There are large amounts of data and information available on above-ground buildings, and infrastructures. For example, it is possible to do visual inspections and attach wireless sensors to monitor such structures and environmental conditions (Rodenas-Herráiz et al., 2016, Stajano et al., 2010). However, there is a lack of information and data on subsurface environments, such as on existing buried infrastructure, including pipes (type, location, and condition), ground water, and geology (Kessler et al., 2008, Marache et al., 2009, van Staveren, 2009). This lack of information can have a severe impact on future planning within urban underground space and on construction activities to maintain, repair, upgrade, and install new buried infrastructure. Moreover, the nature of the subsurface environment is often chaotic. Furthermore, the routes and the condition of buried infrastructure are normally not visible. These facts can be attributed to the historic lack of planning and regulation of subsurface space usage (Hou et al., 2016, Marache et al., 2009, Royse et al., 2008).
Trenchless operations, such a pipe jacking, microtunnelling, and horizontal directional drilling (HDD), offer many advantages over traditional open cut (or trenched) excavations. Advantages include reduced traffic congestion, less impact on the local community, and reduced damage to the infrastructures in the vicinity of the work. However, risks can increase due to the lack of reliable information on underground characteristics, such as the location of existing buried services or ground conditions (De Rienzo et al., 2008, Dong et al., 2015, olde Scholtenhuis et al., 2016). Access to 3D models containing data on the ground and its contents would help to minimise these risks Providakis et al. (2019). In this context, the quality of data and the manner in which it can be visualised and utilised are crucial aspects of such models.
This study aims to demonstrate that information on buried infrastructure and ground conditions combined with above ground information provide a useful tool. This tool can aid projects involving trenchless construction or maintenance operations. It can also reduce risks (to potentially less than that of open cut operations) and improve the sustainability and resilience of both buried and surface transport infrastructures. To achieve these goals, a Building Information Model (BIM) for underground applications has been proposed.
This paper briefly describes BIM and the current mapping procedure of buried infrastructure, which presents greater confidence due to marked improvements over recent years. The formation of the basis for an underground BIM based on the obtained data and ground conditions is also explained. The obtained model can be combined with risk assessments or similar analysis methods for a more dynamic operation. These ideas are then applied to a trenchless construction operation that used microtunnelling.
Building information modelling
The creation of a BIM includes generating, storing, and managing digital information for all objects and elements of a building or structure throughout its life (Eastman, Teicholz, Sacks, & Liston, 2011). Figure 1 shows the use of BIM throughout the life of a constructed artefact, from its conception, through the design and construction process, operation and maintenance, and even demolition. As part of a BIM creation process, all elements and objects are stored and located in the model to be viewed at any scale. For example, the model can store data on the structure and ground properties, on pipelines located in this space, and pipe material characteristics assigned to different elements within the system.
Buried infrastructure mapping
When work is performed in the streets, for example to repair existing or install new infrastructure such as pipes or cables, there is a risk of damaging existing buried services. Therefore, to minimise these risks, it is of paramount importance to have good quality data on the location of existing buried infrastructure, such as water and gas pipes, electricity cables, and telecom cables. However, because the road surface and ground are opaque and of uncertain composition, the identification of new buried infrastructure require geophysical techniques combined with visual observation of surface identifiers (e.g. valve locations, manholes) and ground truthing (e.g. vacuum excavation).
The use of ASCE38-02 (USA, ASCE, 2002) or PAS128 (UK, BSI, 2014) approaches to locate buried infrastructure can significantly improve the quality of mapping surveys and hence reduce subsequent construction risks (Armstrong, 2014). These approaches combined with better training and excavation practices, improved geophysical techniques such as ground penetrating radar technology, and data analysis techniques can increase the safety of streetworks. However, there is still a level of uncertainty regarding underground structures, particularly with respect to the depth of buried infrastructure (Metje et al., 2007, Metje et al., 2015). This uncertainty needs to be recognised in any streetwork activity, and information on the infrastructure locations in a 3D map can aid the planning and activity of on-site works. There have been initiatives to develop 3D buried infrastructure visualisation systems (augmented reality) for on-site activities (KPMG, 2018). Nevertheless, there is still a need for an integration to a larger model that incorporates all the data on the subsurface, including road structure and ground, to which the buried infrastructure is strongly interdependent (Rogers et al., 2012, Rogers et al., 2017). The model should also be continuously updated as more data become available.
BIM for the underground
Many new buildings these days have an as-built model detailing all their structural and construction information. However, most BIMs do not contain any information on the subsurface ground conditions or buried infrastructure in the vicinity of the building. Therefore, a BIM for underground applications (called in this study ‘BIM for the Underground’) that contains information on both above and below ground infrastructure would allow for enhanced planning and engineering risk analyses. The proposed modelling environment would overcome the lack of detailed 3D representations offered by BIM and issues associated with tools such as GIS, which are mainly used for large-scale spatial applications (Amirebrahimi, Rajabifard, Mendis, & Ngo, 2016). The proposed model would also complement specific advanced modelling environments that are currently being developed, such as the Tunnelling Information Model (TIM) at the Ruhr University (Meschke, 2018).
The buried infrastructure information obtained from mapping surveys can be added to the building models to create a 3D overview of all surface and subsurface physical infrastructure. However, to create a more complete BIM that includes underground information, geological and ground conditions must also be added to the model.
This geological information can be obtained from ground investigations of the site and its locality. Borehole information can be converted into a 3D geological model. However, 3D geological models are an interpretation of the discrete location information, and experienced geologists and geotechnical engineers are often required to produce the final model (i.e. their experience, opinion, and judgement are inherently embedded to the information). Hence, it is important to understand such characteristic of the model and not perceive it as an absolute truth. This judgement and experience captured in geological ground models or BIMs into which they are integrated would be extremely useful if the models are to be used by future engineers. Nevertheless, this integration is not currently performed.
Once the information has been incorporated into the BIM for the Underground system, the data can potentially be used more dynamically, i.e. data can be exported to other packages or bespoke software (such as SketchUp software, Inc, 2016) to run specific analyses. The enhanced data (or interpreted information) can then be returned to the BIM environment (again via the SketchUp software) for visualisation and future use (Providakis, Rogers, & Chapman, 2019). Figure 2 shows the export and import of information via the SketchUp software. The visualisation of the information that would be held in the BIM framework is also shown, and it includes surface structures, subsurface geology, and buried infrastructure.
The proposed methodology provides a new approach for data storage and visualisation of geotechnical features and other information using BIM. The obtained model can be combined with other information such as from a settlement risk assessment (as demonstrated in Section 5). As a result, the user has a comprehensive and informative tool that can be used for early assessments and to support decision-making. Further details of this method can be found in Providakis et al. (2019). Thus, only brief additional comments are provided here to illustrate the capabilities and potential of the model.
There are promising options to manage underground applications with BIM, including using multi-source geotechnical data combined with construction data within BIM (Zhang et al., 2018). Similar geotechnical data storage and modelling applications have also been explored by Meschke (2018). However, as an alternative approach, the methodology developed in this study utilises a comprehensive visualisation tool (SketchUp) that can act as an interface between the BIM database and analysis tools. Underground information (e.g. geological strata or pipes) is stored using 3D solid objects or elements, as introduced by Providakis et al. (2019). Any topological or qualitative attributes of these 3D underground elements can be assigned and adapted as element properties, i.e. in a classic finite element fashion. Subsequently, they can be converted using bespoke coding in MATLAB (Inc, 2016) and SketchUp to tabulate the information as readable to the BIM environment.
SketchUp (Inc, 2016) has been used in the presented examples because it provides easy-to-use multidimensional (BIM) visualisations and can convert any generated model to the IFC format (buildingSMART, 2017). Therefore, the information can be easily converted, edited, and used by any commercial BIM software. The information stored in BIM can be extracted via SketchUp or a similar software, and it can be made readable from platforms such as MATLAB to be used for analyses.
Currently, 3D solid elements used for analyses are based on common meshing procedures (Mei, 2014, O’Rourke, 1998). However, depending on the information to store or the analysis being performed, it might be more efficient to use more advanced meshing techniques such as a high order mesh for complex geological structures (Mei, 2014). Thus, different elements can be easily included depending on the analysis requirements.
There is a need for the ability to capture changes or potential changes within the underground environment (due to an anticipated action on the underground system), such as the installation of a new pipeline (via HDD) or suspected corrosion in an existing pipe due to a change in ground conditions. For that, existing information is transferred from BIM using SketchUp into either a bespoke code in MATLAB or a finite element package for an appropriate analysis. The modified information at the end of the analysis can be transferred back into BIM as a refined model. This updated information could be stored under the proposed methodology in numerous ways. For example, it could be stored as additional attributes to existing elements. Alternatively, as an original element, such as the section of a pipeline, it could be subdivided (as an output, for example, from the MATLAB analysis) to store changes along its length. It could also be stored as a new layer of information such that a history of change can be constructed (Providakis et al., 2019). The resulting layer from the MATLAB analysis provides all the information that is finally stored in BIM (Providakis et al., 2019). In addition, volume elements of soil could be handled in the same way using the proposed methodology, allowing differential changes in properties to be captured.
Uncertainty is often associated with underground environments. For instance, there is uncertainty in geological interpretation and parameters, location and positional coordinates, and physical parameters of buried infrastructures. These can all be captured within the proposed method as additional attributes associated with a particular element within BIM or as a new layer within the BIM model. For example, a section of buried pipe could have an additional attribute of positional uncertainty associated with it, i.e. ±50 mm horizontally and ±100 mm vertically. This information can then be utilised as required in subsequent MATLAB analyses or as a parametric input into a finite element analysis. Geological elements within BIM could similarly have uncertainty associated with parameters by including this uncertainty as additional attributes. If there are different geological interpretations of boreholes information, these alternative geological models can be added as individual layers within BIM. As a result, engineering judgements are more easily made when this information is required by subsequent projects or extensions to an existing project.
Based on the analyses, SketchUp visualisations can exemplify the generation of new layers. A preliminary analysis on settlement associated with a new sewer construction is presented in Section 5. In this example, a risk analysis is conducted based on ground surface settlements. The distance from the tunnel centreline was essential for this analysis. The subsequent risk assessment was performed on the ground surface by projecting the information and attributes of the rest of the subsurface model onto the ground surface, i.e. as a separate layer of information (Providakis et al., 2019). The proposed method can also be used to handle information affecting the soil volume, including strains within the soil, i.e. by conducting a finite element analysis. These results can also be handled and visualised within SketchUp and, subsequently, stored back into the BIM database.
Section 5 presents examples of the proposed methodology applied in general terms to trenchless construction operations. In these examples, only SketchUp visualisations based on the original information (that would have come from the BIM database) and subsequent results from the MATLAB analyses are shown. Therefore, they exclude the aspects of the methodology concerning the conversion of BIM data to the SketchUp visualisation and, subsequently, to the MATLAB analysis, and from the SketchUp visualisation to BIM data.
Application to trenchless construction operations
Trenchless operations for new constructions need to minimise the risk of damage to existing infrastructure, particularly to the surface road infrastructure, pipes, and cables in the immediate vicinity. This section examines how the concept of dynamic BIM for the Underground can be applied to a new 1.0-m diameter sewer constructed using a microtunnelling (pipe jacking) operation. Figure 3(a) shows the proposed site, and Fig. 3(b) shows the existing buried infrastructure from BIM in 3D. The model includes all BIMs of neighbouring buildings and subsurface geological information.
The 3D subsurface information incorporated in the model consists of typical superficial (i.e. soil and made ground) and bedrock layers. The BIM-building files were co-located with the subsurface pipe information. The combined model in Fig. 3 shows a detailed example including typical geometries of features for an urban utility network. This model was generated in the SketchUp visualisation environment using information from BIM.
Considering the proposed sewer construction project, the model allows for:
(1) investigation of the alignment for the new sewer to minimise disruption of existing services, and identification of services that need to be relocated. Figure 4 shows a possible location for the sewer within the BIM environment. This information can be viewed from any angle and distance. Thus, it is possible to view specific levels of detail within the model;
(2) an understanding of the ground conditions at the jacking and reception pits (which may ultimately become manholes) associated with the microtunnelling operation, and along the sewer line. The model can also be used to identify the need for further on-site investigations if the information is lacking for any reason (design construction, operational lifetime, etc.). This additional information can be used to update the model, and hence it can be used for subsequent projects in the vicinity;
(3) information from the model to be exported to provide the basis for risk analyses and assessments. For example, information on ground displacements caused by the sewer construction on adjacent services. The results of such analyses can be stored and presented within the BIM environment;
(4) the compilation of a repository for underground information and for each utility, including geometry, material, and other characteristics. This repository could be shared and reviewed by engineers and other professionals involved in the project. The information would be useful for many aspects of local planning and future long-term urban designs.
When conducting microtunnelling operations, the surrounding ground is disturbed, and ground settlements may occur during the construction process, both around the jacking and reception pits and above the tunnelling operation as the pipeline is created. To demonstrate the interaction of information between BIM and an external analysis package, in this case a bespoke MATLAB (Inc, 2016) programme, a settlement risk assessment was conducted above the line of the proposed sewer. This assessment was based on an analysis for surface and subsurface ground displacements proposed by Loganathan and Poulos (1998). The ground displacements were converted into levels of risk based on tensile strains, the results are shown in Fig. 5.
A coarse risk analysis grid was chosen in this example for demonstration purposes. However, the grid could easily be refined as needed. The narrow location directly above the line of the sewer was directly affected by the tunnelling activities and indicated a moderate risk of damage to the road surface. Nevertheless, the subsurface infrastructure would likely be more affected by these ground displacements. Thus, the possible displacements can be combined with details of existing pipes and cables to produce a risk analysis for these assets. Such analysis can be conducted as the construction of the sewer progresses, an example of the output of such an analysis is shown in Fig. 6. The risks are based on levels of tensile strain induced to the various utilities caused by the relative displacements (estimated in the previous ground settlement analysis) created as the construction progresses. This analysis allows potential issues, particularly at pipe joints and connections, to be highlighted and remedial measures to be taken prior to construction.
During the construction phase of the project, the monitoring information collected, e.g. the surface settlement along the road, can be compared to the pre-construction estimations, and appropriate risk mitigations can be adopted if any issues are identified. The key feature is that all information is available in one model, thus being easily accessible and shareable.
After the construction is finished, the as-built information of the project can be incorporated into the BIM and hence be available for future projects (Fig. 7). At the sewer site, for example, the installation of a new electricity cable was proposed along the same road after the sewer construction using horizontal directional drilling (HDD). Figure 8(a) shows the proposed route of the new electricity cable, and Fig. 8(b) shows the existing buried infrastructure, including the new sewer construction. Thus, the new cable route can be determined using the most up-to-date information, which demonstrates that BIM for underground applications can aid planning and decision-making. Figure 9 shows the proposed route for the new cable to ensure that it misses the existing infrastructure and minimises risk. The ability to consider all available information in one model, in which this information can be easily shared and visualised in 3D, is a powerful tool for planners and engineers.
The studied example shows that BIM for the Underground can do more than store data. It can be used, for example, to visualise data for planning and design purposes, utilise data more dynamically to show the results of risk analyses, and for sustainability analyses (Hojjati, Jefferson, Metje, & Rogers, 2017). In addition, it can become a continuously evolving model as new constructions or changes occur and the information is incorporated into the model. The availability of data in one easily accessible environment can help with decision-making processes regarding streetworks and disseminate the use of trenchless construction and maintenance operations, the economic, social, and environmental benefits of which are often significant (Hojjati et al., 2017).
BIMs are well established for above ground structures. However, the use of such approach for subsurface infrastructures is highly limited. The subsurface environment, particularly in urban areas, is an important asset that can aid the planning and design within this space. The concept of BIM for the Underground that encompasses both above ground and subsurface structures as well as 3D geology and ground conditions is advocated herein.
Due to the complex nature of subsurface environments, i.e. the inability to readily see through the ground and the consequent lack of knowledge on the position and condition of buried infrastructures and ground conditions, it is challenging to create a BIM for underground applications. One of these challenges is the determination of the accurate location of existing buried infrastructure, such as pipes and cables, to enable a 3D map of the infrastructure. The paper discusses the developing approaches and techniques that can achieve this, although there remain uncertainties that need to be appreciated. However, once this information is obtained, it can be combined with ground conditions to create a framework of BIM for underground applications.
This study demonstrates the development of a BIM for the Underground, and, in particular, how to access and utilise the obtained information to perform analyses (e.g. risk analyses). The proposed model can aid the planning, construction, and future operational activities (e.g. maintenance). The example of a microtunnelling (pipe jacking) operation for the installation of a new sewer was investigated.
The key messages from this paper are:
(1) The concept of a BIM for the Underground offers a powerful tool to aid streetworks activities, including trenchless construction and maintenance operations. However, the data needs to be accurate and the uncertainties acknowledged.
(2) The proposed model can be combined with analysis tools, such a risk assessment. It then becomes progressively more influential for urban decision-making, allowing different scenarios to be investigated expediently.
(3) BIM for the Underground can also be used dynamically during construction. Monitoring data are collected, incorporated to the model, visualised in real time, and compared to pre-construction assessments. Therefore, deviations from expected behaviour can be identified, and actions can be taken as needed.
(4) As-built information can be immediately input to the BIM for the Underground and used for future projects and maintenance planning.
(5) This BIM development could be scaled up into a comprehensive programme of urban subsurface transformation: Engineering the Underworld. Measures to support this transformation are being developed. This builds on various initiatives to detect, locate, identify and map the buried infrastructure (Mapping the Underworld; Royal et al., 2011) and determine the condition of this existing infrastructure alongside that of the overlying transport (typically road) infrastructure (Assessing The Underworld; Rogers et al., 2017).
Source: BIM for the Underground – An enabler of trenchless construction
Authors: David Chapman, Stylianos Providakis, Christopher Rogers