It is now 49 years since Eastman theorised what would become known as BIM: Building Information Modelling. Despite this, we can observe that the methodology, together with its associated tools, is still considered an exception to established practice, an eternal novelty with clearly something unfinished. If we exclude a few excellences, such as the United States and the United Kingdom, and countries vying to keep up, such as France and Italy, there are still a lot of regions where BIM is completely unsystematised. As a result, it is first implemented in large design studios and public projects and only then, with difficulty, does it spread to the rest of the market. While we could accept the idea of a silent revolution that takes time to gain a foothold, it is now clear that too much pressure has been applied to the AEC (Architecture Engineering Construction) sector, which was not ready for such a radical change, first in thought and then in practice. Putting this aside, the time required for innovation is in any case not compatible with that necessary for digitisation in other sectors, generally between five and ten years (NBS’ 10th National BIM Report). The Fourth Industrial Revolution (Industry 4.0), focused on data and connections, has brought out the limits of our domain, which is unable to keep up with other sectors of production and services. While it is right that transition can only be triggered by an awareness of needs, it is also true that managing interactions with external fields is an equally relevant factor. This paradigm can also be extended to the associated tools, which must interact and be connected to the web to ensure proper data management and the realisation of the so-called “digital twin”. The new AEC software has metabolised the BIM methodology, or at least it is oriented towards it, although consistent and significant examples are still linked to large projects and established professional actors. There are no reports or analyses in the literature that disprove the inevitability of BIM for any product (infrastructure, buildings, etc.), application (new construction, renovation, restoration, etc.) or stakeholder (clients, designers, companies, etc.). Despite commercial maturity and a broad spectrum of technical standards that seem to be converging towards organicity, the most common image is that of an unfinished revolution. Apart from the abovementioned excellences, we can identify an uncertain use of tools, very specialised, based on approximations through trial and error, accompanied by a limited knowledge of the IT (Information Technology) and complexity behind the software front-end. They are compounded by the weaknesses of a mistaken approach to change. On the one hand, there is a fragmented experimentation, which has difficulties in dissemination and systemic interaction, and on the other hand, we have the inefficiency of a top-down body of rules and laws, which risks excluding the bottom from participation. In this panorama, research can play a fundamental role in the dissemination and systematisation, encouraging proper use of tools that can go beyond contingent needs. A hybrid approach of practice and theory, spiced up with training in the basic principles of IT, might be the desirable solution. If the AEC sector is not able to innovate and govern digital change, it will have to undergo it to adapt to this widespread need. The first experiments, which were strongly linked to IT innovations and software tools, have given way to an excessive theorisation of the method which leaves us without any practical feedback, a sign of general uncertainty in the direction to be taken. A representative example of this lack is the answer to the need for sharing data, information and thus knowledge, a problem common to all disciplines. In the field of BIM, this critical issue is ‘solved’, in conceptual and normative terms, with the introduction of a dedicated digital environment, defined Common Data Environment (CDE) first in BS 1192-1:2007 and later in PAS 1192-2:2013. Since then, CDEs described by other standards have been proposed, together with those developed by the academic world and those promoted by software houses, capable of responding to the problem in different and not always compatible ways. Similarly, the interaction between instruments cannot be left to the intensive work of the operator due to the inefficiency of the software. The lack of an AEC ontology cannot be solved by using only IFC classes, which are incapable of pursuing targets incompatible with their nature. The transition to object-oriented programming, with its specific elements, has not been completed in work scheduling and cost management appliances. Not all the seven ‘dimensions’ of BIM can be realised. For example, we do not have Project Management (PM) tools capable of independently predicting possible interactions between modelled objects. When BIM is employed as a graphical support for administration, it does not always blend perfectly with data storage systems. Its tools are too often used to replace the old 2D representations (PDF or DWG), without the perception of a strategic and not just instrumental change. Most of the focus for BIM is on new construction, with protocols being developed to make the production process more efficient. Its use concentrates on planning, design and integrated project delivery for buildings and infrastructure, but since recently, research interest shifted from earlier lifecycle (LC) phases to maintenance, refurbishment, deconstruction and end-of-life considerations, especially of complex structures. Promising benefits of efficient resource management motivate investigation to overcome uncertainties of building condition and deficient documentation, prevalent in existing heritage. A BIM protocol for the latter might end up being very similar to the one applied for new constructions, but this might not be the most effective way of approaching the problem. The public debate on BIM is often confusing and on occasion lacks a clear vision of final goals. To completely reformulate the problem and articulate it appropriately could be the first step to take to clarify the scenery just described. A tool in the gearbox of creative thinking is the so-called Kipling technique, whose archetypes can be found in the structure used by Greek-Roman philosophers to formulate the argumentations. It is a list of six seemingly trivial questions (5W+1H), yet they oblige you to re-examine each element or point of view related to a specific topic. The field of documentation and management of the built heritage is not an exception and a deepening of the details, developed by asking the right questions, can give an overview of the situation. To be extremely concise, it can be observed that in the sector of architecture, engineering and construction (Where?), data flows are today (When?) disorganised and productivity levels do not excel. This is because the world of construction, by its nature, is characterised by a certain level of disorder that does not allow the coordination of the figures involved in processes (Why?). Technicians (Who?) must therefore work hard to develop and adopt digital systems aimed at the effective and efficient management of the information at stake (What?). If the ‘Wh’ questions help to clarify and organise thinking, it is only with the ‘H’ of ‘How’ that we move on to action. How can we solve the problem, and especially how has it been addressed in literature? The BIM could be the answer to the first question if we critically analyse the many facets in which it has been presented since its introduction in the early 2000s. As mentioned for CDEs, the biggest problem of BIM related to research is probably the fragmented treatment, which is unable to contribute to the definition of a best practice. The processes of creating a model can be completely different for new and existing buildings. In the first case, the purpose is to provide a product that is articulated in the different phases of the building lifecycle (ISO 22263:2008-R2022), from inception to demolition. As the implementation of such models is not complete, isolated solutions, designed for a specific purpose, are too often employed. For existing fabrics, depending on the availability of previously developed BIMs, the repository can be updated or re-created. In Italy, structures from the 1970s account for more than 60% of all constructions and they are mainly without documentation in digital format. Therefore, in practice, complex and costly reverse engineering processes are almost always used to retrieve the necessary information. The panorama previously outlined is therefore very articulated and the complex problems that derive from it can be more extensive. This research is oriented towards a broader dissertation to provide solutions to the issues arising from a fragmented treatment of the topic (BIM, existing BIM, Historical/Heritage BIM). The methodology is interpreted as a system consisting of four interconnected elements: functional aspects, which analyse the capabilities or services provided by the BIM in the narrow sense (model construction) or by its accompanying software for data output. The functionalities can be internal (the seven ‘dimensions’) or connected through independent applications. Think of the structural calculation or any operation on specific requirements. This also includes analyses of the accuracy and efficiency of digital reproduction; informational aspects and interoperability, which include issues related to the structured organisation of knowledge and data exchange, to ensure interoperability between different software systems without loss of information; technical aspects, which refer to the construction of the model and depend on the Level of Development (LOD) relative to the designated functionalities. Some examples are data acquisition, processing, object recognition and modelling. The procedural pipeline can be differentiated between new and existing buildings; organisational and legal aspects, which define the general features of the model, the roles of the parties involved, their rights and responsibilities regarding information, their access to the model (reading and writing) or their obligation to provide a defined functionality. The four elements just introduced are interconnected and can be interpreted as nodes of a graph (leaving out here the presence of some elements external to the system). The arcs that connect them can be grouped into two fundamental paths: the flow of information, which moves from the technical aspects towards the organisational ones, and the flow of definition, which has the opposite orientation. The former coordinates the transfer of data coming from the model, while the latter the instances that, after the processing of such information, define or update the model itself. It is not easy to establish which of the two flows originates first because it depends more on the characteristics of the examined object. On this basis, two opposite expert categories are involved. On the one hand, those who operate in input, providing services for importing, acquiring, and monitoring data, transforming them into BIM models, and on the other, those who operate in output, producing reports or technical analyses (energy simulations, structural calculations, etc.). Once again, it is not easy to establish a hierarchy, as these are complementary roles within a cyclical process. The framework presented here certainly does not claim to identify and analyse all aspects of the methodology but is intended to provide a structured guide to reading the contents. All the proposed experiments can always be traced back to the four fundamental aspects described above. These will not be mere containers but will have the task of fostering the construction of connections between the elements investigated, an indispensable step for a systematisation of the methodology. The project follows two lines of research: the first is related to the technical aspects of BIM applied to existing constructions. The main objective is to formalise a procedural pipeline for reverse engineering implementations, especially with Scan-to-BIM techniques. Although the literature is rich in contributions analysing this topic, an organic treatment is lacking and there are many punctual experiences, related to the contingencies of the case study. Instead, our approach aims to generalise the results of applications and contribute to the outline of a best practice for the management of data derived from digital surveying. The proposed solutions attempt to foresee possible scenarios and offer valid alternatives to ensure a holistic treatment of the methodology. The structured organisation of models and outputs is not simply the product of factors emerging from the case study investigation, adapting to a wide range of situations without neglecting the requirements of current legislation and technical regulations. There is also no lack of in-depth studies on the processes of integrating survey data, mainly oriented towards low-level solutions, which are still not very widespread and therefore susceptible to refinement, contextualising the conclusions based on the design requirements. Downstream of the acquisitions and their processing, we devoted ourselves to object recognition as a preparatory and support phase for the semantic classification. Here again, the aim is to propose a cataloguing system that is flexible and compatible with building regulations; the second line of research focuses on the topics of data reliability and accuracy. The possibility of updating and reusing a model depends on precisely these two factors and, despite this, there is a lack of a unified framework to solve this critical issue. As far as the first topic is concerned, valid solutions emerge from the literature, but they struggle to establish themselves because they are not well integrated within the tools outlined by the technical standards. For this reason, our proposal for assessing reliability does not introduce any further novelties but aims to seek out solutions already used in parametric modelling or related fields, reforming them if necessary and lightening the notional load on technicians, who could make use of tools they know and master. Turning to the subject of accuracy, the main proposals focus on the survey phases, presenting modelling solutions that are either expeditious or in any case tied to the plug-ins of commercial software platforms. Alternatively, we suggest differentiated frameworks for survey operations and source-based virtualisation, focused on statistical data processing and implementable in any workflow, without worrying about the specificities of the software used. The choice of the case study is not random. The building block analysed, located in the historic centre of a municipality in the province of Salerno, stands out for its stratigraphic complexity and articulated relationship with the surrounding urban spaces. These elements, although strongly characterising, fully reflect the qualities of many historical centres, produced by centuries-old stratifications. Moreover, they present a wide range of criticalities, both for the surveying and modelling phases, which allow us to identify and field-test potentially the best solutions for the specificities of the case, contributing to enriching the range of experiences necessary to generalise the results of the research. As far as the structure of the book is concerned, Chapter 1 reconstructs, through an in-depth study of the state-of-the-art, the formation process of the BIM methodology, proposing a framework for the classification of its distinctive elements and framing within it the experiences of our research path. Chapter 2 focuses on technical aspects, formalising a workflow for Scan-to-BIM processes oriented towards a correct semantic classification of information content and traceability of data implemented in the virtualisation. Chapter 3 examines the issue of geometric attribute accuracy, proposing evaluation systems compatible with any case study, acquisition technique or parametric modelling platform. In conclusion, we critically analyse the objectives achieved and the possibilities of transferring the results. For easier reading and contextualisation, we have preferred to place the bibliographical references at the end of each chapter. The apparatus of this monograph builds on the results achieved during the PhD in Risk and Sustainability in Civil, Architectural and Environmental Engineering Systems at the Department of Civil Engineering - University of Salerno (Italy).
Protocols for tracking information content in the existing BIM
Andrea di Filippo
2023-01-01
Abstract
It is now 49 years since Eastman theorised what would become known as BIM: Building Information Modelling. Despite this, we can observe that the methodology, together with its associated tools, is still considered an exception to established practice, an eternal novelty with clearly something unfinished. If we exclude a few excellences, such as the United States and the United Kingdom, and countries vying to keep up, such as France and Italy, there are still a lot of regions where BIM is completely unsystematised. As a result, it is first implemented in large design studios and public projects and only then, with difficulty, does it spread to the rest of the market. While we could accept the idea of a silent revolution that takes time to gain a foothold, it is now clear that too much pressure has been applied to the AEC (Architecture Engineering Construction) sector, which was not ready for such a radical change, first in thought and then in practice. Putting this aside, the time required for innovation is in any case not compatible with that necessary for digitisation in other sectors, generally between five and ten years (NBS’ 10th National BIM Report). The Fourth Industrial Revolution (Industry 4.0), focused on data and connections, has brought out the limits of our domain, which is unable to keep up with other sectors of production and services. While it is right that transition can only be triggered by an awareness of needs, it is also true that managing interactions with external fields is an equally relevant factor. This paradigm can also be extended to the associated tools, which must interact and be connected to the web to ensure proper data management and the realisation of the so-called “digital twin”. The new AEC software has metabolised the BIM methodology, or at least it is oriented towards it, although consistent and significant examples are still linked to large projects and established professional actors. There are no reports or analyses in the literature that disprove the inevitability of BIM for any product (infrastructure, buildings, etc.), application (new construction, renovation, restoration, etc.) or stakeholder (clients, designers, companies, etc.). Despite commercial maturity and a broad spectrum of technical standards that seem to be converging towards organicity, the most common image is that of an unfinished revolution. Apart from the abovementioned excellences, we can identify an uncertain use of tools, very specialised, based on approximations through trial and error, accompanied by a limited knowledge of the IT (Information Technology) and complexity behind the software front-end. They are compounded by the weaknesses of a mistaken approach to change. On the one hand, there is a fragmented experimentation, which has difficulties in dissemination and systemic interaction, and on the other hand, we have the inefficiency of a top-down body of rules and laws, which risks excluding the bottom from participation. In this panorama, research can play a fundamental role in the dissemination and systematisation, encouraging proper use of tools that can go beyond contingent needs. A hybrid approach of practice and theory, spiced up with training in the basic principles of IT, might be the desirable solution. If the AEC sector is not able to innovate and govern digital change, it will have to undergo it to adapt to this widespread need. The first experiments, which were strongly linked to IT innovations and software tools, have given way to an excessive theorisation of the method which leaves us without any practical feedback, a sign of general uncertainty in the direction to be taken. A representative example of this lack is the answer to the need for sharing data, information and thus knowledge, a problem common to all disciplines. In the field of BIM, this critical issue is ‘solved’, in conceptual and normative terms, with the introduction of a dedicated digital environment, defined Common Data Environment (CDE) first in BS 1192-1:2007 and later in PAS 1192-2:2013. Since then, CDEs described by other standards have been proposed, together with those developed by the academic world and those promoted by software houses, capable of responding to the problem in different and not always compatible ways. Similarly, the interaction between instruments cannot be left to the intensive work of the operator due to the inefficiency of the software. The lack of an AEC ontology cannot be solved by using only IFC classes, which are incapable of pursuing targets incompatible with their nature. The transition to object-oriented programming, with its specific elements, has not been completed in work scheduling and cost management appliances. Not all the seven ‘dimensions’ of BIM can be realised. For example, we do not have Project Management (PM) tools capable of independently predicting possible interactions between modelled objects. When BIM is employed as a graphical support for administration, it does not always blend perfectly with data storage systems. Its tools are too often used to replace the old 2D representations (PDF or DWG), without the perception of a strategic and not just instrumental change. Most of the focus for BIM is on new construction, with protocols being developed to make the production process more efficient. Its use concentrates on planning, design and integrated project delivery for buildings and infrastructure, but since recently, research interest shifted from earlier lifecycle (LC) phases to maintenance, refurbishment, deconstruction and end-of-life considerations, especially of complex structures. Promising benefits of efficient resource management motivate investigation to overcome uncertainties of building condition and deficient documentation, prevalent in existing heritage. A BIM protocol for the latter might end up being very similar to the one applied for new constructions, but this might not be the most effective way of approaching the problem. The public debate on BIM is often confusing and on occasion lacks a clear vision of final goals. To completely reformulate the problem and articulate it appropriately could be the first step to take to clarify the scenery just described. A tool in the gearbox of creative thinking is the so-called Kipling technique, whose archetypes can be found in the structure used by Greek-Roman philosophers to formulate the argumentations. It is a list of six seemingly trivial questions (5W+1H), yet they oblige you to re-examine each element or point of view related to a specific topic. The field of documentation and management of the built heritage is not an exception and a deepening of the details, developed by asking the right questions, can give an overview of the situation. To be extremely concise, it can be observed that in the sector of architecture, engineering and construction (Where?), data flows are today (When?) disorganised and productivity levels do not excel. This is because the world of construction, by its nature, is characterised by a certain level of disorder that does not allow the coordination of the figures involved in processes (Why?). Technicians (Who?) must therefore work hard to develop and adopt digital systems aimed at the effective and efficient management of the information at stake (What?). If the ‘Wh’ questions help to clarify and organise thinking, it is only with the ‘H’ of ‘How’ that we move on to action. How can we solve the problem, and especially how has it been addressed in literature? The BIM could be the answer to the first question if we critically analyse the many facets in which it has been presented since its introduction in the early 2000s. As mentioned for CDEs, the biggest problem of BIM related to research is probably the fragmented treatment, which is unable to contribute to the definition of a best practice. The processes of creating a model can be completely different for new and existing buildings. In the first case, the purpose is to provide a product that is articulated in the different phases of the building lifecycle (ISO 22263:2008-R2022), from inception to demolition. As the implementation of such models is not complete, isolated solutions, designed for a specific purpose, are too often employed. For existing fabrics, depending on the availability of previously developed BIMs, the repository can be updated or re-created. In Italy, structures from the 1970s account for more than 60% of all constructions and they are mainly without documentation in digital format. Therefore, in practice, complex and costly reverse engineering processes are almost always used to retrieve the necessary information. The panorama previously outlined is therefore very articulated and the complex problems that derive from it can be more extensive. This research is oriented towards a broader dissertation to provide solutions to the issues arising from a fragmented treatment of the topic (BIM, existing BIM, Historical/Heritage BIM). The methodology is interpreted as a system consisting of four interconnected elements: functional aspects, which analyse the capabilities or services provided by the BIM in the narrow sense (model construction) or by its accompanying software for data output. The functionalities can be internal (the seven ‘dimensions’) or connected through independent applications. Think of the structural calculation or any operation on specific requirements. This also includes analyses of the accuracy and efficiency of digital reproduction; informational aspects and interoperability, which include issues related to the structured organisation of knowledge and data exchange, to ensure interoperability between different software systems without loss of information; technical aspects, which refer to the construction of the model and depend on the Level of Development (LOD) relative to the designated functionalities. Some examples are data acquisition, processing, object recognition and modelling. The procedural pipeline can be differentiated between new and existing buildings; organisational and legal aspects, which define the general features of the model, the roles of the parties involved, their rights and responsibilities regarding information, their access to the model (reading and writing) or their obligation to provide a defined functionality. The four elements just introduced are interconnected and can be interpreted as nodes of a graph (leaving out here the presence of some elements external to the system). The arcs that connect them can be grouped into two fundamental paths: the flow of information, which moves from the technical aspects towards the organisational ones, and the flow of definition, which has the opposite orientation. The former coordinates the transfer of data coming from the model, while the latter the instances that, after the processing of such information, define or update the model itself. It is not easy to establish which of the two flows originates first because it depends more on the characteristics of the examined object. On this basis, two opposite expert categories are involved. On the one hand, those who operate in input, providing services for importing, acquiring, and monitoring data, transforming them into BIM models, and on the other, those who operate in output, producing reports or technical analyses (energy simulations, structural calculations, etc.). Once again, it is not easy to establish a hierarchy, as these are complementary roles within a cyclical process. The framework presented here certainly does not claim to identify and analyse all aspects of the methodology but is intended to provide a structured guide to reading the contents. All the proposed experiments can always be traced back to the four fundamental aspects described above. These will not be mere containers but will have the task of fostering the construction of connections between the elements investigated, an indispensable step for a systematisation of the methodology. The project follows two lines of research: the first is related to the technical aspects of BIM applied to existing constructions. The main objective is to formalise a procedural pipeline for reverse engineering implementations, especially with Scan-to-BIM techniques. Although the literature is rich in contributions analysing this topic, an organic treatment is lacking and there are many punctual experiences, related to the contingencies of the case study. Instead, our approach aims to generalise the results of applications and contribute to the outline of a best practice for the management of data derived from digital surveying. The proposed solutions attempt to foresee possible scenarios and offer valid alternatives to ensure a holistic treatment of the methodology. The structured organisation of models and outputs is not simply the product of factors emerging from the case study investigation, adapting to a wide range of situations without neglecting the requirements of current legislation and technical regulations. There is also no lack of in-depth studies on the processes of integrating survey data, mainly oriented towards low-level solutions, which are still not very widespread and therefore susceptible to refinement, contextualising the conclusions based on the design requirements. Downstream of the acquisitions and their processing, we devoted ourselves to object recognition as a preparatory and support phase for the semantic classification. Here again, the aim is to propose a cataloguing system that is flexible and compatible with building regulations; the second line of research focuses on the topics of data reliability and accuracy. The possibility of updating and reusing a model depends on precisely these two factors and, despite this, there is a lack of a unified framework to solve this critical issue. As far as the first topic is concerned, valid solutions emerge from the literature, but they struggle to establish themselves because they are not well integrated within the tools outlined by the technical standards. For this reason, our proposal for assessing reliability does not introduce any further novelties but aims to seek out solutions already used in parametric modelling or related fields, reforming them if necessary and lightening the notional load on technicians, who could make use of tools they know and master. Turning to the subject of accuracy, the main proposals focus on the survey phases, presenting modelling solutions that are either expeditious or in any case tied to the plug-ins of commercial software platforms. Alternatively, we suggest differentiated frameworks for survey operations and source-based virtualisation, focused on statistical data processing and implementable in any workflow, without worrying about the specificities of the software used. The choice of the case study is not random. The building block analysed, located in the historic centre of a municipality in the province of Salerno, stands out for its stratigraphic complexity and articulated relationship with the surrounding urban spaces. These elements, although strongly characterising, fully reflect the qualities of many historical centres, produced by centuries-old stratifications. Moreover, they present a wide range of criticalities, both for the surveying and modelling phases, which allow us to identify and field-test potentially the best solutions for the specificities of the case, contributing to enriching the range of experiences necessary to generalise the results of the research. As far as the structure of the book is concerned, Chapter 1 reconstructs, through an in-depth study of the state-of-the-art, the formation process of the BIM methodology, proposing a framework for the classification of its distinctive elements and framing within it the experiences of our research path. Chapter 2 focuses on technical aspects, formalising a workflow for Scan-to-BIM processes oriented towards a correct semantic classification of information content and traceability of data implemented in the virtualisation. Chapter 3 examines the issue of geometric attribute accuracy, proposing evaluation systems compatible with any case study, acquisition technique or parametric modelling platform. In conclusion, we critically analyse the objectives achieved and the possibilities of transferring the results. For easier reading and contextualisation, we have preferred to place the bibliographical references at the end of each chapter. The apparatus of this monograph builds on the results achieved during the PhD in Risk and Sustainability in Civil, Architectural and Environmental Engineering Systems at the Department of Civil Engineering - University of Salerno (Italy).I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
