THE EFFECTIVE APPLICATION OF BIM ON CONSTRUCTABILITY PROBLEMS

Abstract

Constructability is an integral part of construction as it defines how we build. Part of designing a building must consider how it will be built. Over the years building techniques have improved to allow us to build bigger and bigger. Therefore, improving the constructability of our designs will allow us to build more efficiently, using less material, less time and less people. With advancements in computer technology the construction industry is changing in the way it designs. In 2011 the Government Construction Strategy mandated the use of Level 2 BIM on all public sector projects by 2016 (Digital Build Britain (2015)). Advancements such as these encourage further improvements to design. One such improvement would be the investigation of how applicable BIM is to reducing constructability problems.

This study has classified and analysed constructability issues on a case study project in Oxford using data from RFIs on the project. Through secondary research it has then assessed how applicable BIM can be in addressing these issues, commenting on both the benefits and limitations of BIM.

Table of Contents

Abstract………………………………………………………

Acknowledgements………………………………………………

Table of Contents……………………………………………….

1. Introduction……………………………………………..

2. Literature Review…………………………………………

2.1. A brief history of BIM……………………………….

2.2. Constructability…………………………………..

2.2.1. A process approach………………………………

2.2.2. Case Study and Survey Research methods……………..

2.2.3. Prefabrication and Modularisation…………………….

2.3. BIM on Constructability……………………………..

2.3.1. 3-D Modelling…………………………………..

2.3.2. 4-D modelling…………………………………..

2.4. Conclusion………………………………………

3. Research Method…………………………………………

3.1. The Case Study…………………………………..

3.2. RFI Categorisation…………………………………

3.3. BIM Applicability………………………………….

4. Findings………………………………………………..

4.1. Initial Breakdown………………………………….

4.2. Constructability Categorisation………………………..

4.3. RFI Origins………………………………………

4.4. Design Coordination……………………………….

4.5. BIM Applicability Matrix……………………………..

5. Discussion………………………………………………

5.1. Constructability vs Project Impacts……………………..

5.2. BIM as a Solution………………………………….

5.3. Limitations of the Study……………………………..

6. Concluding Remarks and Recommendations…………………….

7. References………………………………………………

8. Appendices……………………………………………..

1.   Introduction

Constructability has long caused problems in the construction industry as a result of a knowledge gap between the design and construction stages of a project (Fischer, M & Tatum, 1997). Improving constructability of designs has been the subject of much research in the last two decades, with particular emphasis placed on providing construction information within the design stage (Odeh, A.M. (1992)). A reason for this is that constructability is a major factor to the success of construction projects (CII (1993)). Problems occur when designers do not consider how components of buildings will be constructed or installed on site. Constructability problems have existed as long as construction has; Ancient Egyptians considered constructability of the pyramids at the design stage to plan the huge movements of stone to its peak (Edwards, J. F. (2003)).

Not all constructability issues are of this scale, smaller constructability issues often pass through the design stage unnoticed, causing delays and additional costs to the project (Fischer, M & Tatum, 1997). Recent advances in the UK health and safety industry have contributed to more constructability issues which previously did not exist. For example, the use of 950mm edge protection is mandatory for all leading edges on a construction project (HSE, 2008). This may prevent a worker from accessing ‘hard to reach’ areas. Before this safety requirement legislation, workers could dangerously gain access to such areas to complete the works. Today, additional design considerations are required to avoid causing overall project delays through inaccessibility.

Improving the design to eliminate or mitigate against these issues should be a focus of the construction industry in order to reduce project delays and unforeseen costs. The introduction of Building Information Modelling (BIM) in the last twenty years has given a new means to solving constructability problems. BIM can be defined as ‘A process for creating and managing information on a construction project across the project lifecycle. One of the key outputs of this process is the Building Information Model, the digital description of every aspect of the built asset. This model draws on information assembled collaboratively and updated at key stages of a project. Creating a digital Building Information Model enables those who interact with the building to optimize their actions, resulting in a greater whole life value for the asset.’ (NBS, 2017). In application, BIM is used to support better design through improved clarity and presentation of information. As part of this improvement BIM can help address some constructability problems through features such as clash detection and visualisation (Codinhoto et al 2011).

The aims of this study were to identify constructability issues on the Big Data Institute, Headington, Oxford. This was completed through the compilation and analysis of all ‘requests for information’ (RFIs) on the project. RFIs were chosen as they lie within close proximity to the later design stages and construction stage of the project. In total there were three hundred and thirty-seven RFIs, not all of which were constructability related. Through secondary research, a comprehensive literature review was used to define different constructability categories. RFIs were sorted into these categories and then analysed using additional classifications. The second aim of this research was to identify the features and functions of BIM that currently benefit construction projects. This was completed through secondary research and the use of a literature review to identify said features of BIM. Finally, the study aimed to understand how applicable these identified features and functions are to the constructability problems identified in the original analysis.

2.   Literature Review

2.1.         A brief history of BIM

The origins of Building Information Modelling (BIM) are unclear with so many technological advances having been made in the last fifty years. BIM was theorised as early as 1962 by Douglas C. Engelbart who envisioned a design where the following occurs, “When he has finished, the revised scene appears on the screen. A structure is taking shape. He examines it, adjusts it…”, this vision has evolved into a similar reality. Since, there have been many advances in technology within construction, one of the most notable has been Computer Aided Design (CAD). Whilst BIM is heavily involved with CAD it goes far beyond 3-D modelling, as explained in Section 1.

Much of the literature reviewed doesn’t specifically refer to BIM, this is due to the evolution of computer usage and the associated names given to computational software in construction. BIM along with CAD and other related computational terms are inexplicably linked within the reviewed literature in terms of the impacts they have in the construction industry. Figure 1: UK BIM maturity model, (http://bim-level2.org/globalassets/pdfs/bim-maturity-model.pdf ).

The last twenty years have seen BIM evolve hugely in its capabilities, with little to suggest it slowing down. Due to the vast number of influencing factors the evolutionary path of BIM is somewhat unclear. Figure 1 can be used to aid the description of how BIM has evolved, depicting the BIM levels of maturity. Taking an alternative, financial approach to assessing BIMs growth, analysing share prices of the principle BIM providers gives a good indication of the BIM industries growth. The main providers of BIM software are Autodesk and Nemetschek, between they own: Revit, AutoCAD, Graphisoft and ArchiCad. Both Autodesk and Nemetschek have seen a sustained rise in their stock price over the last ten years as how n in Figures 2 and 3.

Figure 2: Autodesk stock price over the last ten years. (Appendix B).

Data sourced from: www.nasdaq.com

 

Figure 3: Nemetschek stock price over the last ten years. (Appendix B).

Data sourced from: https://www.nemetschek.com/en/investor-relations/stock/stock-charts/

This growth supports the success and growth of BIM as an industry. Additionally, the growing stock values of these companies correlates to an increased investment in the product, which in this case is BIM software (amongst other software packages). These investments will help BIM develop further and have the potential to unlock yet more applications.

As BIM continues to evolve, research interests have grown, as shown in Figures 2 and 3. This systematic literature review was undertaken to review related research and identify the key issues and potential gaps within it. Google Scholar was used to conduct this research which consists of one hundred and ten journal articles and theses. Five different searching criteria were used to cover the differing aspects of the research field, Table 1 shows the search criteria and the corresponding number of articles identified and reviewed.

Table 1: Search Criteria for Google Scholar Literature Review

Word combination used	Number of entries	Number of entries recorded
BIM Constructability in Construction	2600	40
Disassembly using BIM	3400	24
Sustainability through BIM in Construction	13500	6
Constructability and Disassembly with BIM	123	22
BIM and Constructability design	2540	8

A set of qualitative parameters were used to eliminate irrelevant articles beyond the original searching criteria, these included:

• No inclusion of BIM with little reference to constructability.
• Avoid BIM awareness studies or BIM vs No BIM project studies.
• Avoid the use of BIM in education or teaching.
• Avoid BIM in past, present, future comparisons as a construction tool.
• Avoid studies with little reference to deconstruction or decommissioning in relation to BIM.
• Only use articles between 2000 and 2017 unless specific to constructability.
• When search results show less than 1-2 related articles per page move onto the next word combination (standard 10 results per page).

After tabulating the results of the literature review, many different categories within the research field were identified. Most of these categories related back to BIM (see Figure 4), however the secondary links between articles were weak, often with only BIM in common. This resulted in the breadth of field being quite wide. Articles were further divided into sub-categories, often linking to main categories with irregular transverse links. The main categories identified by the literature review were as follows:

1. Constructability
2. Modular Construction / Prefabrication
3. Deconstruction
4. BIM implementation
5. Sustainability

These main categories have many articles relating to each of them, most of which link to another main category and or sub-category. Sub-categories were identified within the literature, which aided the classification of many articles and provided greater detail into different areas within the field. Figure 4 highlights these subcategories and provides a visual aid into the links between articles and the most popular categories researched.

The literature review focused first on constructability, and the on the benefits of BIM to constructability. The categories of sustainability and BIM implementation were dependent on constructability or deconstructability to feature and could therefore be incorporated within them. Modular and prefabrication construction techniques were linked to constructability with some also referring to deconstruction. For this reason, they too were incorporated into the constructability section. Many authors focus on constructability over deconstruction in the literature reviewed, a reason for this is that the search results in Table 1 feature constructability far more heavily, to keep within the limitations of this study, thus maintaining an appropriate scope. The benefits of BIM to constructability were then reviewed using much of the literature previously mentioned. This was done to re-examine on the literature reviewed in order to support the second part of the research: How applicable is BIM to the identified constructability issues.

Figure 4: A visual representation of the reviewed literature and identified topic areas.

Many of the research methods used across the range of literature reviewed were not viable to base this research on due to a variety of factors. Much of the literature reviewed focused on specific areas of construction, for example Dossick, C. S. & Neff, G. (2009) focuses specifically on MEP in the implementation of BIM. In total there were nine examples of material specific literature identified similar to Dossick. Many other authors focused on specifics such as building types, areas within the project lifecycle and project team member roles (See Table 2). As a result of this, numerous constructability issues found in these articles couldn’t be applied as a general constructability principle to guide this study’s research method. Authors taking a broader research approach identified constructability issues that apply to all construction projects, for example Jergeas, G., & Put, J. V. D. (2001) used a survey approach and identified issues that reoccur in many other papers. This broader approach allowed better comparisons to be drawn between different authors, facilitating a wider understanding of the common constructability problems identified across projects. These problems range from construction methods, to trade coordination and to different types of building.

Table 2: Different focus groups of recorded constructability literature (Appendix C)

Frequency	Focus of study
8	Material specific (e.g. Steel design, Concrete formwork design, etc.)
9	Area of construction (e.g. MEP or Scaffold design, etc.)
2	Project Team Role (e.g. Construction manager)
5	Type of building (e.g. High-rise building, retrofitting, etc.)

2.2.         Constructability

Extracting constructability related literature from the one hundred and ten articles in the overall literature review led to sixteen articles being identified, the earliest dating back to 1987 by Clyde B. Tatum. This early paper focuses on constructability in the conceptual planning stages of a project. Tatum introduces three issues of constructability, developing the project plan, laying out the site and the construction methods. These issues are further explained within the paper however they are limited in their complexity due to their age, nonetheless, the overlying ideas behind the issues still remain in construction today. For example, planning works usually follow guidelines provided by a company template, especially on larger construction projects.

The specific issues mentioned by Tatum have since evolved into modern day equivalents as the building industry has evolved. For example, Tatum suggested having increased construction flexibility to maintain productivity when an unexpected restraint occurred. This was considered a constructability issue. Today many of these unexpected restraints have been removed (such as delays due to weather have been reduced through all-weather installation systems and insulated façade/cladding panels), which reduces the benefits of a flexible schedule. On top of this modern day construction is comprised of more specialist trades, this provides an increase in available works to be completed, as long as the critical path is not restricted. Tatum’s paper makes no mention of computer involvement in any aspect of construction, this vastly changes the problems faced and the possible solutions we can use to overcome them.

In 1997 Tatum published another paper focusing on the constructability of reinforced concrete (RC) structures [4]. This paper explores constructability issues across the entire construction process compared to just conceptual planning phases of the project. Despite focusing on RC, Tatum identifies many integral issues associated with constructability. This includes the requirement for construction knowledge in the design stage of the project to eliminate common problems that are unforeseeable to a designer with no practical experience. Tatum also comments on how some issues can have a general rules applied to them, for example “use lap splices wherever possible”, such rules can be modelled into a computer program, facilitating BIM integration. Tatum begins to introduce the basic ideas behind the use of computers to improve constructability, on a basic level. Inclusions such as companywide standards of constructability and statements such as: ‘designers will start a project from a higher level’ promote improved constructability standards. He also begins to introduce the ideas of storing constructability knowledge to be used on future projects (compared to relearning or relying on individual experiences). Overall, Tatum’s 1997 paper highlights many features of constructability that are still an issue in modern day construction, for example, improving communication between construction personnel and designers. Additionally, he also comments on how corporate constructability handbooks are overlooked in the heat of design development, a case in point being could be integrated into a computer model that prevents designers from ‘overlooking’ such handbooks.

As the literature reviewed moves closer to the present day, the use of computers becomes more and more apparent. The increase in computer usage and application of use created a new area for research methods. This literature review compares varying computational research approaches to tackle constructability issues in the field of BIM.

2.2.1.    A process approach

Following this rise in computer use a construction review process was developed in March 2000 for the transportation sector. Anderson, S. D. et al used Integrated DEFinition modelling (IDEF0) alongside a process approach to review constructability at different stages of the project lifecycle. An IDEF0 is a modelling methodology used by many governments and businesses to model actions, decisions and activities of a system (Varun Grover, William J. Kettinger (2000)). This technique was applied to a constructability system, and tested to investigate its effectiveness. By breaking down the construction process into its constituent parts it could be analysed, identifying where inputs, constraints and outputs lay within the process. Constructability issues were identified through the use of an expert team. As the transportation sector often has to work under constant operability circumstances a key factor for construction is to reduce time on site, for example, UK network rail replaced a bridge overnight when trains were not running to avoid affecting the services the following day (Network Rail 2017). Projects like these require immense planning and coordination to ensure success, otherwise significant costs are incurred. Constructability is a huge player in transport projects as unforeseen circumstances and projects delays have to be minimised and anticipated thoroughly.

Anderson (2000) identifies seven constructability criteria within each of the following phases, planning, design and construction from his survey results. These are divided into individual tasks in a companion paper (Fisher, D. J., Anderson, S. D., & Rahman, S. P. (2000)). This research method is very different to Tatum (1997), whilst the obvious difference is the computer usage, the actual approach does have some similarities. Firstly, the constructability areas are similar, breaking it down from planning, designing and construction. At this point the studies begin to differ, where Tatum (1997) identifies constructability issues such as ‘lack of coordination between designers and construction experts’, Anderson (2000) locates areas in the project cycle where issues should be resolved but lacks a definitive solution. Sub phases such as ‘identify major constructability issues’ relies heavily on expert knowledge to know and identify those issues. Anderson (2000) doesn’t raise the issue of designers needing to communicate better with construction experts to improve constructability. This lack of communication can be seen in both papers. Finally, despite the heavy use of computing technology and a process approach there is no mention of BIM as at the time it was still in the early stages of its development.

In 2012 Wang, L., & Leite, F explored the implementation of BIM at the design stage using a process approach. This is very similar to the for-mentioned Anderson (2000) paper in which there are many similarities despite being twelve years apart. Both Anderson (2000) and Wang (2012) use localised knowledge of project team members to identify information required for a constructability review process (CRP). A limitation of this is that using only a small number of people reduces the exposure to constructability issues across a range of projects. Nevertheless, the majority of constructability issues are likely to be picked up. The most common issues will be picked up due to their higher frequency. Further to this, such issues tend to be more significant because an accumulation of problems causes a greater overall effect.

Wang (2012) focuses on the design stage of the project, where change have the least impact on cost (See Figure 5). This focus allowed Wang (2012) to explore examples of constructability issues on her case study project such as differences in field and model tolerances. Addressing these issues in the design stage by understanding that model space is more refined than actual space led to the avoidance of plant collisions on the project. Anderson (2000) tested the CRP on two projects however there is no evidence of success of either models.

Many progressions have been made in BIM between the two authors papers above, with many sub categories listed in Anderson (2000) assumed to be included within the BIM implementation of Wang’s (2012) more recent CRP. This concludes the literature reviewed for a process approach addressing constructability issues on projects. Many of the ideas from these authors overlap to the more common case studies and survey approaches taken by the subsequent authors.

Figure 5: An adaptation of the RIBA stages of work to the ability to influence construction cost over time on a generic project. RIBA component source: https://www.ribaplanofwork.com.

2.2.2.    Case Study and Survey Research methods

The following authors used a range of research approaches to tackle constructability issues that encompass the use of BIM. What they all have in common is that they identify specific problems in more depth than process approaches that focus on a CRP derived from lessons learned. It looks at issues individually and begins categorising them, assessing potential mitigation techniques, either by a general rule, or other means. Because this approach identifies issues first, it shows a better evolution of the constructability problems faced, and how the industry has tried to solve them over time. Many then assess whether BIM can be a more effective approach to solve such issues, utilising constructability knowledge from expert’s industry-wise, to deliver a more efficient construction project.

Jergeas, G., & Put, J. V. D. used a survey method in 2001 to evaluate the effectiveness of potential constructability principles against realised principles. The survey sample used consisted of construction industry professionals in high level management. Their opinions are considered a reliable source of data as they represent different stakeholders across the industry. The survey indicated that the largest gap between potential and realised principles was in ‘advanced computer technology’, supporting Tatum’s (1997) paper. This paper highlights the benefits of using computer technology to improve construction across the industry, showing that in the fifteen years that separates these papers the industry did not maximise its potential to capitalise on available technology. Interestingly, the rest of the paper analyses the findings and concludes that a number of other principal groupings were more important to tackle. These focused largely on ‘a willing to change’, current practises and collaboration between firms to better deliver a project. Problems such as these still exist today, however with BIM’s potential uses it could be an aid to encourage change in the industry and reduce discrepancies between firms allowing them to better collaborate Jergeas (2001). One of the key points that Jergeas (2001) raises is that there are too many barriers to entry to maximise computer technology in construction, and hence there is a large gap between realised benefits and potential benefits. This gap refers back to his earlier point of ‘a willingness to change’ and barriers to entry being a key learning point of the study.

The theme of change was also picked up on by W. O’Brian et al (2012) through the benefits of 3-D and 4-D CAD models on a project. O’Brian (2012) focused on projects that have tried to change through the inclusion of such models and outlines the benefits achieved through their use. The key benefits O’Brian (2012) outlined were improvements to communication and visualisation, these reduced incompatibility between trades. Many of these points are highlighted by Jergeas (2001) as problems to address, while O’Brian (2012) commented on how 4-D CAD can improve a CRP through saving time and automating easy processes within them. O’Brian doesn’t comment on how the 3-D and 4-D models are integrated to the project team and how well they are supported by project team members. No drawbacks to using the models are outlined by O’Brian. Such drawbacks would include substantial costs from the initial software purchase and training of personnel to update drawings and maintain models (Bryde, D., et al (2013)). There is no indication of why this study was completed, or who it was funded by, this leaves the study open to potential bias.

Many authors use case studies for their research, and while the issues addressed are often project specific the overall constructability issues remain the same. For example, the 2012 study by W. O’Brian et al assessed the benefits of Three- and Four-dimensional computer aided design model applications on transportation projects. The study [105] identified constructability issues such as communication and visualisation that BIM can positively influence. These issues align with much of the literature previously mentioned.

Being focused on transportation involves more planning than for standard projects (Network Rail). This is a likely reason as to why more papers involving BIM (focusing often on planning stages) are focused on transportation. Many of the benefits and points made about the inclusion of BIM/CAD can be applied to general areas of construction despite the additional focus on transport.

A case study in 2013 by R Leicht looks into automating the CRP process through the use of BIM. Many benefits outlined by O’Brian (2012) apply to some of the issues mentioned by Leicht, for example ‘material of wall pier needs to be changed due to access restraint’ could be addressed more easily through the use of 3-D models to better visualise 3-D space. Leicht (2013) categorises his constructability issues differently to most other authors as they do not conform to the RIBA project stages. He categorises constructability issues by their fault, e.g. Design Omission. The majority of categories are independent of the RIBA stages, while ‘Unforeseen Conditions’ lies almost exclusively within the construction stage. This analysis included information fed from projects where multiple solutions are possible depending on the stage the project is in. It also does not identify when constructability issues are identified, which can have a large effect on their implications for the project (See Figure 3).

Leicht used one case study for his research and collated twenty constructability issues. This sample size was unlikely to encompass all generic constructability issues; however, the most common issues aligned with previous authors and so is deemed reasonable. This is the first paper reviewed to comment on a BIM implementation method which uses a rule based approach. Whilst this is simple it still eliminates many errors and can perform simple checks on the design to improve constructability.

His following paper published in 2014 aimed to address the problems mentioned by authors such as Jergeas (2001) and O’Brian (2012), bridging the gap between construction knowledge and design. This leads to the question: can BIM utilise constructability knowledge to guide designers and alert them to constructability issues only experts would have? Leicht (2013) tried to address this issue using simple functions, for example, an IF function to check the height of retaining walls meets a minimum value.

A common knock on effect of improving processes such as constructability is improved efficiency (Tatum 1997). When designing improved efficiency, sustainability often follows through a reduction in waste and time savings. S. H. A. Seoud (2013) investigated how constructability can be used to reduce project waste. During the investigation many constructability issues, such as the lack of involvement of construction personnel in the design stage were identified. Seoud puts this down to the type of building contract, recommending a design and build contract for maximum collaboration. Seoud (2013) also commented on how clients should be made more aware of the benefits of constructability can have on time and money. If clients were aware of this they are more likely to include it in the project budget, which feeds through to construction firms who in turn, can incorporate this into their budget. Similarly, to O’Brian (2012) the additional costs to implement good constructability processes were not mentioned. The proposed solution was not tested further to assess its effectiveness in improving constructability on a project.

2.2.3.    Prefabrication and Modularisation

Moving away from traditional building methods, prefabrication and modularisation offer new ideas about how to construct. Being a more recent method of construction and assembled in factory conditions, CAD is heavily used to achieve higher tolerances than traditional methods (Gunawardena, (2013)).

This was further explored in 2009 by J. Neelamkavil to examine the amount the construction industry can pre-fabricate to improve tolerances and buildability. This raised a new problem; that the benefits of using BIM vary hugely between trades. Significant costs to implement BIM properly are incurred as it fundamentally changes the way buildings are designed (Bryde, D., et al (2013)). Automation of certain building processes varies trade to trade as many trades such as carpenters work off hand-drawings. Therefore, some trade companies resisted change as skills, such as hand drawings, became redundant (Jergeas (2001)). Other changes such as the increased collaboration between trades in order to coordinate modular components of a building could also prove challenging (Neelamkavil (2009)).

In 2013 T Gunawardena developed a model for optimising prefabricated modular buildings. Interestingly he did not touch on the issues that many prior authors have, such as the coordination of trades and the challenges faced to ‘change’ the industry. Instead he focused on the likely problems faced. From the study it was not clear what kind of contractors carried out the construction works, making it hard to understand how contractors collaborated on the project, or if only a single contractor was used. Many of the ideas put forward by Gunawardena (2013) claim benefits such as reduced waste, reduced construction time, re-usability options, and better quality due to factory assembly conditions.

2.3.         BIM on Constructability

The benefits of BIM on constructability have been touched on by some previously mentioned authors, however this section investigated authors who focused on this topic in more detail. Ricardo Codinhoto et al (2011) aimed to investigate the BIM implementation process from an owner’s perspective and to devise guidelines for them to implement BIM in the Facilities Management (FM) of the building. Within the report, BIM deliverables are outlined, these include visualisation and decision making, 4-D construction programming and clash detection to name a few. These deliverables were organised by the RIBA Plan of Works (RIBA 2011). It could still be seen how BIM implementation varied across different stages of the project, with far more input in the design stages. The benefits of BIM to constructability were clearly outlined by Codinhoto et al (2011). The BIM deliverables were most effective in the design stages; resolving issues that had potential to become constructability related problems.

Interestingly, Codinhoto et al (2011) also discussed ‘mirages’ of BIM. These outlined some key limitations of BIM that the industry currently faces. One example was that the design process suffers from intrinsic problems that have to be managed. The use of BIM can help facilitate improved management, however the industry must adapt to utilise the benefits of BIM to improve the overall design process. Limitations such as these are interesting because they link back to constructability issues outlined by authors such as Tatum (1987), showing that some constructability issues have not been addressed in the twenty-four years between publications. Moreover, they can’t be tackled through the sole use of BIM, a change in how the industry operates must also occur to address such complex issues.

Focusing more on constructability, Wang, L., & Leite, F (2012) identified information required for a constructability review and then used IDEF0 models to implement BIM through the use of clash detection software. Weekly meetings were held to manage identified clashes and upload revised models to be checked the following week. Whilst this process required a significant amount of management time, it uses BIM to help identify constructability issues on the project. The main BIM benefits outlined by Wang & Leite (2012) are similar to those listed by Codinhoto (2011), with many 3-D modelling benefits mentioned. Benefits such as better visualisation, improved design coordination, and clash detection software helped bring potential constructability issues to the attention of the relevant management.

2.3.1.    3-D Modelling

The benefits of 3-D modelling are significant and diverse across the entire construction industry; however, this research focuses on the benefits it can provide to improving constructability problems. Firstly, 3-D models allow a better visualisation of space for complex areas, and also provide support for individuals who have less experience of reading 2-D drawings. A result of this is an increased opportunity for more members of the project team to spot mistakes. Additionally, it is far clearer to spot potential clashes, small spaces or undersigned spaces within the design. These benefits are gained just by having a 3-D model, with no additional benefits of BIM.

There are many additional benefits provided by BIM to address constructability issues. Features such as clash detection provide a degree of automation to identifying constructability issues and provoke design meetings to mitigate the identified problems. Other features are more passive and include improved design co-ordination through better integration of drawings and specifications. Features like this aid the reduction of design discrepancies and also aid the for mentioned benefit of visualisation without having extra information that improves the model from a simple 3-D CAD drawing to a BIM model.

Another feature of BIM 3-D modelling is the inclusion of the different models, these provide additional clarity and detail for the design team and facilitate further visual analysis from the design team to identify problems. These additional features can simply be categorised as part of improved visualisation, however the additional clarity and detail provided by a BIM model vastly improves upon a solo 3-D model.

2.3.2.    4-D modelling

The benefits of 4-D modelling include all of the features covered in 3-D modelling with some additional benefits. The fourth dimension is the incorporation of time. This allows a 3-D model to become a construction sequence. A key benefit of this is that the construction process can be frozen at any time. This provides a different view and provides even more opportunity for the design team to identify problems within it (O’Brian, (2013)). This can also be used to model logistical movements on site and optimise deliveries and site storage/ laydown areas.

2.4.         Conclusion

This literature review has looked at the identification and categorisation of constructability issues across a wide range of projects and how BIM has been implemented with the aim to address such issues. Constructability issues have existed in construction from its early beginnings. This literature review has identified the key constructability issues facing the construction industry, across a range of projects and construction methods. The key principles many authors commented on included a lack of construction knowledge in the design stage of the project, and poor design coordination between sub-contractors. Some authors classified constructability issues, the majority of which were located within the design stage of the project. These classifications help identify ways in which BIM can be implemented to address them (R. Leicht et al (2013)).

Two authors that focused on the implementation of BIM used this method, firstly their research identified and grouped constructability issues into logical categories. The research methods then divided. Wang (2012) used IDEF0 modelling to implement BIM to the constructability review process while Leicht (2013) assessed each constructability category individually. He ascertained how, if at all BIM could be used to improve the associated constructability problems. This approach identifies problems that have passed through the design stage unnoticed. Constructability issues should be dealt with in the design stage as they incur less cost, compared to in the construction phase, Leicht’s (2013) method identified common constructability issues that passed design stage and therefore incurred the largest cost for the project. Both Wang (2012) and Leicht’s (2013) studies had limitations to the implementations of BIM. Other authors such as Codinhoto (2011) outlined the limitations BIM still has to improve constructability without changing how we design and utilise tools such as BIM.

Other categories identified within the literature included Sustainability and Modularisation/ Prefabrication. Sustainability features throughout the publications as resultant benefit from improved constructability and BIM implementation. As time has progressed the use of computers has grown more and more in conjunction with the drive to be more and more sustainable in our designs. For this reason, the involvement of BIM has increased with time in the publications reviewed.

3.   Research Method

3.1.         The Case Study

The project used for this research was based in Headington, Oxford, and was called the Big Data Institute. The two-year project was a Design and Build contract. The building itself is a high-spec office building, built for the University of Oxford with a value approximately £35 million. The project included a 3-D model for the plantroom roof while the remainder of the building was detailed using 2-D drawings and other traditional methods. Three hundred and thirty-seven RFIs were raised on the project control collaboration software and were organised by the document controller at Mace.

3.2.         RFI Categorisation

The research method used in this study is partly derived from the literature reviewed above. The first part of this study used previously identified constructability categories to identify constructability related issues in RFIs. RFIs were chosen as the means to collect the data, as they are raised during the stages of the project closer to, and during the construction. From the reviewed literature, it was clear many constructability issues were identified in stages three four and five of the project, hence the use of RFIs as the data source.

The categories used were derived from Leicht (2013), who derived them from the Constructability Information Classification Scheme (Hanlon & Sanvido, 1995). Below, Table 3 shows the categories used as well as their definition. Their definition was used as part of the qualitative analysis in order to categorise the RFIs. Some of the definitions have been expanded upon for this study from Leicht’s work.

Table 3: Constructability Categories (R. Leicht 2013)

No.	Constructability Category	Definition
1.	Design Inadequacy	The existing design is inadequate to meet expected performance.
2.	Design Omission	The information of the existing design is incomplete or missing.
3.	Design Ambiguity	Information of the existing design is inconsistent across drawings, schedules, models, details etc.
4.	Design Coordination	Design concerns regarding coordination with other project participants and other trades.
5.	Unforeseen Conditions	Design concerns due to unforeseen external impacts from/ to the environment, the infrastructure and adjacent sites.
6.	Resource Constraints	Design concerns due to resource requirements or impacts, including material, time, equipment, tools, space, etc.
7.	Construction Performance	Design concerns based on construction performance, including cost production rate, quality, safety, etc.
8.	Not Constructability related	The RFI is not constructability related.

The seven constructability categories used in this research are shown above, with the eighth category identifying issues that are not constructability related within the RFIs. The categorisation of the RFIs was completed by reading their contents and assessing which category best suited them. Some RFIs were constructability related, however on further inspection many were not, an example of this would be an RFI requesting setting out of specific works. This is not a constructability issue, as the information required for the subcontractor had just not been released.

Many RFIs were difficult to categorise as they were not explicitly a constructability issue when raised, however if not addressed they could become a constructability issue. A reason for this is the stage of the project in which the RFI was raised. For example, in the Design Detail stage constructability issues are easily addressed and expected to be found, e.g. the drawings and specification do not align. In the construction stage the same issues have the potential to create more significant consequences, for example incurring a project delay while the drawing and specification are checked to work out which is correct. On top of this, the BDI was a design and build contract (D&B), meaning the RIBA (2013) stages overlap and design development continues as the project is under construction. As a result of this, understanding the significance of each RFI and its overall impact proved more difficult to identify.

To aid with the analysis an additional definition of Design Coordination was included and can be described as: a high-level concept of the planning, scheduling, representation, decision-making and control of product development with respect to time, tasks, resources and design aspects (Duffy, A. H. B. et al. (1993)).

To mitigate these additional difficulties, extra parameters were added to analyse the RFIs. These additional parameters included:

• RIBA 2013 classification
• Significance to project
• Company raised by
• Format

These additional parameters were included with the aim to aid the analysis of the data later in this study. The RFIs were categorised using the constructability criteria labelled one to eight and also with the four criteria above (see Tables 3, 4 and 5).

Table 4:  Definition of project significance for RFI classification

Significance	Description
1	A minor issue that has no structural implications and isn’t on the critical path of construction. If unresolved no major consequences occur.
2	A common issue, if not addressed it could cause small delays to the project. Could include issues such as Drawings and Specification discrepancies.
3	A major issue, if unresolved serious time and costs are incurred on the project. Likely to be on the critical path, or key to releasing further work.

Stages 0,1,2 were not included in RFIs as RFIs began when the project team had set up on site, this is usually at the near end of stage 2 of the RIBA PoW (2013). For this reason, the RFIs were likely to only focus on stages 3 4 and 5.

It was difficult to identify constructability issues during stages 3 and 4 because at this point you would expect to identify issues and therefore resolve them early. The problems arise when these are in phase 5 as the cost is higher for a lower influence value. A limitation of this is that because the job was a D&B contract multiple phases were occurring simultaneously, so each RFI must be classified into its RIBA stage.

Constructability issues identified in stage 4 (Technical design) were identified through errors in the design, the simple release of documents for design wasn’t considered a constructability issue, nor were instructions to “set out works” for different trades.

Table 5: The RIBA 2013 Plan of Works (PoW). Sourced from: www.ribaplanofwork.com

After the RFIs had been categorised into their relevant location they were analysed to identify trends in the data. The next part of Research looked at the benefits of BIM to constructability issues and how they could be categorised.

3.3.         BIM Applicability

From the literature reviewed there are many common ideas about the benefits of BIM to construction projects. Many of these benefits are difficult to quantify beyond a certain point as they offer passive support to designers such as visualisation (O’Brian. (2013)). This study identifies the key benefits that BIM can have on constructability through secondary research as shown in the literature above. From this literature, the BIM benefits can be split into 3-D modelling and 4-D modelling, with further classification below these levels.

This study used the previously defined constructability criteria (see Table 3) to assess the extent to which BIM can be used to improve constructability issues. This was completed through the use of a BIM Applicability Matrix (Figure X). The benefits of BIM were assessed against the seven different constructability criteria. This was a qualitative study and is therefore partial to a level of subjectivity.

Table 3 (Leicht 2013) shows the Constructability categories numbered 1-8, these correspond to the y-axis of the Matrix with the exclusion of the non-constructability category (category 8). Table 6 shows the benefits of BIM for constructability issues. The benefits have been numbered for reference in the Matrix.

Table 6: The Benefits of BIM to Constructability Problems

No.	BIM             Feature/ Function	Description
1.	Visualisation (3-D)	Visualisation is the broadest benefit of BIM as it is a result of both 3-D and 4-D modelling, enabling designers to see the model very clearly and better understand how it will look when constructed.
2.	Clash Detection (3-D)	A software feature that draws designers’ attention to intersecting building components.
3.	Design Coordination (3-D)	This benefit is closely linked to visualisation but has more emphasis on areas such as different sub-contractor work interfaces. It also focuses on the alignment/ consistency of different models and the linked information to the model (e.g. specification to model building component).
4.	Alternative Models (3-D)	This feature supports clearer modelling and divides structural and architectural details. It is very closely linked to visualisation, however it provides additional benefits through the linked, split simplified models.
5.	Sequencing of Works (4-D)	The process by which the project is constructed, this allows designers to view the project at different construction stages, facilitating better development of design.
6.	Logistical Planning (4-D)	Through 4-D modelling a site layout model can be used and linked to a delivery schedule to optimise site storage and onsite logistics/ lay down areas.

These categories were then aligned against the categories in Table 3 to assess the extent to which BIM can benefit constructability issues identified on the BDI project in Oxford.

4.   Findings

This section presents the findings of the research carried out on three hundred and thirty-seven RFIs on the BDI project, Headington, Oxford. As mentioned in the previous section the data was gathered from the project control collaboration software system that contained all the RFIs raised throughout the project lifecycle.

4.1.         Initial Breakdown

Of the three hundred and thirty-seven RFIs collated one hundred and forty three were not constructability related, leaving one hundred and ninety-three constructability related issues within the Constructability criteria adapted from Leicht 2013 in Table 3. Figure 6 shows the initial breakdown of these results. This simple classification did not take into account any other criteria of the RFIs.

Figure 6: Categorisation of all constructability RFIs analyse. Source: Appendix A

 

As can be seen in Figure 6 the constructability issues identified vary in frequency across the project, with Design Ambiguities and Design Coordination being the most frequent. Unforeseen conditions and Resource constraints are least frequent, with only six RFIs in each category. The remaining three categories were similar in division, with design inadequacies making up 19% of constructability related issues.

As previously discussed, the influence of change is highest in the early stages of design (see Figure 5), for this reason constructability issues have the highest cost to the project in the later stages of design, such as RIBA 2013 stage 5 – Construction.

4.2.         Constructability Categorisation

This section presents the results of the RFI classification with the additional consideration of their location within the RIBA PoW. The study identified thirty-seven RFIs in the construction stage of the project that related to constructability. This breakdown can be further refined through the use of the significance ratings given to each RFI. The research methodology explains significance in Table 4. It also explains that more significant problems are likely to be within the construction stages as their influence on cost is higher at this stage. The results show that of the thirty-seven RFIs within RIBA PoW stage 5, twenty-nine have a significance above 1. The figure below shows a graphical representation of the RFI breakdown into their respective categories, using significant RFIs only.

Figure 7: The breakdown of significant, constructability related RFIs within RIBA stage 5.

Within the RFIs recorded, twenty-nine were in the construction stage. These RFIs are presented in Figure 7 above. It is clear that Design Coordination caused the most constructability issues on the BDI project in the construction phase attributing 36% of all issues. This was followed by Design Ambiguities and Construction Performance. Design Ambiguities decreased by 6% with the refinement of significance to the results, however, at 22% it was still 10% lower than Design Coordination. Design Ambiguity issues decreased by 10% when only the construction stage of the project was considered. Construction Performance, by comparison, rose from 11% to 16%. The significance of these results will be discussed in section 5. Design Omissions and Inadequacies both contributed five RFIs to significant constructability issues. Unforeseen Conditions and Resource Constraints both remained at 3%, each contributing only one RFI to the results. Of the construction stage RFIs, fourteen were in text format and sixteen were in drawing format.

4.3.         RFI Origins

As part of the raw data, details such as which company raised the RFI were included. This allowed the RFIs to be categorised accordingly, highlighting features of the results such as which company raised the most RFIs. These were then segregated further into their respective RIBA stages.

Figure 8: RFIs organised by which company raised them. (Appendix A).

Figure 8 shows that over the life of the project, Mace (General Contractor) raised the most RFIs relating to constructability. RFIs raised by Mace were often secondary RFIs, whereby Mace acted as the intermediary between Sub-contractors, aiding to coordinate works through the raising of RFIs. This is one contributing factor to the higher number of RFIs raised.

From the figure it is clear that the Concrete sub-structure, MEP and Joinery contractor raised the majority of the remaining RFIs on the project. These were the three largest packages on the project so it makes sense that they were responsible for the majority of RFIs.

The smaller contributors to the RFIs realised include the Landscaping, Façade, Roofing, Drylining, Steel and Piling contractor. With the exception of the Drylining contractor, the results shown roughly correlate to the amount of work completed on the project (based on the value of each contract). Considering that Drylining works involved a vast amount of design co-ordination it was surprising that the number of RFIs raised was so low. This however could be due to a range of factors mentioned in section 5 of this study.

Figure 9 shows how these RFIs fit into the RIBA PoW construction stage of the project. As illustrated, the majority of constructability issues identified are by the Joinery Contractor. There are many possible reasons for this high value, however taking significance into consideration the value drops by 8% (Appendix A).

Figure 9: RFIs organised by which company raised them within RIBA stage 5. (Appendix A).

4.4.         Design Coordination

The findings have shown that 35% of constructability issues significant to the project and located in the construction phase, are Design Coordination issues. Table 7 displays three example RFIs from this category.

Table 7: Three example RFIs

No.	Contents	Raised by:	Format
11.	There is a reinforced concrete slab between gridlines A3 & C5 in the basement. It is 600mm deep and 1200mm wide. There are services to pass by it and I am looking for confirmation we can attempt the following. The ceiling drops 250mm below the slab and there seems to be space between the slab and ceiling to route services. If we route here will these services be visible between the slab and the dropped ceiling partition or is this ‘bulkhead’ going to be completely enclosed. Is it permissible to route in this space? I don’t see an alternative (other than drilling the slab). Screen shots attached	MEP contractor	Drawing
22.	We will not be able to put 120mm insulation and 75mm screed onto the slab in the B1 Level Good’s lift circulation area as drawing 839-BDI-4400. The SSL here is 91.985 as 30251-BDI-S-028, Level B1’s SSL is 91.455 as 30251-BDI-S-002 add 120mm insulation, 75mm screed and 350mm raised access and the level is 92.000. Will we need to insulate the slab’s soffit on Level B2 i.e. in the Labyrinth? If so would we need to insulate the walls too to prevent cold bridging? On a related item, do the Labyrinth’s air shafts need insulating?	Mace	Text
33.	Issue immediately the size setting out for your preferred openings in the covered plant room roof to facilitate the craneage of plant into the covered plant room. this proposal will need agreement with the steelwork && decking S/C, and as such may be subject to comment or amendment.	Mace	Text

Table 7 shows three typical RFIs that highlight different issues that cannot be mitigated against by BIM in the same way, despite being from the same category. This will be further explored in section 5.

4.5.         BIM Applicability Matrix

Below is the BIM Applicability Matrix (Figure 10), this compared the BIM benefits outlined in Table 6 against the constructability categories outlined in Table 3. Each comparison is given a rating of one to three. One indicates that BIM cannot be used to mitigate this constructability issue currently; 2 shows that BIM can be used to partly improve constructability issues and 3 suggests that this BIM feature can be used to mitigate almost all constructability related RFIs in this category, inclusive of the need for management and decision making to solve problems. (e.g. clash detection drawing attention to issues, but requiring meetings to go through potential problems).

		Constructability Criteria
		1	2	3	4	5	6	7	Total
BIM Benefit or Feature	1	3	3	2	2	1	2	1	1.9
	2	2	2	1	2	3	1	1	1.9
	3	2	2	2	2	1	2	2	2.2
	4	2	2	2	2	3	2	2	1.9
	5	1	2	2	2	1	1	1	1.4
	6	1	2	1	2	1	1	1	1.2
	Total	1.8	2.2	1.7	2.0	1.7	1.5	1.4

Figure 10: The BIM Applicability Matrix. (Appendix A).

The Matrix above was filled using data from RFIs gathered. Each RFI was assessed against each BIM function or feature. The resulting Matrix shows the average integer values given in each constructability category. Totals are in the end column and row, showing how applicable BIM was over all constructability categories, and across all BIM functions. These values are given to one decimal place to indicate which BIM functions are more applicable than others.

The BIM Applicability Matrix shows that the constructability categories that benefitted most from BIM are, Design Omissions and Design Coordination (see Table 3). It also shows that the most useful feature of BIM is Design Coordination (see Table 6). Design Coordination is a collective term used for the features of BIM that link the 3-D model to design specifications and ‘coordinate’ the BIM model so there are no discrepancies within it. Alternative Models, Visualisation and Clash Detection are other features of BIM that this study identified as beneficial to reduce constructability issues on the project. The remaining features; Sequencing and Logistical planning have a limited impact for the constructability issues identified on the BDI.

5.   Discussion

5.1.         Constructability vs Project Impacts

In this part of the study the results and analysis are discussed to identify the key findings and outcomes of the research. When classifying constructability related issues time is a key factor. For example, an RFI regarding a column’s close proximity to a lining wall is dependent on the time that this problem was identified. If the problem was identified in RIBA stage 2 or 3 or even 4 then it can be resolved with no or little additional cost or delay to the project. If it is identified late in stage 4 or in stage 5 then it will more likely create a delay or cost to the project. For this reason, identifying constructability issues early in the design process is key to reducing their overall impact on the project. This is supported by Tatum 1997 as he recognised that improving constructability knowledge within the design stage will help eliminate common problems that are unforeseeable to a designer with no practical experience. This reasoning formed the basis for the choice to use RFIs to identify constructability related problems on the project.

This study has used a different research approach to much of the literature preceding it. As previously discussed, identifying problems earlier in the design stage reduces their impact on the project. Many authors therefore focused their research on these areas to improve a designs ability to identify issues at these stages. This study identified the constructability related issues that have passed this point and therefore had a higher impact on the project. By identifying the types of issues passing through the design stage, it can be understood where the industry needs to focus on improving in order to reduce the impacts of such issues.

The constructability categories used in this study are based on R. Leicht’s (2013) paper. The reason for this was that, despite it being the only paper to classify issues in the way it did, it was independent of the stages of construction. The majority of other papers focused on the design stage of the project. This was because the majority of constructability issues were identified in this stage (Tatum (1997), Anderson (2001)). Using R Leicht’s (2013) method allowed this study to focus on the issues that are passing through the design stage, into the construction stage, where costs are higher (Figure 5). Categories were defined using a case study research approach. Other authors, such as Anderson (2001), break down the construction process and identify constraints, inputs and outputs of the process. These are then analysed to identify the constructability issue categories.

To summarise, this study and R Leicht’s (2013) used a case study approach to identify constructability problems. Anderson and others, such as Wang (2012) and Jergeas (2001), use a process approach and survey methods to identify constructability problems. These methods do not identify specific constructability related issues, but rely on industry expertise to identify common issues that apply to a range of construction projects.

The results show that of three hundred and thirty-seven RFIs analysed, one hundred and ninety-four are constructability related, of which a further one hundred and forty are located in the technical design stage. Figure 11 illustrates the split of RFIs between the respective RIBA stages. A likely reason for the higher number of technical design RFIs is that as the project team reviews developed design drawings and schedules, mistakes are identified. Further to this, Sub-contractors review concept and developed design drawings as part of their process to create their technical drawings. This helps identify constructability issues such as design omissions, inadequacies and ambiguities. Identifying issues at this stage of the project is beneficial as the incurred costs are very minimal. This study contrasts much of the preceding literature which focuses more on design stages of projects.

Figure 11: RFIs organised by their respective RIBA stage location. (Appendix A).

Thirty-seven constructability related RFIs were identified in the construction stage of the project, as seen in Figure 11. Of these, twenty-nine were considered significant and were therefore likely to cause delays and costs to the project. Figure 7 shows the breakdown of constructability categories within the construction stage, 36% of which are located within Design Coordination. A likely reason why this represented so many of the constructability issues is that it involves interfaces between differing trades. O’Brian 2012 commented on this as problems such as sequencing of works and poor construction quality arise. This has created a drive to reduce the discrepancies between trades and improve how well they interface. Following this, many of the problems identified in Design Coordination relate to details in the design. It is likely, as these details were small, that they were overlooked in the design stage of the project. Below is an example RFI from this section:

The information missing to avoide the above issue was checking floor level continuity between different drawings. As a result, the area was redesigned, which caused delays to the project. A possible solution to this specific example would be a BIM model that could flag up discrepancies in floor levels as the different drawing elements are added to the model. This way the issue is identified when the drawings are input to the BIM model, which should be in the design stage, where the cost of change is very low (Figure 5).

5.2.         BIM as a Solution

The analysis of the RFIs showed that while constructability categories can be used to classify and group different RFIs, it does not provide a general BIM solution to each of the constructability categories. Each RFI was assessed individually against the six BIM feature to ascertain how good BIM was to provide a solution. Whilst many RFIs could be solved with similar BIM features there was not a single valid category of constructability issue that could be solved by a singular BIM feature. To clarify, there was no clear BIM feature that could solve all RFIs within a specific constructability category.

Table 7 shows three RFIs that all lie within the Design Coordination category for constructability as well as all being in the RIBA Construction stage of the project. Despite all these similarities, each RFI is missing a different type of information to solve it. For example, number one is missing the information in the design dictating whether or not the bulkhead is enclosed. If this was given, the clear solution would present itself. RFI two refers to a design ambiguity whereby different finished floor levels are given. BIM could be used to eliminate this issue through design coordination and visualisation. To clarify, the BIM model could recognise that the floor levels inputted from different sub-contractors do not match, and then highlight the discrepancy to the user. Failing this, visualisation from a 3-D model offers an increased chance of a designer spotting the mistake compared to spotting it on two comparative 2-D drawings. Finally, RFI 3 refers to roof openings being maintained to allow access for MEP roof plant. This is a more complex constructability issue and therefore the information missing is more complex. The minimum size opening required is equal to the largest piece of plant to be landed, plus a safety margin. This information is within the designs but is outside the scope of this report to know whether or not it is within current BIM capabilities to recognise and identify. Other benefits of BIM still apply to solve this problem, using 4-D sequencing a clear craneage and laydown plan could be designed to maximise safety and optimise the quantity of lifts required, especially as large plant would prove difficult to move once landed. This supports the above paragraph as the three RFIs in question all belong to the design coordination category within the construction stage of the project, while the BIM features required to solve them vary from visualisation to 4-D sequencing.

O’Brian (2012) identified key constructability issues such as visualisation and communication which aligns with the results of this study as design coordination is the most heavily involved BIM function that links to communication. It also benefitted all constructability categories covered in this study except for unforeseen conditions (See Figure 10). A reason for this being that design coordination is a very general feature of BIM that applies to many areas of the project. It utilises simple features of BIM such as model consistencies, for example RFI 2 in Table 7 could have been avoided if a BIM model had been used. Different drawings would have been created from the BIM model and could not contain differing floor levels.

The second most beneficial feature of BIM this study identified was visualisation. This key feature of BIM could provide improved constructability on a project, it scored 1.9 as an average total across all constructability criteria in this study. The benefits of visualisation are some of the most basic in BIM. While it may not solve constructability problems it provides a huge amount of clarity to designers. This in turn creates more opportunities for design issues to be identified before construction. This relates back to O’Brian’s (2012) study. While he took an entirely different approach to identifying the benefits BIM can provide, the key results have a commonality with this study, identifying visualisation and design coordination as key benefits BIM can provide. This corroborates the results of this study.

From the BIM Applicability Matrix (Figure 10) the two functions that are least applicable to constructability are sequencing and logistical planning. Both of these functions are part of 4-D BIM, which includes the incorporation of time. A likely reason why these functions are least applicable is that they are quite specific. For example, logistical planning is limited to RFIs relating to site layout and logistics. As a result, the overall applicability is graded lower than other more ‘generally’ applicable BIM functions. For this reason, the results of the study must be looked at carefully as not to not regard BIM functions such as logistical planning as less effective than others, such as visualisation, which as discussed can be applied to a wide variety of constructability problems.

The two constructability categories that BIM functions applied to least were Resource Constraints and Construction Performance. These two categories have something in common, the problems that lie within them are as a result of unplanned results within the design and construction. Issues such as quality control and safety regulations can cause a need for redesign. A reason why BIM is less applicable to these two categories is that they follow the design that BIM is supporting, therefore it is very difficult for BIM to identify the issues as the work has often been completed in accordance with the design. An example RFI from the Joinery contractor highlights this is below:

“Due to the H131-490 Pivot and H131-106 emergency release stop requirement we cannot achieve the requested 34 Db audio rating. Note in order to try and achieve the 34Db rating we propose to supply the doorset to accommodate the GEZE TS 500 INVERS transom closer. As such the H131-490 will not be used. In lieu of a mechanical drop seal at the base of the door we will use 6mm projecting wipe fins (because of the bottom pivot), which will require a raised profile threshold plate (by others) in order to seal the doorset for audio. As this seal configuration is not tested with a transom closer we could no guarantee the audio performance. Please confirm your acceptance?”

By analysing this RFI to assess what information is missing it can be seen how challenging it is for BIM to be implemented to address the issue. The missing information in this example is a solution to fit the doorset while maintaining the 34Db rating and H131-490 Pivot and H131-106 emergency release stop requirements. This level of complexity is assumed to be hard to model. As such, there is less chance current BIM capabilities could aid in solving this RFI (Figure 1).

The two constructability categories that could benefit most from BIM are Design Omissions and Design Coordination. Starting with Design Omissions, there were twenty-three RFIs in the project, of these, only four were within the construction stage of the project. The BIM Applicability Matrix was created from RFIs in the construction stage, therefore only four RFIs were used to assess the BIM applicability level. This limitation is explored further within section 5.3. Despite this limitation the actual RFIs provide clear evidence that BIM could be of a huge benefit to reduce problems raised. Table 8 below contains the four RFIs in question.

Table 8: Design Omission RFIs in RIBA stage 5. (Appendix A).

No.	Contents	Raised by:	Format
1.	There is a wall opening shown on drawing BDI-S-002 starting near gridline intersection C-07, if this is the same as the other openings then the TOC level of 91.900m will be lower than the adjoining slab and there will be an opening into the void to be filled. Can you please specify what is to be constructed in this area (highlighted drawing attached)?	Superstructure Concrete Frame	Drawing
2.	Please provide locations for the cast in openings as per the attached.	Superstructure Concrete Frame	Drawing
3.	50.04.04 – We are unable to measure for the S/O. Can you clarify a frame thickness we should manufacture to so as to ensure there is no delay to the project?	General Joinery	Text
4.	Can you provide a final drawing showing the following…? – Dimensions / Set out of the 2nr S/S inserts to the Tread. – Dimensions / Set out of the 1nr S/S insert to the Riser.	Hard Floor Finishes	Drawing

RFI one could be solved through the use of Design Coordination and Visualisation BIM features. It may still be overlooked by designers until the RFI was raised if the empty space was modelled as such. RFIs two, three and four could all be solved through the use of Design Coordination and Visualisation. Queries that sub-contractors have, such as the above, could be avoided as they too would have access to the BIM models and can therefore see where ‘cast in openings’ are located. The model could also show the specification required to ascertain the ‘frame thickness’ to produce. Failing this, it is likely to be on the 3-D model for them to view as an alternative measure.

Many of the benefits BIM provides for Design Coordination have been covered under Design Omissions, however the common problems within this category differ slightly. Often in Design Coordination problems arise at trade to trade interfaces, this leads to sequencing problems amongst other issues such as tighter quality management to ensure the flow from one handover to the next remains smooth. The benefits BIM can provide for this are largely 4-D sequencing and modelling. This provides clarity to sub-contractors for their works and would help reduce trade to trade discrepancies. The problem with this idealistic view of BIM solving the above problems relates back to Jergeas’s study in 2001. He comments on the differing drive of sub-contractors to use BIM as part of their design process. These differences hinder the effectiveness of the BIM model to improve the construction process as some trades have more BIM capabilities than others. This uneven level of capability is one of the barriers to entry preventing BIM from being better incorporated into construction design (O’Brian (2012)).

In summary the use of BIM for the problems highlighted above would reduce the amount of time contractors spend checking what to do, and provide them with a more reliable information source. A limitation of this is that some individuals will still want to check the works they are doing due to a lack of experience, or to ensure they are faultless should something go wrong.

5.3.         Limitations of the Study

This study has identified the common constructability issues on a single project, the BDI in Oxford. This is a limitation in itself, however common constructability problems found are likely to be representative for the industry. Further to this, the quantity of three hundred and thirty-seven RFIs analysed improved the reliability of the study.

Another key limitation of the study was of the data collected. On a construction project there are many constructability issues raised and resolved that do not pass through the RFI system. Therefore, it was likely that some constructability issues were not picked up by this study.

RFIs are raised by contractors, sometimes falsely, some of these RFIs have been excluded from the study through the Non-constructability related RFI category. However, without being engaged in the project it was hard to distinguish between erroneous RFIs and genuine constructability issues, this further reduced the reliability of the study.

The analysis of this study was predominantly a qualitative one, as such obtaining the results have an inherent level of subjectivity. This was another limitation of the study; as if this study was repeated the results could vary slightly.

Some RFIs identified as constructability issues in RIBA stage 4 had potential to be constructability issues, however if addressed immediately the likelihood of them having a cost or time related impact on the project was negligible. For example, the misalignment of specifications and drawings was a common issue that if left would cause constructability issues as an ambiguity. Many of these issues were picked up in stage 4 and could also be classed as design development, this was a limitation of the study as the difference between a design development and a constructability issue was not defined.

6.   Concluding Remarks and Recommendations

This study analysed three hundred and thirty-seven RFIs qualitatively into different classifications, the foremost being constructability categories derived from previous literature by R Leicht et al (2013). The non-constructability related RFIs were excluded from many of the analyses. One hundred and ninety-four RFIs were classed as constructability related and were further defined by seven constructability categories shown in Table 3. Design Ambiguities and Design Coordination RFIs were the two most frequently recorded. Other classifications used to analyse the RFIs included: their RIBA stage (as shown in Table 5), company raised by, project significance and format. From the reviewed literature and Figure 5 this study concluded that constructability issues identified in the later stages of the project have a higher impact on project delays and increased costs. For this reason, additional focus was put upon the RFIs situated within RIBA stage 5 – Construction. The results showed that 35% of issues were in the Design Coordination category, with the next best being 17% (shared between Construction Performance and Design Ambiguities). This indicates that to reduce time and cost impacts of constructability on projects, reducing Design Coordination issues should be a focus.

The second part of the analysis investigated the applicability of BIM functions and features to the constructability issues identified. This was carried out through the use of a literature review on the benefits of BIM on construction projects to identify what functions and features BIM currently offers within the industry. These features can be seen in Table 6. Again, using a qualitative analysis approach, each RFI was graded 1 to 3 on how well each feature or function of BIM could be used to mitigate it.

Grade 1 indicated BIM offering minimal or no benefits to the constructability problem.

Grade 2 suggested that BIM could reduce the given problem, or highlight it to designers, however this does not prevent the problem from being overlooked.

Grade 3 dictated that BIM can be used to address the problem earlier in the design stage, or provide much improved clarity for designers to solve it.

From this BIM applicability analysis, a final quantitative analysis was undertaken to match constructability categories against their average BIM applicability. All RFIs were filtered into their respective constructability categories, then the mean value of each BIM feature for that constructability category was placed in the BIM Applicability Matrix (Figure 10).

There are many options to further this study through additional research. One such way would be to analyse more projects through RFIs. To get a maximum effectiveness the research should be conducted while the construction takes place, adding extra emphasis on the RFIs to support a deeper, more accurate analysis. Interviews could be undertaken with project team members to try and identify additional RFIs or constructability issues that have not gone through the system to enhance the study.

Further research into constructability categories could be undertaken to try and identify a more refined definition of constructability issues. This could be designed in collaboration with BIM functions in an attempt to apply a more general rule to solve problems in a single constructability category. To do this effectively more research into the functionality of BIM would need to be undertaken. Primary research into BIM functionality was outside of this study’s scope, however researching the uses of BIM on a primary level would give far more detailed BIM feature and function categories. It would also grant a better understanding to the researcher classifying each RFI a score of one, two or three. These advancements would improve the reliability and accuracy of the study; they would also provide a better base to take this research to the next stage.

The next stage of this research would be to use the results of the improved study to investigate how BIM could be more applicable to constructability issues, focusing on issues with highest cost to the project. These issues would be a combination of frequency and significance. Further research should investigate how current functions of BIM that benefit constructability issues could be improved. In an ideal world, this research would be undertaken by the BIM providers such as Autodesk and Nemetschek, as BIM applications can be derived from constructability issues.

7.   References

Anderson, S. D., Fisher, D. J., & Rahman, S. P. (2000). Integrating constructability into project development: a process approach. Journal of Construction Engineering and Management, 126(2), 81-88.

Apacible, K. B. B. (2016). Constructability Assessment of Different Formwork Methodologies Using Building Information Modeling (Doctoral dissertation, De La Salle University).

Bryde, D., Broquetas, M., & Volm, J. M. (2013). The project benefits of building information modelling (BIM). International journal of project management, 31(7), 971-980.

CII (1993). Constructability Implementation Guide. CII Special Pub. 34-1, Univ. of Texas, Austin, Texas.

Digital Build Britain, HM Government. (2015) Level 3 Building Information Modelling – Strategic Plan. Published by Digital Build Britain. Available from: http://digital-built-britain.com/DigitalBuiltBritainLevel3BuildingInformationModellingStrategicPlan.pdf

Dossick, C. S., & Neff, G. (2009). Organizational divisions in BIM-enabled commercial construction. Journal of construction engineering and management, 136(4), 459-467.

Duffy, A. H. B., Andreasen, M. M., MacCallum, K. J., & Reijers, L. N. (1993). Design coordination for concurrent engineering. Journal of Engineering Design, 4(4), 251-265.

Edwards, J. F. (2003). Building the great pyramid: Probable construction methods employed at Giza. Technology and Culture, 44(2), 340-354.

Engelbart, D. C. (1962). Augmenting Human Intellect: A Conceptual Framework. Summary Report AFOSR-3223 under Contract AF 49 (638)-1024, SRI Project 3578 for Air Force Office of Scientific Research. Stanford Research Institute. Retrieved March, 1, 2007.

Fischer, M., & Tatum, C. B. (1997). Characteristics of design-relevant constructability knowledge. Journal of Construction Engineering and Management, 123(3), 253-260.

Fisher, D. J., Anderson, S. D., & Rahman, S. P. (2000). Integrating constructability tools into constructability review process. Journal of Construction Engineering and Management, 126(2), 89-96.

Gambatese, J. A., Dunston, P. S., & Pocock, J. B. (2007). The Way Forward: Recommendations for Future Constructability Research and Practice. In Constructability Concepts and Practice (pp. 142-145). ASCE Publications.

Gunawardena, T., Ngo, T., Mendis, P., Aye, L., Crawford, R., & Alfano, J. (2013). A holistic model for designing and optimising sustainable prefabricated modular buildings.

Hanlon, E. J., & Sanvido, V. E. (1995). Constructability information classification scheme. Journal of construction engineering and management, 121(4), 337-345.

Health and Safety Executive, HSE. 2008. Working on roofs. Published by the Health and Safety Executive. Available from: www.hse.gov.uk/pubns/indg284.pdf . [First accessed 20/03/2017]

Jergeas, G., & Put, J. V. D. (2001). Benefits of constructability on construction projects. Journal of Construction Engineering and Management, 127(4), 281-290.

Jiang, L., & Leicht, R. M. (2014). Automated rule-based constructability checking: Case study of formwork. Journal of Management in Engineering, 31(1), A4014004.

Jiang, L., Leicht, R. M., & Kremer, G. E. O. (2014). Eliciting Constructability Knowledge for BIM-enabled Automated, Rule-based Constructability Review: A Case Study of Formwork. In CONSTRUCTION RESEARCH CONGRESS, Atlanta.

Jiang, L., Solnosky, R. L., & Leicht, R. M. (2013). Virtual prototyping for constructability review. In Proc., CSCE 2013 Conf., Canadian Society for Civil Engineering, Montreal, QC Canada.

Lee, C., Ham, S., & Lee, G. (2009). The development of automatic module for formwork layout using the BIM. system, 7(3), 1-6.

National Building Specification, NBS. 2016. What is Building Information Modelling (BIM)? Published by NBS. Available from: https://www.thenbs.com/knowledge/what-is-building-information-modelling-bim [First accessed 18/11/2016]

Neelamkavil, J. (2009, June). Automation in the prefab and modular construction industry. In 26th Symposium on Construction Robotics ISARC.

Network Rail. (2017). Planned Works. Published by Network Rail. Available from: https://www.networkrail.co.uk/running-the-railway/looking-after-the-railway/planned-works/ [First accessed 10/03/17]

O’Brien, W., Gau, P., Schmeits, C., Goyat, J., & Khwaja, N. (2012). Benefits of three-and four-dimensional computer-aided design model applications for review of constructability. Transportation Research Record: Journal of the Transportation Research Board, (2268), 18-25.

Odeh, A.M. (1992). CIPROS: Knowledge-based Construction Integrated Project and Process Planning Simulation System. Ph.D. Diss., Dept. of Civil & Envir. Engrg., Univ. of Michigan, Ann Arbor, MI.

Seoud, S. H. A. (2013). Towards achieving sustainability in construction: constructability as a tool for reducing project waste (doctoral dissertation, architectural engineering).

Spitler, L. E. (2014). The Effect of Inter-Team Dynamics on the Constructability of the BIM Model. In 22nd Annual Conference of the International Group for Lean Construction (pp. 957-968).

Tatum, C. B. (1987). Improving constructibility during conceptual planning. Journal of Construction Engineering and Management, 113(2), 191-207.

Varun Grover, William J. Kettinger (2000). Process Think: Winning Perspectives for Business Change in the Information Age. p.168.

Wang, L., & Leite, F. (2012). Toward process-aware building information modeling for dynamic design and management of construction processes. In Proceedings of the 19th Annual Workshop of the European Group for Intelligent Computing in Engineering (EG-ICE). Herrsching, Germany: Technische Universität München.

8.   Appendices

All appendices are included on the accompanying USB stick with this report.

Appendix A contains all data for the Analysis, Figures, Tables and BIM Applicability Matrix that is not mentioned below.

Appendix B contains the data for Figure 2 and 3.

Appendix C contains the systematic literature review and data for Table 2