Handover defects: comparison of construction and post-handover housing defects

1 Handover defects: comparison of construction and post-handover housing defects Nuria Forcadaa, Marcel Macarullab, Marta Gangolellsa, Miquel Casalsa a Department of Construction Engineering, Group of Construction Research and Innovation (GRIC), Universitat Politècnica de Catalunya, C/ Colom, 11, Ed. TR5, 08222 Terrassa, Barcelona, Spain E-mail: [email protected], [email protected], [email protected] b Department of Engineering Design, Group of Construction Research and Innovation (GRIC), Universitat Politècnica de Catalunya, C/ Colom, 11, Ed. TR5, 08222 Terrassa, Barcelona, Spain E-mail: marce[email protected] Abstract: Although inspections occur during construction or at handover, customers do not normally participate. This situation creates a gap between quality perceived by contractors and customers. An analysis of 52,552 handover defects in 2,179 flats in Spain is presented which identified their nature, the building element and trade where these defects are located. These results are compared to previous studies that analyzed defects detected during the construction stage and those that remain after handing over the building to the client. The research reveals that structural defects are resolved during construction due to existing quality standards. However, other aesthetic and functional defects remain and/or arise at handover. Some defects are not resolved until customers complain after they first occupy the dwelling. Many functional defects arise due to the lack of involvement of end users in the early project stages. Keywords: construction defects, defects, customer complaints, housing, quality control, Spain 2 Introduction Numerous studies have analyzed the factors affecting the quality of housing (e.g., Craig, et al. 2010, Chong & Low, 2005, 2006, Johnsson, 2009, Georgiou et al., 1999, Ilozor et al., 2004, Mills et al., 2009, Sommerville & McCosh, 2006). In Spain, research on housing defects has been confined to the studies undertaken by Forcada et al. (2012, 2013a, 2013b, 2014). Within the housing sector, there are two major opportunities for a builder to rectify defective elements:  during construction  prior to the building handover when a range of inspections occur by site management, and relevant warranty and guarantee providers. In previous studies, Forcada et al. (2014) analyzed those defects detected during construction in 68 residential building developments undertaken by two large Spanish contractors. The research revealed that the most common defects that arise during construction technical faults related to the stability of the structure and inappropriate installation of roofs and façades caused by poor workmanship rather than the quality of the materials or products used (Forcada et al., 2014). An alternative approach to understanding defects is to examine client complaint forms. Forcada et al. (2012, 2013a, 3013b) analyzed the defects that remain after handover from four Spanish builders’ databases. The most common defects identified after handover by customers were predominately functional rather than technical in nature (Forcada et al., 2013a). In general, post-handover defects were found to be incorrect or missing grouting / sealant in tiles, fixtures and fittings in toilets, failure to apply second coats of paint to walls or surface/appearance defects such as floor or wall unevenness, 3 stains, mess, small cracks and marks mainly caused by lack of protection (Forcada et al., 2013b). Although building defects have been widely addressed in previous research in relation to their concepts, profiles and causes (Georgiou 2010; Yung and Yip 2010; Macarulla et al 2013) they have mostly been examined within the context of their associated studies, while there is little cross-context comparative analysis and a lack of fundamental exploration of the nature and features of building defects. The aim of the current research is to detect if quality control measures adequately fulfil their roles. Therefore, the research presented in this paper examines the nature of defects that remain at handover and compares it with those defects identified by customers when they first occupy the dwelling (Forcada et al., 2013a; Forcada et al., 2014). Understanding the nature of defects, who detects them and when are they resolved can enable appropriate quality strategies to be developed and implemented. Therefore, to support the implementation of these strategies, this research provides knowledge of those elements and trades where builders are likely to make errors, mistakes or deliberately take short-cuts. The current study and the results drawn on from other studies used the same classification system to analyse data derived from the non-conformances, checks and/or clients complaints forms obtained from Spanish contractors’ databases. This enables a consistent approach to the analysis of the defects at different lifecycle stages. The housing sector In Spain, the demand for housing increased significantly in the mid-1990s, leading to a rise in prices and increased activity in the construction sector. When the economic situation changed, leading to higher unemployment and interest rates, the construction 4 sector faced challenges that affect its future viability and that of the entire national economy (Forcada et al 2012). Nevertheless, the construction industry still accounts for 7.2% of gross domestic product, and the housing sector represents 66% of the total building sector (Asociación de empresas constructoras de ámbito nacional (National association of construction companies) 2013). The Spanish housing construction boom of the late 1990s and the first decade of the 21st century, along with the ease of entering the market, led to an influx of inexperienced workers and an increase in competition within the industry. This, in turn, gave rise to an observed decline in quality (Forcada et al 2012). Moreover, the marginal role played by end users in defining functional and quality requirements has fostered a perception of poor quality at the time of purchase. This lack of quality is observed in the form of defects.Construction defects can exert significant impacts on project performance, time and cost increase. In fact, the cost of rework on residential, industrial, and commercial building projects were estimated to range from 2% to 6% of their contract values (Josephson and Hammarlund, 1999). Similarly, Love and Li (2000), found rework direct costs to be 3.15% of the contract value in residential projects and Mills et al. (2009) found defects represent 4% of the contract value of the new dwelling or renovation. Consequently, the costs of defects reduce the profitability of the builder and the estate management organization. In addition, building defects can damage the reputation of the builder and reduce customer satisfaction (Sommerville and McCosh 2006; Forcada et al. 2012). Therefore, building defects impose significant impacts on industry and society, and are a critical issue to be addressed. 5 Methods For the purpose of the research reported in this paper, the definition of the term “defect” proposed by Watt (1999) is adopted: “Defect is the term used to define a failing or shortcoming in the function, performance, statutory or user requirements of a building, and might manifest itself within the structure, fabric, services or other facilities of the affected building.” (Watt 1999) Regarding the stage were defects occur and are manifested, different terms are used:  ‘Construction defect’ is defined as “that defect that is manifested during the construction stage” (Forcada 2014).  ‘Handover defect’ (i.e. 'snags' in the UK) is defined as “that defect which is absorbed during the construction/building process and which is usually corrected before practical completion; and, that which is “visible” to the contractor and home buyer once the home is deemed ready for occupation“ (Sommerville and McCosh 2006).  ‘Posthandover defect’ is used to describe “that defect which is still remaining after handing over the building but only during the liability period, which usually lasts 12 months” (Forcada et al. 2013a).  ‘latent defect’ is used to describe “that defect that appear during the occupancy of the Building” (Chong and Low 2006). Handover defects data were collated from handover check forms for 16 developments from one of the largest Spanish building company’s database. 6 To enable the analysis of data and extraction of conclusions, information must be organized, possibly re-formed and expanded where necessary (Georgiou 2010). Therefore, the original structure of handover checks used by the company was analyzed and adapted to the standardized classification system used by Forcada et al (2013a, 2014). Similar approaches have been adopted by other authors such as Trotman (1994), Georgiou et al. (1999), Olubodun and Mole (1999), Mills et al. (1999), Chew (2005), Chong & Low (2005 and 2006) and Craig et al. (2010). Building characteristics include gross floor area (GFA), number of flats and construction cost. Handover defects include the type of defect (e.g. appearance, stability/movement, etc.), the affected building element (e.g. internal wall, window, etc.) (Watt, 1999) and the respective trades (e.g. foundations, coatings, etc.). These data were used to:  Determine the most common defect types  Determine the distribution of defects by building element  Analyse the influence of the building element on the defect type  Determine the distribution of defects by trade  Analyse the influence of the trades on the defect type  Compare the nature of defects and the building elements where these defects are detected with those detected during construction and at handover. When selecting data source, an important consideration is to minimise the subjectivity of inspections, accuracy and reliability. To reduce the variation in subjectivity, this study uses data from the same database and from the same inspectors. However, caution should be taken when generalising these results. A limitation of this study is the data 7 capture and data source are derived from one main contractor. Although 2,179 dwellings are analysed, all of them were constructed by the same contractor and therefore may not be representative of the whole Spanish housebuilding industry. It should also be noted that some contractors who are responsible for registering inspection results might neglect to register some of them. The standardized format used by the company for the handover check forms and its translation to the standardized method adapted in previous studies (Forcada et al (2013a, 2014) increased the accuracy and reliability of the analysed data. The data was analysed using the Statistical Package for the Social Sciences (SPSS) for Windows (Version 17.00). Chi-square (χ2) test was used to determine the dependence between the building element and the trade and the defect type. This test allows comparison of the observed and expected frequencies. For a chi-square test, the null hypothesis is that the two sets of frequencies (i.e., observed and expected) are equal. The alternative hypothesis is that they are unequal. To identify those variables with significant correlations at the 95% confidence intervals, the asymptotic significance should be less than 0.05. Data Collection A total of 52,552 handover defects from 16 building developments were identified and analysed. The number of dwellings within each of these 16 developments ranged from 60 to 369. The size of the developments ranged from 6,270 - 41,697 m2. All building projects were private construction projects undertaken by a major contractor and there were no temporary joint ventures with other contractors during the project. The cost per 8 development ranged from €4,493,447 to €23,449,039. Table 1 identifies the main characteristics of the analyzed developments. Insert <Table 1. Building characteristics> Results The analysis of the defect data revealed that the most common defects, as noted in Table 2, were: ‘surface appearance’ (64.5%), ‘tolerance errors’ (9.3%) and ‘affected functionality’ (6.8%). ‘Surface/appearance’ defects include colour, type, uneven surfaces, hit and scratches, peeling and cracks. ‘Tolerance errors’ include those dimensional errors in pavements such as parquet flooring, woodwork, etc.‘Affected functionality’ defects refer to disabled building elements or systems that must be replaced because their functionality is completely affected. Typical ‘affected functionality’ defects during handover include problems with the boilers, noises in heating tubes, sockets located in not accessible places, impossibility to open the door of the fridge, door scrapes on the floor, etc. Insert <Table 2. Handover defects by type of defect> Analysis of Defects by Construction Element Tables 3 and 4 present the distribution of defects by building element. The results show that ‘internal walls’ (59.9%), ‘windows’ (17.2%) and ‘mechanical and electrical Systems (M&E)’ (8.5%) were the building elements where most defects arose. Table 4 presents the results of a Chi-square (χ2) analysis that sought to determine the independence of the defect type and the respective building element. The analysis revealed that the defect type and building element variables were not independent (p < 0.05). 9 Insert <Table 3. Defects by building element> Insert <Table 4. Building element and defect type> Insert <Table 5. Chi-square test of independence: Building element and defect type> Results revealed that the most common ‘surface appearance’ defects were mainly detected in interior walls, pavements and doors. For interior walls, the most common ‘surface appearance’ defects were related to color, dents, scratches and uneven surfaces. For floor surfaces, defects in the polish and stains were also very common. Spilled paint and chemicals caused most of the stains. Other complaints were related to plaster work on uneven walls and ceilings and protruding joints. Most of the ‘inappropriate installation’, ‘missing item or task’ and ‘tolerance errors’ defects were mainly identified in interior walls. The ‘inappropriate installation’ defects were mainly related to setting out the walls and inadequate dimensioning of wall elements such as the joint between the wall and the floor slab, while ’missing item or tasks’ were mainly due to by the lack of the second coat of paint. The majority of the ‘affected functionality’ defects were mainly related to door and window locks, handles or doorbells. Aluminium frames and glass were also parts of doors and windows with ‘affected functionality’ defects. ‘Water problems’ were mainly detected in P&S systems, and they took the form of leaking pipes, goods not plumbed in or pipes not earth bonded and in the roof. However, results revealed that ‘water problems’ only accounted for 3.2% of the construction defects. The most important M&E defects were mainly ‘detachment’ of electrical and mechanical elements such as tubes, pipes, fluorescents and sockets. 16 waterproofing membrane and concrete work) or design (failure to provide moisture barriers) (Chong and Low, 2005). However, defects may also arise from poor maintenance, degradation or local conditions ,(meteorological or climatic conditions or settlement such as stains caused by moisture from rain, dirt from occupants, capillarity, cracks, dampness or efflorescence. In relation to M&E systems, customers might not initially detect improper functioning but this may become apparent over time and use (e.g. actual energy consumption is much higher than initial predictions). This gap might be caused by poor management of systems but often is a result of design and/or installation problems. Addressing these latent defects and comparison can be done through a systematic adoption of postoccupancy evaluation. This has the potential to integrate actual performance and satisfaction with the remediation of defects. Conclusions Although inspection can occur during the construction stage of residential buildings, not all defects are addressed prior to handover. Therefore, the rework entailed by these defects has an inconvenient and negative impact on efficiency, productivity and competitiveness. The detailed analysis of 2,179 flats in Spain identified that the most common defects detected at handover are ‘surface appearance’ (64.5%) including colour, type, uneven surfaces, dents and scratches, peeling and cracks; ‘tolerance errors’ (9.3%) such as dimensional errors in pavements for instance parquet flooring, woodwork; and ‘affected functionality’ defects (6.8%) such as problems with the boilers, noises in heating tubes, sockets located in not accessible places, impossibility to open the door of the fridge or door scrapes on the floor. 17 The results also show that ‘internal walls’ (59.9%), ‘windows’ (17.2%) and ‘M&E systems’ (8.5%) were the building elements where most defects arose. Comparing the results of this study with those obtained from the analysis of construction defects (Forcada et al 2014) and post-handover defects (Forcada, et al., 2013), it can be concluded that, although many defects during construction are similar to those detected at handover stage, they are different in nature. These results also revealed that structural defects and inappropriate installation of roofs and façades during construction are resolved at during the construction stage due to existing quality standards, while other aesthetic and functional defects remain and/or arise at handover. These remain unresolved until customers make complaints when they first occupy the dwelling. This forces reluctant subcontractors to return to the site. Finally, although minor in nature, many functional defects are only detected by customers due to the lack of involvement of end users at the first stages of the project. This study provides evidence that the special characteristics of the housing sector, with inexperienced customers, quality standards set and managed by the contractors, lack of registration and licensing of subcontractor trades and high levels of standardization contribute to inadequate quality inspections. Quality construction regulations and certification exists. However, adaptation of these regulations to the real needs of the sector, emphasizing functional aspects and the involvement of customers at the very beginning of the process, are needed to improve the quality of housing and customer satisfaction. The identification and comparison of the typical nature of defects, building elements and trades where defects arose in residential buildings provides useful information about those areas where builders are likely to make errors, mistakes or deliberately take short-cuts during construction. 18 Further research will be focused on investigating the financial implication of defects identified at different stages of the lifecycle of new residential buildings. This will enable an understanding of the severity of the problem of defects on the performance of house developer organization, and consequently on the productivity of the construction industry. At present, an accurate estimate of the financial impact of defects does not exist in Spain. An estimating model that can accurately calculate the cost of defects to both the house building sector and the wider construction industry would be beneficial. Further research will also focus on analysing residential latent defects (for buildings more than 2 years old) where construction defects caused by poor workmanship or poor material performance might become visible. Tables Table 1. Building characteristics Development Number of flats Size (m2) Cost [€]/development Development 1 104 12,896 11,800,000 € Development 2 100 14,253 17,299,000 € Development 3 113 14,916 13,200,000 € Development 4 80 11,760 7,466,000 € Development 5 172 21,151 10,379,000 € Development 6 135 22,465 23,449,039 € Development 7 138 14,766 13,401,303 € Development 8 235 25,145 19,556,314 € Development 9 60 6,270 5,996,021 € Development 10 141 17,343 10,699,328 € Development 11 132 15,708 12,886,381 € Development 12 369 41,697 19,695,986 € Development 13 72 8,064 5,566,032 € Development 14 128 9,085 8,324,077 € Development 15 128 9,342 11,041,593 € Development 16 72 6,946 4,493,447 € 19 Table 2. Handover defects by type of defect Defect Type Number of defects % Surface appearance 33,890 64.5 Tolerance errors 4,905 9.3 Affected functionality 3,559 6.8 Detachment 2,960 5.6 Inappropriate installation 2,424 4.6 Missing item/task 2,011 3.8 Water problems 1,705 3.2 Soiled 575 1.1 Misalignment 337 .6 Broken/deteriorated 145 .3 Flatness and levelness 41 .1 Total 52,552 100.0 Table 3. Defects by building element Element Number of defects % Internal wall 31,454 59.9 Window 9,017 17.2 Mechanical & Electrical Systems 4,474 8.5 Door 2,644 5.0 Plumbing & Sanitary Systems 2,015 3.8 Pavement 1,936 3.7 Furniture and Devices 505 1.0 General 377 0.7 Roof 77 0.1 Exterior wall 53 0.1 Total 52,552 100.0 20 Table 4. Building element and defect type Type of defect Building element Total Window P&S Systems General M&E systems Furniture and devices External wall Internal wall Pavement Door Roof Surface appearance 0 33 0 100 1 6 2 2 1 0 145 Soiled 0 1,365 136 204 0 0 0 0 0 0 1,705 Misalignment 2 1 0 20 1 0 0 1 16 0 41 Detachment 3 43 0 870 1 0 1,502 0 3 2 2,424 Tolerance errors 1,285 0 0 276 357 0 0 1 1,640 0 3,559 Stability 78 196 0 44 18 34 1,350 46 228 17 2,011 Missing item/task 2 373 0 0 1 0 4,279 250 0 0 4,905 Affected functionality 0 0 0 2,960 0 0 0 0 0 0 2,960 Inappropriate installation 223 0 0 0 0 0 113 0 1 0 337 Flatness and levelness 39 1 241 0 17 0 1 143 131 2 575 Water problems 7,385 3 0 0 109 13 24,207 1,493 624 56 33,890 Broken/deteriorated 0 33 0 100 1 6 2 2 1 0 145 Total 9,017 2,015 377 4,474 505 53 31,454 1,936 2,644 77 52,552 21 Table 5. Chi-square test of independence: Building element and defect type Value df Asymp. sig (2tailed) Pearson chi-square 112,254.52a 99 0.000 Likelihood ratio 61,167.50 99 0.000 No. of valid cases 52,552 a 29 had an expected count of < 5. The minimum expected count was 0.25. Table 6. Defects by subcontractor Subcontractor Number of defects % Partitions and enclosures 26,835 51.2 Doors and windows 10,896 20.7 Facilities 6,488 12.3 Coatings 5,017 9.5 Pavements 1,840 3.5 Furniture and devices 1,268 2.4 General 158 0.3 Total 52,552 100.0 22 Table 7. Subcontractor and defect type General Facilities Furniture and devices Partitions and enclosures Pavements Doors and windows Coatings Broken/deteriorated 0 133 1 7 2 1 1 145 Water problems 0 1,569 0 136 0 0 0 1,705 Flatness and levelness 0 21 1 0 1 18 0 41 Inappropriate installation 0 913 1 1,502 0 6 2 2,424 Affected functionality 0 276 1120 0 1 2,162 0 3,559 Missing item/task 0 240 18 4 46 306 1397 2,011 Tolerance errors 0 373 1 4,279 205 2 45 4,905 Detachment 0 2,960 0 0 0 0 0 2,960 Misalignment 0 0 0 113 0 224 0 337 Soiled 158 0 17 2 226 168 4 575 Surface appearance 0 3 109 20,842 1,359 8,009 3,568 33,890 158 6,488 1,268 26,885 1,840 10,896 5,017 52,552 Ç 23 Table 8. Chi-square test of independence: Subcontractor and defect type Value df Asymp. sig (2-tailed) Pearson chi-square 92,208.13a 77 0.000 Likelihood ratio 60,809.48 77 0.000 No. of valid cases 52,552 a 18 had an expected count of < 5. The minimum expected count was 0.25. Table 9. Comparison of defect type among construction, handover and posthandover Construction period (Forcada et al. 2014) Handover Post-handover (Forcada et al. 2013) Inappropriate installation 24% Surface appearance 65% Missing item/task 37% Surface appearance 15% Tolerance errors 9% Surface appearance 19% Affected functionality 12% Affected functionality 7% Inappropriate installation 16% Missing item/task 12% Detachment 6% Soiled 10% 24 Table 10. Comparison of elements where defects are detected among construction, handover and post-handover Construction period (Forcada et al. 2014) Handover Post-handover (Forcada et al. 2013) Pillar 14% Internal wall 60% Fixture and fittings 19% Facilities 14% Window 17% Doors 15% Internal wall 12% Facilities 8% Windows 14% External wall 11% Door 5% Internal wall 14% References Asociación de empresas constructoras de ámbito nacional (SEOPAN). (2013). “Informe Económico 2013.” Madrid, Spain. Auchterlounie, T. (2009). "Recurring quality issues in the UK private house building industry." Structural Survey, 27(3), 241 - 251. Chew, M.Y.L. (2005). Defect analysis in wet areas of buildings. Construction Building Materials. 19(3), 165-173. Chong, W.K., & Low, S.P. (2005). Assessment of Defects at Construction and Occupancy Stages. Journal of Performance of Constructed Facilities, 19(4), 283-289. Chong, W.K., & Low, S.P. (2006). Latent Building Defects: Causes and Design Strategies to Prevent Them. Journal of Performance of Constructed Facilities. 20(3), 213-221. 25 Craig, N., Sommerville, J., & Auchterlounie, T. (2010, September). Customer satisfaction and snagging in the UK private house building sector. 26th Annual ARCOM Conference. Leeds, UK. Forcada, N., Macarulla, M., Fuertes, A., Casals, M., Gangolells, M., & Roca, X. (2012). Influence of Building Type on Post‐Handover Defects in Housing. Journal of Performance of Constructed Facilities. 26(4), 433–440. Forcada, N., Macarulla, M., & Love, P.E.D. (2013a). Assessment of Residential Defects at Post-Handover. Journal of Construction Engineering and Management. 139(4), 372– 378. Forcada N; Macarulla M; Gangolells M; Fuertes A; Casals M; Roca X. (2013b). Post handover housing defects: sources and origins, Journal of Performance of constructed facilities, 27(6): 756-762. Forcada N; Macarulla M; Gangolells M; Casals M (2014) Assessment of construction defects in residential buildings in Spain, Building Research & Information, 42(5): 629640. Georgiou, J., Love, P. E. D., & Smith, J. (1999). A comparison of defects in houses constructed by owners and registered builders in the Australian State of Victoria. Structural Survey. 17(3), 160-169. Georgiou, J., Love, P.E.D., and Smith, J. (2000). "A review of builder registration in Victoria". Structural Survey 18(1), 38-45 Georgiou, J. (2010). Verification of a building defect classification system for housing. Structural Survey, 28(5), 370-383. Hall, M., & Tomkins, C. (2001). A cost of quality analysis of a building project: Towards a complete methodology for design and build. Construction Management and Economics. 19 (7), 727–740 .