 (3.6%)	10 (17.9%)	5 (8.9%)	24 (42.9%)
C19	Difficulties in cooperation between the site manager and the contract team	0 (0.0%)	5 (8.9%)	2 (3.6%)	34 (60.7%)	15 (26.8)
C20	Low experience (or no experience) of contract team members	0 (0.0%)	2 (3.6%)	4 (7.1%)	45 (80.4%)	5 (8.9%)
C21	Delays in contracting individual scopes of work (e.g., contracting a given scope of work several months after the planned date)	0 (0.0%)	2 (3.6%)	17 (30.4%)	26 (46.4%)	11 (19.6%)
C22	Contracting companies that offer the lowest prices	0 (0.0%)	0 (0.0%)	0 (0.0%)	6 (10.7%)	50 (89.3%)
C23	Contracting a given scope of work several times (due to the need to introduce substitute execution)	0 (0.0%)	5 (8.9%)	5 (8.9%)	23 (41.1%)	23 (41.1%)
C24	Contracting companies without analyzing their ability to perform a given scope of work	0 (0.0%)	0 (0.0%)	4 (7.1%)	34 (60.7%)	18 (32.1%)
C25	Lack of complete contracts that cover the entire scope of works to be performed—the need to supplement orders after detecting incomplete scope of works to be performed	0 (0.0%)	6 (10.7%)	18 (32.1%)	23 (41.1%)	9 (16.1%)
C26	Lack of financial possibilities for conducting large-scale constructions by contractors	0 (0.0%)	5 (8.9%)	21 (37.5%)	24 (42.9%)	5 (8.9%)
C27	Expectation of advance payments by contractors who do not have financial capabilities	4 (7.1%)	14 (25.0%)	18 (32.1%)	19 (33.9%)	0 (0.0%)
C28	Loss of financial liquidity of subcontractors (who also carry out other investments)	0 (0.0%)	12 (21.4%)	4 (7.1%)	31 (55.4%)	8 (14.3%)
V	STAGE V—acceptance of the investment
C29	Lack of internal acceptance of the completed works	0 (0.0%)	1 (1.8%)	0 (0.0%)	21 (37.5%)	33 (58.9%)
C30	Lack of executive potential to prepare the facility for technical acceptance	0 (0.0%)	0 (0.0%)	1 (1.8%)	5 (8.9%)	50 (89.3%)
C31	Not enough engineering staff involved in the preparation of premises for technical acceptance	0 (0.0%)	0 (0.0%)	3 (5.4%)	31 (55.4%)	22 (39.3%)
VI	STAGE VI—carrying out the defect removal process
C32	Lack of responsibility of the construction management staff (site manager, construction engineers) for the removal of defects in the investment	3 (5.4%)	6 (10.7%)	1 (1.8%)	7 (12.5%)	26 (46.4%)
C33	Lack of executive potential to remove defects in a short time	0 (0.0%)	0 (0.0%)	0 (0.0%)	25 (44.6%)	31 55.4%)

.

4.3. Discriminant Analysis

Three groups of residential premises were selected. The belonging of the premises to a given group is determined by the total number of identified defects. The groups are:

• Group G1, which includes premises for which the total number of defects <650;
• Group G2, which includes premises for which the total number of defects is in the range of 540–1508;
• Group G3, which includes premises for which the total number of defects >1508.

These ranges were selected based on the variance of the number of defects.

Table 7. Analysis of the discriminant function with a posteriori probability that defines the discriminatory power of a given factor in relation to the total number of defects.

Variable	A Posteriori Probability $\mathit{P}\left({\mathit{G}}_{\mathit{p}}/\mathit{d}\right)$	Wilks’ Partial Lambda	F	p	Tolerance	1-Tolerance R2
C1	0.580	0.488	333.625	<0.001	0.052	0.948
C15	0.472	0.528	283.382	<0.001	0.018	0.982
C30	0.463	0.691	141.651	<0.001	0.043	0.957
C14	0.407	0.593	217.978	<0.001	0.021	0.979
C10	0.405	0.722	122.430	<0.001	0.034	0.966
C21	0.388	0.628	188.439	<0.001	0.046	0.954
C26	0.373	0.627	188.798	<0.001	0.024	0.976
C35	0.365	0.635	182.314	<0.001	0.005	0.995
C5	0.348	0.683	147.025	<0.001	0.012	0.988
C29	0.347	0.653	168.869	<0.001	0.032	0.968
C31	0.343	0.657	165.481	<0.001	0.012	0.988
C34	0.331	0.669	156.767	<0.001	0.007	0.993
C4	0.277	0.770	94.579	<0.001	0.036	0.964
C33	0.267	0.757	101.857	<0.001	0.024	0.976
C27	0.235	0.765	97.739	<0.001	0.041	0.959
C2	0.235	0.932	23.277	<0.001	0.037	0.963
C23	0.212	0.801	78.708	<0.001	0.027	0.973
C9	0.190	0.885	41.101	<0.001	0.072	0.928
C18	0.185	0.848	57.031	<0.001	0.052	0.948
C19	0.178	0.877	44.678	<0.001	0.042	0.958
C3	0.156	0.974	8.420	<0.001	0.052	0.948
C6	0.155	0.949	17.129	<0.001	0.069	0.931
C12	0.132	0.941	20.053	<0.001	0.049	0.951
C8	0.129	0.871	46.963	<0.001	0.032	0.968
C16	0.121	0.939	20.767	<0.001	0.016	0.984
C32	0.112	0.952	16.177	<0.001	0.006	0.994
C7	0.074	0.926	25.278	<0.001	0.014	0.986
C24	0.072	0.928	24.554	<0.001	0.023	0.977
C25	0.060	0.940	20.152	<0.001	0.041	0.959
C28	0.052	0.948	17.533	<0.001	0.036	0.964
C17	0.049	0.951	16.215	<0.001	0.033	0.967
C13	0.046	0.994	2.062	0.128	0.025	0.975
C11	0.033	0.967	10.754	0.116	0.051	0.949
C22	0.023	0.992	2.648	0.072	0.053	0.947
C20	0.006	0.994	2.007	0.045	0.073	0.927

presents the results of the analysis.

The entire model explains 78% of the total variance in the number of defects, and also shows which of the factors differentiate the number of defects the most. The p-value for the F test in the model indicates the statistical significance of the discrimination. The F value is a variance statistic that shows whether a factor, taking into account the possibility of other factors, significantly affects the sum of defects. The results are statistically significant if p is less than 0.05.

To avoid poor conditioning of the covariance matrix, a tolerance factor was introduced, the value of which was calculated as 1-tolerance of a variable when including other variables in the model’s system. If the tolerance value is close to zero, there is a poor fitting of the model. In the case of the analyzed variables, the tolerance is high and there is a good fitting of the model. Tolerance is a measure of the fitting of the model, i.e., the quality of the contribution of the independent variables to the explaining of the number of defects. Variables for which the values of the F statistic are the highest, and for which the values of Wilks’ partial lambda are the lowest, contribute to the discrimination of the number of defects in dwellings. These are:

• C1—lack of internal control of the design documentation before the start of the construction of the facility;
• C15—lack of stability of the team (high staff turnover) for conducting contract tenders;
• C30—lack of executive potential for preparing the facility for acceptance;
• C14—lack of a designated person from the contracting department who cooperates with the site manager on an ongoing basis;
• C10—lack of interior design for common areas—works carried out based on current arrangements during construction;
• C21—delays in contracting individual scopes of works (e.g., contracting a given scope of work several months after the planned date);
• C26—lack of financial capacity for conducting large-scale constructions by contractors; C34—The usable area of a dwelling [m2];
• C5—lack of analysis of offers for the execution of works, and a lack of confronting them with the investor’s cost estimate
• C29—lack of internal acceptance of completed works;
• C31—too few engineering staff involved in the preparation of premises for technical acceptance;
• C35—number of dwellings in the building.

These factors were also noticed by other authors of publications [3,4,20,28,29]. Four variables were not entered into the model, namely:

• C13—lack of a prepared contract team that is capable of carrying out several dozen tenders in a short time interval;
• C11—lack of participation of the construction manager in the process of contracting contractors for finishing works;
• C22—contracting companies that offer the lowest prices;
• C20—low experience (or no experience) of contract team members.

For these variables, the p-statistic values are greater than 0.05. The results are not statistically significant.

5. Discussion

Based on the Pareto-Lorenz diagram, it is possible to indicate the building elements in which the largest number of defects occur and therefore which elements should be particularly controlled by the construction supervision [36]. In the conducted tests, the largest number of defects was found in plasters and windows, followed by flooring, electrical installations, common areas, water and sewage installations, doors, glazing, and shortcomings in the cleaning of elements. The above-mentioned places with the largest number of defects were noticed by other researchers who deal with quality in housing construction, namely in papers [9,13,14,16].

The authors’ own research and the review of the literature show that in different countries, regardless of geographical location, similar defects can be found in housing construction. The reasons for such situations cannot just be poor quality building materials. The rules for introducing construction products in the construction industry, which are applicable in many countries including those in the European Union, oblige manufacturers to meet a number of regulations. In the European Union, harmonized technical specifications are used such as European harmonized standards and European technical approvals (in the case of CE certification). They allow for the circulation of the product throughout the economic area of the European Union. The rules for manufacturing and marketing construction products have been sufficiently formalized in legal, controlled, and organizational acts. This directly translates into the high quality of construction products that are delivered to a construction site [44].

The reasons for the large number of defects in residential premises that are commissioned for use should therefore be sought in the second group of factors, i.e., human factors (Table 4). The discriminatory power of a factor is determined by the value of the F statistic [41]. The values presented in Table 2 indicate that individual factors have different discriminatory power.

According to the research carried out by the authors, factor C1 has the greatest discriminatory power of defects—a lack of internal control of the design documentation before the start of the construction of the facility. The analysis of the documentation of the facility, before submitting it for the preparation of the offer and the implementation of the investment, is crucial with regards to the quality of the facility because it will allow errors, deficiencies, and shortcomings in the documentation - such as: C33—lack of executive potential to remove or be detected, and the impact of the following factors on the quality of the building to be eliminated: C2—errors in the bill of quantities, C3—errors in documentation, C5—errors in the cost estimate. The introduction of design documentation with technical errors in the implementation will result in the need to solve problems at the construction stage, which will in turn affect the quality. Defects in the architectural andI confirmed - it is ok structural design, incomplete detailed drawings, defects resulting from material specifications, lack of design standards, structural changes made by the owner at the implementation stage, failure to adapt the design to the existing technology, failure to take into account the conditions of using the facility (its maintenance) at the design stage, and the assessment of the conditions of using the facility are factors that were also indicated in the paper [23].

The conducted research also showed how important it is to have a well-trained and stable contracting team involved in the preparation for the bid for the investment. This is indicated by the factors that largely discriminate the number of defects, such as C15—lack of stability of the team (high staff turnover) to conduct contract tenders, and C14—lack of a designated person from the contracting department who cooperates with the site manager on an ongoing basis. Other factors that significantly affect the generation of defects are C30—lack of executive potential to prepare the facility for acceptance, C29—lack of internal acceptance, C31—too few engineering staff involved in the preparation of premises for technical acceptance, and C33—lack of executive potential to remove defects in a short time.

The subject literature most often mentions human factors such as design errors [23], lack of proper supervision and construction management skills, lack of communication between the designer and contractors [22], and also insufficient skills of contractors [24]. In addition, the authors of some publications also mention weather conditions [20], financial problems, and poor planning and scheduling [22]. However, the literature does not mention factors related to the contracting of construction works (C14, C15), or the contractor’s capabilities related to the acceptance of facilities (C29, C30, C31, C33). The elimination of these factors in the investment process will significantly contribute to the improvement of the quality of residential premises.

Factors that are indicated in the literature, and those not included in this study, should also be noted. These involve, among others, the type of building, the location of the building, the incorrect assessment of the conditions of use, poor planning and scheduling, and poor organizational culture. These factors will be taken into account in subsequent research conducted by the authors concerning quality management in housing construction in Poland.

6. Conclusions

The article presents the results of research that aimed to identify defects in residential buildings, identify factors at the stages of the investment process that influence the generation of defects, and assess the impact of individual factors on the occurrence of defects. The measure of quality is the number and type of defects that were found during the technical acceptance of building objects.

A methodology for identifying and evaluating the impact of selected factors on the generation of defects in residential buildings was developed. The methodology consists of three stages:

• In situ tests carried out on construction sites which involve the identification of defects in residential buildings during the technical acceptance procedure;
• The identification of factors (based on conducted surveys) that occur at individual stages of the construction process and which affect quality;
• Determining the strength of the influence of factors on the generation of faults using discrimination analysis.

The largest number of defects were found in building elements such as plasters (22.43%) and windows (15.05%), followed by floors (7.57%), electrical installations (7.12%), common parts (5.99%), water and sewage installations (5.53%), deficiencies in cleaning elements 5.39%), doors (5.23%), and glass (4.51%).

Based on a literature review and surveys conducted among construction managers, 33 factors were identified that influence the generation of defects.

Factor C1 has the greatest power to discriminate faults—lack of internal control of design documentation before starting the construction of the facility. The analysis of the facility’s documentation, before sending it to prepare an offer for construction works and investment implementation, is crucial for the quality of the facility because it will allow detection of errors, omissions and shortcomings in the documentation and eliminate the impact on the quality of the building of factors such as C2 errors in bills of quantities, errors in the C3 documentation, and errors in the C5 cost estimate. The research also showed that it is very important to have a well-trained and stable team responsible for preparing an offer for the investment.

The results of the conducted research are very important for construction practice in the field of quality management in residential construction. They indicate which elements of a residential building have the most defects identified and what factors have the greatest impact on generating defects. This knowledge is useful for participants of the investment process (construction manager, investment supervision inspector, designer, investor), construction companies and, above all, the client. The application of research results in practice will result in the customer receiving a product of the highest quality, free from defects, which should be the goal at every stage of planning, implementation, and operation of the facility. An important practical tip is to point out that the most important impact factor is the quality of the design documentation and the need to control the documentation before starting the construction of the facility.

The results of the conducted research contributed (certainly only to a limited extent) to filling the research gap in the area of assessing the impact of identified factors on the generation of defects in housing construction. The knowledge obtained will help in the future to develop quality management procedures and minimize the costs incurred for removing defects. Knowledge about phenomena occurring in residential buildings can therefore be used to better plan the investment budget.

The authors point out two limitations in the research conducted:

• A relatively small set of residential buildings consisting of 8 objects.
• Research territory limited to one region of Poland, Lower Silesia. These two limitations may constitute the basis for stating that the research results are not representative of the whole of Poland. However, the authors of the article intend to conduct further research in other regions of Poland, which will allow for a comparison of the obtained results.

The conducted research and analyzes indicate directions for future research, namely:

• Research on the perception of the impact of identified factors on quality by people holding managerial positions in construction and with various professional experiences. The results of such research will help in making decisions regarding the employment of people for managerial positions in the construction industry.
• Develop procedures to control the investment process at every stage of the facility’s construction so that the number of defects found is as low as possible. It is impossible to completely eliminate defects, but the key thing from the point of view of managing the investment process is to significantly reduce them.