Measurement Uncertainty in Off-site Construction (OC) and Building Information Modelling (BIM)

Measurement Uncertainty in Off-site Construction (OC) and Building Information Modelling (BIM)

Literature Review

This section is reserved for the review of previous publications that offer empirical and theoretical evidence that can aid in this paper’s objectives of examining the role of measurement uncertainty in Off-site construction (OC) and Building Information Modelling (BIM). We will firstly consider the merits and drawbacks of OC, making known benefits explicit and outlining areas of concerns for OC.

It is necessary to comprehend the nature and sources of measurement uncertainty in the manufacturing process. Therefore, we will consider previous published literature concerning measurement uncertainty to gain an understanding on the sources of uncertainty, why they exist and potential remedial actions that can be taken to mitigate the effects thereof.

Off-site Construction

Off-site manufacturing is a broad term which is applicable to a large range of construction projects involving preassembled modules to be erected at a site separate to where manufacturing occurred. The structures are typically fabricated under controlled conditions in a factory with the view of them being transported and assembled thereafter. OC is an innovation within the construction industry that is said to increase efficiency and cost economics of a project. This section will review empirical reports to see if these beliefs are upheld by theory and evidence.

Oakley (2017) posits that OC conveys certain benefits and opportunities to the economy. They state that if the share of the UK construction sector utilising OC increases from the current level of seven percent to twenty-five percent, then productivity of the sector could increase by 3.6% by 2020. Moreover, a gravitation to a more OC centric industry could be a catalyst to equality within the United Kingdom (UK) with regions outside of the capital well poised to gain from OC hubs. This is because manufacturing centres can be positioned in areas of high unemployment. This is reinforced by Oakley (2017) by stating that “by boosting employment and utilising skills and expertise outside of the Capital, the approach would provide a spur to growth outside of London worth some £15 billion up to 2020”. This could attract the attention of governments who may lend support in the way of policies to encourage prefabrication factories. 

This viewpoint is underpinned by a government report which states, “The need to take action to improve the performance of the construction sector has been a recurring theme in the national policy debate for many years. For example, in 2013, the Coalition Government and construction industry put forward a joint desire to: Lower the initial construction and whole life costs of buildings by 33%;  Reduce the time from inception to completion for new build and refurbished assets by 50%;  Reduce greenhouse emissions in the built environment by 50%; and Improve export performance by 50%” (HM Government, 2013).

Benefits of OC

Perhaps the most pressing question that should be addressed is one that appraises the benefit to moving construction off-site when it has traditionally been located on-site. One of the alleged benefits of OC is reliability of construction delivery. On-site construction is subject to weather and site conditions in which an adverse environment can have implications on construction activity, whereas OC occurs within a factory where conditions can be controlled and held constant irrespective of seasonality. This allows for greater certainty on delivery time of the modules which will be construe as a clear benefit. This is underpinned by Simon Ambler, Director of the Portakabin Group who claimed, “Off-site construction is highly predictable, which helps construction clients reduce risk with far greater assurance of completion on time and on budget – often to extremely challenging deadlines and on complex, fully operational sites”. (Yorkon, 2015). It seems, from a report issued by Yorkon (2015) that the claims are justified given that, as of 2015, “Portakabin Group outperforms the construction industry with 99.7 per cent of projects completed on time and on budget (every year since 2003) in comparison to the industry average of 40 per cent delivered on time” (Yorkon, 2015). Based on this account, it seems that most of the major hinderances in construction projects will be circumvented by moving much of the process off-site to where there is a greater control over the manufacturing process. According to KPMG (2016), subsequent benefits of improved quality of OC includes avoiding costs and time lost associated with re-design and re-work. They continue by intimating that offsite methods exhibit a fifty percent reduction in the costs of projects that are subject to “snagging” (KMPMG, 2016). Prefabrication will reduce the number of elements that are required to be checked on-site which will save time and resources. This, of course, is only realised as a benefit if the prefabricated produce was manufactured to the required standard.

Other benefits that are offered by pro-OC commentators is the reduction in on-site injury from operations. HSE (2018) posits that the overall cost in 2016/17 that injury conveys was estimated at £1,062 million.  This is a significant cost to the industry and can be remedied, based on this account, by outsourcing more of the process to the OC model. The HSE enlists several factors that can mitigate risk of injury that OC offers. The more relevant being operations being conducted in a clean and controlled environment which can conform to minimalistic standards and technique adoption available on a production line. OC also reduced the exposure of risks associated from working from height or below ground and reduces the risk of workers suffering from UV related illness.

A key aspect of OC is that productivity can be optimised due to the nature of the production chain permitting multiple endeavours being conducted concurrently which reduces overall build time. This view is considered accurate as “by increasing the level of offsite activity in the factory environment less time will be spent on-site because activities can be run concurrently rather than sequentially such as the placement of foundations while components are being assembled” (Hairstans, n.d.). Traditionally, certain parts of the project may be delayed as they await site-specific prerequisites being in place, however, this is becoming a lesser concern as the nature of the OC manufacturing process allows parts to be completed irrespective of the status of other fabricated parts.

The nature of the produce derived from OC is said to be more energy efficient than buildings from more traditional construction methods. This is because “Offsite construction techniques have the potential to reduce energy-in-use because the finished quality of the buildings is generally to a higher standard. Examples include structural quality (leading to improved air-tightness, for example)” (Krug, 2013). They continue to claim that reduced energy-in-use can be as high as twenty-five percent over the asset life which is a significant benefit to using this method of construction.

The Steel Construction Institute conducted research on construction projects and identified “benefits of modern methods of construction with particular regard to the disruption caused to local residents from dust, noise and commercial vehicle movements and the environmental impact of site-generated waste” (Taylor, 2009). Presumably, some of the other most relevant environmental benefits will be a function of less on-site diesel generator emissions and prevention of pollution in urban areas. Residents in existing areas of natural beauty may be heavily resistant to large scale construction works in their area due to the preconceived expectations of by-products of on-site activities, however, prefabricated units will help to ease concerns of this nature due to the volume of the process which is now occurring in separate locations.

One of the most valuable allures for gravitating from on-site construction to OC is the reliability of results and quicker turnaround times.  “Off-site construction is highly predictable, which helps construction clients reduce risk with far greater assurance of completion on time and on budget – often to extremely challenging deadlines and on complex, fully operational sites”. (Yorkon, 2015). This is especially important given the track record that the industry has shown over time. “A 2016 survey showed that just 68% of projects finished within budget and only 41% came in on time or better” (Glenigan, 2016). This is a very concerning statistic as there is more projects failing to complete within deadline than there are projects that do complete in advance of the deadline. Should a large proportion of the delays stem from environmental factors, then one can expect these figures to markedly improve if the OC model is integrated into the projects hereafter.

OC can have a cost reducing effect on the overall delivery of the project due to benefits of the learning curve. This is because in lieu of construction occurring on-site where different companies are completing different projects, the same company can continually fabricate modules and benefit from learning from experience. Inefficiencies present for inexperienced firms are no longer a factor as the same companies are continually manufacturing similar units and structures. An example offered by a Laing O’Rourke spokesperson is “Today, building a nuclear station is an intensive on-site operation. But if we can move a lot of that work to factories and work in a more modular way, we’d see major benefits, not just in terms of time and costs, but also in terms of establishing quality systems, improving learning and health and safety. Of course, we have to be realistic about what can be done, but this has to be the direction of travel. Japan has already demonstrated that it’s possible to work in this way and get big improvements. In South Korea, this approach is now the standard. They’ve cut about 30% off the cost of their reactors in real terms over 10 years by building a reactor every year, using the same team, and setting about it in this systematic and progressive way” (Davis, 2013). Given the success mentioned in the example, it is conceivable that similar results can be experienced within the UK construction sector. Moreover, efficiencies can be reduced by iterative improvements through pressures of competition. If international firms can supply the UK market with prefabricated goods, it will encourage incumbent companies to increase efficiency.  This is a viable concern given the high value of parts which makes transportation from adjacent countries economical. “The general rule for anything that is manufactured in volume is that it gets better in terms of quality and performance and cheaper in terms of affordability over time. This is because volume attracts competition and suppliers then invest to achieve competitive advantage. Competitive advantage typically comes from 3 routes; cost leadership, differentiation and focus, or expressed differently product, price and service” (UK Government, 2015). Both competitive forces and the learning curve can help to increase efficiency in the sector which will ensure benefits resonate throughout the economy.

Drawbacks

Despite the embellished perception that some engineers may hold, there are some aspects of the OC process that can be construe as being undesirable and welfare reducing for project stakeholders. These challenges will be outlined hereunder so that a comprehensive and balanced view of OC can help appraise the overall value of moving from an on-site to off-site construction process.

In the event of a boom within the OC manufacturing sector, it is feasible that agents enter the market who lack the dexterity and ability to manufacture units at an expected standard. An official report published by the UK Government states that “as more manufacturers enter the market, there is a risk that some may not deliver homes to an acceptable standard and this has the potential to impact on the acceptance of this type of housing from a customers and a funders perspective” (UK Government, 2015).  Situations in which subpar units are provided will likely be a source of project delays, increase costs and tarnish the reputation of manufactures if the problem is seen on a wide scale.

A cardinal criticism of prefabricated units is the life-expectancy of the structure. A publication from the UK Government lists this as a drawback to OC and raises concerns on the impact that this may have on securing funding for purchases of finished units. The report states “Depending on the form, some units have a shorter life expectancy. This raises questions on how such units might be treated from a funding perspective, such as:  If they are seen as depreciating assets, it may push their value down. This would need to be weighed up against any cost efficiencies gained in the construction. Funders generally require a dwelling to have a lifespan of the length of the loan being secured plus 30 years. This could mean as dwellings age they become more difficult to raise finance on (or are even rendered unmortgageable) and their values decrease, much like leasehold properties (which have a finite and reducing lifespan).” (UK Government, 2015). Although largely specific to houses, these concerns are valid and may, unless corrected, retard growth in the demand for modular houses.

Accord Housing Association (n.d) continues to explain that the OC method of construction is disadvantaged in the sense that there is a distinct absence of supervision for the contractor as they are unable to observe or supervise any of the off-site process. This prevents them from checking standards and specifications during the manufacturing process. Such barriers would not be evident if the whole process was conducted on-site as the contractor could make regular site visits to ensure contentment with the standard of fabrication and progress being made (UK Government, 2015). This may discourage some purchasers who may regard ongoing control and supervision of the project throughout as the costs of travelling to the location where construction is being completed may be uneconomical.

Other disadvantages are damage or loss of units in transit.  Transporting by road has the risk of damage from collision and risk of damage could be a more prominent issue for parts imported internationally where the units are conveyed over water. This is as loss of a transportation shipping vessel could result in large disruption to the project through delays whilst alternative substitutes are manufactured and delivered to the site. Furthermore, Kyjakova and Baskova (2016) cite that security can pose an issue as prefabricated parts are stored onsite which may entice thieves. These scenarios are all possible drawbacks to the OC approach as units are damaged or stolen during transit or in storage.

Further, contractors are exposed to non-delivery risk as if a supplier liquidates, it may be very difficult to find an alternative company who is able to manufacture units by the exact same specifications as previously agreed with the insolvent company. “Manufacturer insolvency during the course of a project would have major implications for modern methods of construction. If the design and production of many building components, particularly pods and panels, are specific to one manufacturer then the choices are to alter the project substantially or to seek another manufacturer prepared to make a compatible product. Either action is costly and causes delay. The effect of insolvency would have greater impact the later into a development it occurs” (National Audit Office, 2005).  Any project could be exposed to a partner reneging on their obligations due to insolvency but the specific risk, as mentioned, is if a design is unique to the manufacturer that has become insolvent. This is a significant risk and challenge that OC faces unless a predetermined contingency provision is put in place at the contractual stage.

Critiques of OC may claim that standardising the manufacturing process may lead to restrictions on the designer’s creativity, leading to mundane and uniform appearances of the structures. However. Osborne Group Holdings conjectures that, “In terms of creative design off-site requires a significant change in mindset to customary design practice. Architectural and structural designs have to be completed and signed off much earlier in the project development process. Once the design is fixed the off-site element makes it far more difficult to change during the project implementation and delivery phases. There is still substantial opportunity for creative design it simply needs to be done at the outset and not drip fed and continually amended through the building phase” (UK Government, 2015). This statement makes explicit that concerns over lack of design liberty are not well founded provided that any aesthetic features are disclosed at an early stage. The drawback is flexibility in changing one’s mind once the process is underway.

The Problem of Measurement Uncertainty

Bell (1999) outlines measurement uncertainty as “the doubt that exists about the result of any measurement. You might think that well-made rulers, clocks and thermometers should be trustworthy, and give the right answers. But for every measurement – even the most careful – there is always a margin of doubt. In everyday speech, this might be expressed as ‘give or take’ … e.g. a stick might be two metres long ‘give or take a centimetre”. 

The notion of measurement uncertainty is captured by the viewpoint that no measurement is necessarily perfect and exhibit multiple sources of variation. A best estimate of measurement uncertainty provides a range of measurements in which one assumes the true value resides within the range. By the law of large numbers, we can surmise that if the sample of measurements is great enough and each measurement procured is plotted in a histogram or related plot, a pattern that resembles the normal distribution bell curve should be prominent. A best-case estimate for the true measurement value will be to equate it to the mean of the sample measurements.

As a means to remedy the effects of measurement uncertainty, users have used ranges and confidence intervals. For example, an object’s length could be declared as being 20cm ± 1mm at a 95% level of confidence. This is shorthand way of stating that, if this object was to be measured one hundred times, we would expect the true value to sit within the interval ninety-five times.

Simply put, measurement uncertainty is a crude quantification of uncertainty of the true value of a measurement. This concept is explained by White (2008) as “The dispersion of results obtained from such repeated measurements (imprecision) can be described approximately by a normal probability (Gaussian) distribution, with some 95% of the results falling within ± 2 standard deviations (SD) of the mean value”. This makes the problem of the lack of measurement certainty in OC a severe one given that small margin of errors on various units will not be evidence until they are transported and erected on-site.

Other factors can impact the measurement uncertainty issue. For instance, a theoretical versus geometric problem can present in which there is a discrepancy between designed specifications and those realised after the manufacturing process has concluded. The unit must first be modelled mathematically using the elected design tools, thereafter, the unit can be manufactured. During this process the parts can be subject to imperfections in the tools used for construction and/or the materials themselves which can lead to deviations from those specified during the design phase.

Said imperfections are an important consideration given that anomalies are prominent as some degree of imperfection will present in every object, no plane is perfectly flat nor a sphere perfectly spherical. This type of imperfection is referred to as a ‘form error’ which can be explained as “a function of the deviation vector. It is important to note that the defining parameters of the reference surface are not always computed by minimizing the zone of the deviation vector. However, after the defining parameters of the reference surface are determined, i.e., the reference surface is determined, the form error is declared to equal the zone of the deviation vector” (Gosavi, n.d.). The practical implications of a form error are that engineers may request a perfectly straight part but, due to form error, the part dispatched is imperfect and not truly straight as required.

A second source of measurement uncertainty is measurement error which typically occurs when data is collected from an object’s surface by a measuring instrument. Fridman (2012) suggests that the effectiveness of a measurement is contingent on the precision of the measurements taken. This is explained as the proximity of the measured value from the true value(s). Measurement errors can be a function of issues with the measuring instrument such as bends in the hardware or probe system imperfections or can derive from the environment such as temperature or humidity differentials. White (2008) goes further to contest that “Measurement results are thus unreliable and should be regarded as best estimates of the true value of the quantities being measured”.  This proves problematic for prefabricated units that have been designed with the intention of fitting in exact spaces or aligning with other structures.

Sources of Measurement Uncertainty

Various sources of measurement uncertainty are known and will be described in this section.

Random Errors

Random errors are such that a measurement discrepancy is observed with the source of the error being unknown. Because unknown, it cannot be eliminated. One can only converge the error ranges by repeating the measurements a number of times whilst holding all other conditions constant. The measurements obtained will likely differ for various reasons already discussed, however, absent any guaranteed precision, the dispersion of measurement scores should differ arbitrarily which will equate to them following a normal distribution. We can assume that a best estimate for the true value will be approximately equal to the mean measured value. By taking a large sample of measurement and accepting the mean as the true value, the effects of the random error are significantly reduced. In large samples of measurements the mean value may tend to the true value to such a degree that the error can be non-existent (Kharagpur, n.d.).  A contrasting measurement uncertainty is the systematic error. “Systematic errors are reproducible inaccuracies that are consistently in the same direction. These errors are difficult to detect and cannot be analyzed statistically. If a systematic error is identified when calibrating against a standard, applying a correction or correction factor to compensate for the effect can reduce the bias. Unlike random errors, systematic errors cannot be detected or reduced by increasing the number of observations” (Anon, 2011). These distinctions are important as “error is the actual true difference between the measured data and the true value. One never knows the true value or the true error, but can only estimate its limits. This is called the uncertainty. In estimating the limits of a systematic error, use systematic standard uncertainty. For random error, use random uncertainty” (Dieck, 2017).

Calibration

As aforementioned, random errors cannot be eliminated due to factors already said. Calibration is a tool that can be employed to help mitigate measurement uncertainty. Calibration is a process in which “a known input signal or a series of input signals are applied to the measuring system. By comparing the actual input value with the output indication of the system, the overall effect of the systematic errors can be observed. The errors at those calibrating points are then made zero by trimming few adjustable components, by using calibration charts or by using software corrections” (Kharagpur, n.d.). This is an example of where technology can be deployed to ratify known impurities in the manufacturing process, however, calibration can also be a source of measurement uncertainty in which the instrument produces a value that is different from the true measurement value.

Data Collection

Measurement uncertainty can be a product of data collection errors too. “Data acquisition errors are those that result from the use of data acquisition equipment, which may be computer data acquisition systems, a data logger, or just a person with pencil and paper reading a meter. Acquiring data to estimate the magnitude of these error sources is usually simple. All that is needed is to take multiple readings over the appropriate time period. This category of errors is also usually smaller in magnitude than calibration errors by a factor of three to ten. As with calibration errors, the exact magnitude of these errors is not known; only their limits may be estimated: systematic and random uncertainties” (Dieck, 2017).

This section has explored different sources of measurement uncertainty which can all feasibly cause disruption in the OC process. The main issue being that it may not be evident until workers that are overseeing installation on-site notice discrepancies that prevent efficient joining of the modules.  The nature of some of the different types of measurement uncertainty can make them difficult to identify and correct until a late stage in the project.

References 

• Anon, (2011). ‘Measurements and Error Analysis’. Advanced Instructional Systems, Inc. and the University of North Carolina, USA.
• Bell, S. (1999). ‘A Beginner’s Guide to Uncertainty of Measurement’. Available Online: https://www.dit.ie/media/physics/documents/GPG11.pdf. [Accessed 7 January 2019].
• Davis, G. (2013). ‘The future of DfMA is the future of construction’. Available Online: www.laingorourke.com/~/media/lor/…/lor-engineering-excellence-journal-2013.pdf {Accessed 7 January 2019]. 
• Dieck, R. (2017). ‘Measurement Uncertainty: Methods and Applications’. Research Triangle Park, NC 27709, USA.
• Fridman, (2012). ‘The Quality of Measurements’. Available Online: < https://www.springer.com/gb/book/9781461414773>. [Accessed 5 January 2019].
• Gosavi, A. (n.d.). ‘Form Errors in Precision Metrology: A Survey of Measurement Techniques’. Available Online: http://web.mst.edu/~gosavia/metrology_survey.pdf. [Accessed 11 January 2019].
• Hairstans, R. (n.d.) ‘Building Offsite An Introduction’. Available Online: http://www.forestryscotland.com/media/320961/building_offsite_an_introduction.pdf [Accessed 10 January 2019].
• HSE, (2018). ‘Construction statistics in Great Britain, 2018’. Available Online: http://www.hse.gov.uk/statistics/industry/construction.pdf [Accessed 4 January 2019].
• Kharagpur, I. (n.d.). ‘Measurement Systems’. Available Online: < https://nptel.ac.in/courses/108105063/pdf/L-10(SS)%20(IA&C)%20((EE)NPTEL)%20.pdf>. [Accessed 4 January 2019].
• KPMG, (2016). ‘Smart construction April 2016 kpmg.com/uk/ibc How offsite manufacturing can transform our industry’. Available Online: https://www.buildoffsite.com/content/uploads/2016/08/KPMG-Smart-Construction.pdf [Accessed 4 January 2019].
• Krug, D. (2013) ‘OFFSITE CONSTRUCTION: Sustainability Characteristics’. Available Online: https://www.buildoffsite.com/content/uploads/2015/03/BoS_offsiteconstruction_1307091.pdf/ [Accessed 7 January 2019]. 
• KYJAKOVÁ, L & BAŠKOVÁ, R. (2016). ‘ADVANTAGES AND DISADVANTAGES OF MODERN METHODS OF CONSTRUCTION USED FOR MODULAR SCHOOLS IN SLOVAKIA’. Available Online: < https://docplayer.net/41650199-Advantages-and-disadvantages-of-modern-methods-of-construction-used-for-modular-schools-in-slovakia.html>. [Accessed 4 January 2019].
• National Audit office, (2015). ‘Using modern methods of construction to build homes more quickly and efficiently’. Available Online: https://www.nao.org.uk/wp-content/uploads/2005/11/mmc.pdf [Accessed 7 January 2019].
• Oakley, M. (2017). ‘THE VALUE OF OFF-SITE CONSTRUCTION TO UK PRODUCTIVITY AND GROWTH’. Available Online: http://wpieconomics.com/WPI-Economics-Value-of-Off-Site-Construction-April-17.pdf [Accessed 10 January 2019] 
• Taylor, S. (2009) ‘Offsite Production in the UK Construction Industry – prepared by HSE’. Available Online: https://www.buildoffsite.com/content/uploads/2015/04/HSE-off-site_production_june09.pdf [Accessed 7 January 2019].
• UK Government, (2013). ‘Off-site manufacture for construction: Building for change Oral and Written evidence’. Available Online: < https://www.parliament.uk/documents/lords-committees/science-technology/off-site-manufacture-for-construction/off-site-manufacture-construction-ev.pdf> [Accessed 10 January 2019]. 
• White, G. (2008). ‘Basics of Estimating Measurement Uncertainty’. Available Online: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2556585/. [Accessed 7 January 2019].
• Yorkon Online, (2015). ‘Project delays in uk construction continue to worsen’. Available Online: https://www.yorkon.co.uk/news-and-views/project-delays-in-uk-construction-continue-to-worsen/ [Accessed 10 January 2019].