Building and Environment

,


Introduction
A building permit is the authorization to start the construction phase of a building project, granted by public authorities.The framework is not only provided by local entities, but also by national respectively federal governments.Hence there is a wide legal diversity as of now.It is part of a process of spatial planning that ensures that the requirements, set to ensure a sufficiently high quality, are met for new constructions, in order to guarantee a sustainable and controlled development of the built environment, benefiting communities, environment and economy [1,2].Several aspects are involved, such as functionality, sustainability, circularity, safety and security, disaster prevention, emergency management, environmental quality (noise, shadow, pollution, temperature), accessibility and more.As it is clear from such definition, several skills and disciplines are involved in the building permit use case.Many researchers have investigated the topic with respect to the many parts of which it is composed, sometimes proposing algorithms applying their findings (e.g., Narayanswamy et al. [7]).Moreover, several examples exist, where experiments, initiatives and pilots have been developed, together with demonstrators and more complete implementations, with different levels of progress.Therefore, the current knowledge achieved about the building permit digitalization and automation and the related implementations consist of a large amount of contributions from both academia and industry.However, such solutions remain somehow fragmented because of the fact that, due to the challenge represented by the many sub-issues entailed, they usually refer to few specific parts of the more complex workflow and related issues.Critical step to grasp the actual progress in the topic is to build a clear overview of the existing efforts with reference to a common framework in order to depict the complexity of the topic itself.For this reason, such an overview is the subject of this paper, with the final goal of supporting a consistent action plan, building upon the current efforts, understood and interpreted with respect to the overall topic.
Literature review works were already published previously: they mainly consider automated code compliance checking while introducing and comparing the state of the art [8][9][10].Automated code compliance checking is pretty much related to building permitting, but it does not cover the whole topic thoroughly.Ponnewitz and Schneider [11] proposed a review searching for process investigations regarding BIM-based building permitting.However, an overall review which classifies the very diverse publications in building permitting according to detailed criteria and refers to the specific components of the complex issue and workflow of digital building permitting is missing so far.This paper analyzes the available literature on the technical aspects of digital building permitting by considering the specific scope of each item.It allows unveiling a reliable picture of the current state of the art on each part of the digital building permitting process, enabling considerations on the overall progress and detecting the major gaps.Although legal aspects are intrinsically related, these and the social side of digital building permitting are not directly in the scope of this paper.
This paper is structured as follows: Section 2 provides an overview on the topic of the digitalization process of building permitting.The parts composing the overarching topic are explained in the Sections 2.1 to 2.3.In particular, the need for managing changes in operational workflows within public administrations, including smallsized municipalities, to manage both innovate processes and the new kinds of data, now digital data, is outlined in Section 2.1.At the end of this section, a building permit workflow is selected and mapped as a reference and the related steps are reported as a common framework.An introduction about the involved digital data, as three-dimensional (3D) information systems, follows in Section 2.2, together with the respective open standards and the efforts and challenges for their integration.Moreover, Section 2.3 shortly reports on the European network for Digital Building Permit (EUnet4DBP), multi-disciplinary and multi-sectoral collaboration within which the work described in this paper was performed.The research methodology that has been adopted for the development of this review is described in Section 3. The results are finally presented in Section 4 and discussed in Section 5, from where it is possible to point out the specific gaps in the overall development of building permit digitalization.Section 6 concludes the paper.

The digitalization of building permit issuing
The recent progress in software and hardware have made it possible to aspire to the great potential given by digitalization in terms of both workflows and data management for the practice of many processes and in many fields.Therefore, large resources are being invested worldwide, for example in the AEC industry, in public administration processes, in information production, management and analysis.
The building permit process is considered among the most promising use cases for automation via digital processes and digital data about buildings and the built environment, bringing relevant savings with respect to the current processing [12,13], which is still mostly manual in the great majority of cases.Fig. 1 presents the evolution of the building permit issuing from the traditional paper-based analogue process to a fully digitalized model-based setting.
Several studies have investigated the state of practice and workflows, by sending questionnaires, interviewing involved stakeholders and analysing current processes with required resources (i.e., time and money), in order to quantify the potential advantages related to the digitalization of building permit issuing in terms of economic savings and efficiency increase.Plazza et al. [14] analysed the process in Italy, while Samasoni et al. [15] studied the New Zealand case, calculating economic benefits due to the use of such a system of approximately $67.3 millions per year, taking into account the time saved by both applicants and building authorities, as well as benefits to contractors.From an Estonian report of the project introducing DBP in the country [16], a cost/benefit analysis revealed a potential saving of more than €500,000 per year, without considering the advantages and savings given by the improvement in rules clarity and interpretations, which are pointed out as source of time and effort savings also from the designers' side.A report by Advisors [17] also highlights a possible saving of 45 FTEs (full time equivalent) workforce per year in responsible authorities alone, with an increase in efficiency of about 8%-10% in workload.Such saved resources could instead be dedicated to a more careful check of noncompliance, the most complex cases and more advanced analysis that could help solving issues currently reported (e.g., exemptions, specific cases, specific plan needs), with a general increase of the quality of the built environment and of the job task.In addition, the proceedings burden would be reduced.The same concepts are confirmed by the Centre for Digital Built Britain (cdbb) report [18], in the United Kingdom (UK).According to the report, digitalizing and automating the system could enable a new level of transparency, and inherently build in the so-called ''Golden Thread of information''. 1 Although these numbers are presented outside their context and come from different cases, they all demonstrate a positive impact deriving from the introduction of a DBP system.The building permit issuing use case is therefore among the priorities for the digitalization process for public authorities.It is also expected that the digitalization of building permit issuing aids to economic development and housing in the context of smart city [19].
F. Noardo et al.Consequently, national and international legislation is moving in the direction that promotes digital transition for construction industry (see, for example the Directive 2014/24/EU of the European Council on public procurement 2 ).The European Commission (EC) 3 recently announced that a framework that enables digital permit issuing in the built environment will be developed within the renovation wave for Europe.However, the current practice is still far from the objective: in the vast majority of cases, a manual building permit process is still used, entailing subjective, error-prone, and time-consuming tasks and decisions, with high risk of ambiguity, inconsistency in assessments and delays over the entire construction process [20].
The process of issuing a building permit consists of several steps, with a great number of stakeholders involved, using many pieces of information.In many countries, this (rather diverse) information is still handled in analogue formats (or, at best, PDF documents) and in two-dimensional (2D) graphic representations.Moreover, in practice, the involved procedures and data are very heterogeneous and, in most cases, they can also be relatively informal -e.g., decisions depend on the local knowledge and expertise of individuals.These aspects make the transformation of the process challenging, since it is composed of many small issues to be solved separately, but still needing important reciprocal coordination and collaboration between several stakeholders and several different disciplines (e.g., code checking, building design, geoinformation management).
Several examples exist about the value of reusing and analysing (digital) building permit data.Gauthiez and Zeller [21] map the urban growth in the city of Lyon in the 17th and 18th century with the help of a Geographical Information Systems (GIS).To restructure the organizational structure of the building permit authorities and to optimize the building permit process in Prague, Hainc et al. [22] analysed the building permit data.Using census data, investigations of urban or residential growth related to the real estate market is exemplary proposed by Davis and Schaub [23], Shakro [24], and Cellmer and Jasiński [25].All approaches reflect the significance and the wide range of opportunities of accessible DBP-related data.

A reference, digital-enabling, building permitting workflow
As previously mentioned, even the current mostly (paper) documentbased building permitting is a very complex topic, articulated in many small components, characterized by high levels of multidisciplinary and inter-sectoral involvement.The first step towards the change was therefore the definition of how such a workflow should be changed in order to take advantage of digital data and digital tools, without neglecting the needs and the steps foreseen by more traditional procedures.
Several works are intended to solve this issue, with different levels of insight [14,[26][27][28][29][30].One example [31] was developed within the EuroSDR GeoBIM project4 by first harmonizing the workflows in use in the countries participating in the project, together with others found in literature [32].On this base, changes useful to facilitate the use of digital data (e.g., 3D city models and Building Information Modelling (BIM) models) by digital and (semi)automated tools were proposed, resulting in the workflow outlined in Noardo et al. [31], the draft of which was validated by several municipalities in the participating countries.
The conversion of the current, analogue workflows to digital workflows would, without doubt, requires an in-depth investigation itself.However, for the aims of this paper, and after checking for consistency with some other proposed examples [3,[32][33][34], we consider the steps depicted in Fig. 2 as a reference to investigate and code the state of the art of previous research efforts in the digitization of building permitting.
Fig. 2 illustrates the considered workflow steps 1-8 and their assignment to the parties in the DBP process.Each step is assigned to either the applicant or the authority (e.g., municipality).In some situations, the step can be assigned to both parties.This means that both parties perform this step together or either one or the other is responsible for the respective step.

From paper-based documents to digital information systems: Digital data for DBP
The key enabling factor to make the digitalization of building permit process achievable was the development of relatively new kinds of data and the technologies to produce and use them.The traditional process is based on 2D documents representing plans, facades and sections of the designed new building drawn according to orthogonal projections, and the context is usually represented by means of the technical and cadastral maps of the city.Even in the early examples of digital platforms to manage the building permit applications and documentations, when a dematerialization approach was implemented rather than a digitalization approach, those kinds of deliverables were the reference approaches, submitted as paper or in PDF format.However, such kinds of documents can only be analysed manually, without offering any potential towards automation.On the other hand, powerful information systems have been developed in the two main fields involved in building permit (city representation and building design), allowing a complex and hierarchical representation of both semantic information and geometry.Information systems allow a more complex data management as well as the automatic analysis of the obtained models; moreover, they enable collaboration environments to manage digital workflows.The following Section 2.2.1 introduces the adoption of BIM models in the field of building design.Section 2.2.2, describes GIS and their 3D evolution as 3D city models.The issue of their integration in a GeoBIM environment is addressed in Section 2.2.3.

Building information models and related open standards
BIM, which is now understood as an expression of digital innovation across the construction industry, is known as ''a set of technologies, processes and policies''5 enabling the ''use of a shared digital representation'' of a facility (i.e., building and infrastructure) through its life cycle in order ''to form a reliable basis for decisions'' [35] in a collaborative environment.A BIM model, in particular, is an objectoriented, data-rich, 3D parametric digital model generated during the modelling process and potentially containing, according to specific use cases, from the smallest elements of a building (e.g., bolts) to the construction site, following consequent semantic structures [36].The adoption of BIM is exponentially increasing in different sectors, particularly in the AEC industry [37].Reference standard for BIM-based data exchange and interoperability is the Industry Foundation Classes (IFC) by buildingSMART. 6IFC is a standardized, digital description of the built environment, including buildings and infrastructures.It is an open, international standard [38] meant to be vendor-neutral and developed to define an extensible set of consistent data representations for exchange between AEC tools and platforms [36].The development of IFC is an ongoing process.Another standard within the scope of this paper is the BIM Collaboration Format (BCF), the development of which started in 2009 and which "allows different BIM applications to communicate model-based issues with each other by leveraging IFC models that have been previously shared among project collaborators". 7odel View Definition (MVD) 8 is another important issue that is also developed by buildingSMART to ensure the interoperability between various applications and domains for data exchange.MVDs allow to define an implementable subset of the very rich IFC schema that encompasses a wide range of scopes, including sensors, permits, and conditions in order to use and focus on the required data within the specific process.By the next version of the IFC ( 5), MVDs will evolve to Information Delivery Specification (IdS) 9 that enables the exchange of models by providing a document that defines the requirements in computer-readable format.

Geoinformation and related open standards
Geospatial data with the derived geoinformation is intended to represent the entire city and broader landscape, i.e., built and natural environment, which are the context of the new planned buildings.In comparison to traditional maps, the digital geospatial data with geometry, topology and attribute data, including 3D city models, is managed within GIS.Geoinformation plays an essential role in the analysis of a city [39][40][41][42][43] and could be effectively used in the assessment of the reciprocal impacts of the new construction and the city.
Several data models exist to structure digital geospatial data and 3D city models, usually developed according to the needs of different cities or different countries, therefore several national data structures exist.In order to obtain internationally interoperable data, further schemas are provided by sovra-national organizations, that are usually considered by the studies intending to provide replicable solutions.Among the most popular data schemas are the data model proposed by the European Directive for a Spatial Data Infrastructure (INSPIRE) 10and CityGML, by the Open Geospatial Consortium (OGC) 11 (now in version 3.0), adopted in its original schema or by means of national Application Domain Extensions (ADEs) [44] in several cities and countries.However, for several reasons, among which the fact that such models are intended for a very wide scope, they are very complex and quite difficult to be implemented [45], also for being based on the Geography Markup Language (GML).Recently, CityJSON 12 was proposed as an alternative solution and approved by OGC, starting from a different implementation of the CityGML v.2.0 schema [46] and was proved to be very effective from an implementation point of view [47][48][49][50][51].

GIS-BIM integration (GeoBIM)
Using geoinformation about the context of a planned building, suitably integrated with design information derived from a BIM model is a critical step for making a number of analysis and for checking many regulations measuring not only the building performances alone, but also considering the impact of the designed building on the planning context and, in turn, of the context conditions on the building performances.For this reason, the interoperability and, optionally, the integration of these two spatial information systems are needed for the automation and digitalization of building permitting.The concept, also known as 'GeoBIM', has attracted increasing interest in recent years [52][53][54][55][56][57].Notwithstanding, several issues are still hindering a completely smooth integration and reciprocal conversions.They are often technical issues, related to the origin data themselves and the used open standards [58].However, most of the challenge lies in the alignment with the use cases needs and the related requirements.In order to solve this, a strict relation to practice is necessary and the automation of building permit use case has the potential to become an effective ground for it.

The European Network for Digital Building Permit (EUnet4DBP)
A complex framework was outlined through singular activities by research, industry and public entities until here, which has to be comprehensively considered for achieving an effective digitalization of building permit issuing.After experiencing it, several researchers and stakeholders working in the fields related to the use case and having approached the topic from several points of view, realized that a wider collaboration, covering different fields and different sectors was essential in order to obtain meaningful results.For this reason, at the beginning of 2020, the European Network for Digital Building Permits (EUnet4DBP) 13 was founded.
It is composed of researchers, public entities and companies combining their different experiences and skills in the common interest of the definition of a common strategy for the digitization of the building permit issuing process, with advantages to interoperability, procedures and data optimization, standardization and good implementations.This paper was conceived within this collaboration.
The three pillars on which the network activities are built are: (1) Process, (2) Rules and requirements and (3) Technology.All of them must be properly tackled and the related issues solved in order to achieve successful results w.r.t.digitalization of building permit process.
Adding higher detail, more specific ambitions were brainstormed within a workshop held on the 29th of May 2020, together with the requirements that should be fulfilled to reach them [59].Those definitions were later re-elaborated by a restricted team of the EU-net4DBP in order to make them clearer and consistent.A relevant part of the methodology for writing this paper was about associating the EUnet4DBP ambitions and requirements to each reviewed document as coding tags, so that we can also picture how much each of the ambitions and requirements are addressed at the moment (see Section 3.2).
During this process, it was possible for the authors to point out the remaining ambiguities of the formulated statements.Discussing their meaning and relevance was a great opportunity to re-define them in a clearer way and adding explanations and examples (see Table 1 and Table 2 for ambitions and requirements respectively).

Methodology
The literature review described in this paper includes both scientific and technical contributions, retrieved and analysed following the research methodology illustrated in Fig. 3.
It began with the collection of contributions by the listed keywords in Fig. 3, as proposed and agreed by the authors.The retrieved papers were object of an initial bibliometric analysis about DBP (Section 3.1).The scope of the bibliometric study relies on analysing trends, in the field of DBP in order to evaluate the international interest of the scientific community.
Later, the initial strong corpus of scientific contributions was integrated with more technical works, derived by the authors' knowledge.Others were collected by means of a questionnaire spread within the EUnet4DBP network and within the First EUnet4DBP workshop for DBP, 14 where relevant audience had attended, including researchers and stakeholders.The contributions presented at the workshop itself were added to our list as well, whether relevant.
A screening of both scientific literature and technical contributions was performed based on the coding schema, i.e., workflow steps, relevant EUnet4DBP ambitions and DBP requirements.Kind of contribution, level of progress, and country was also considered (see Section 3.2).The screening and coding process of retrieved papers was repeated twice, by different people, to reduce the possible bias in interpretation of the papers.

Paper retrieval
A bibliometric analysis was performed to investigate the DBPrelated research worldwide.The consultation and selection of the research works among scientific publications was carried out via the Scopus 15 and Web of Science 16 databases in the period from 2001 to 2021, which was assessed as being inclusive of the earliest experiments about DBP until the day of the search (3rd February 2021).The terms used for the search are reported in Fig. 3.These databases were chosen because they are accepted as the most reliable and comprehensive scientific databases and used by various researchers to conduct literature review (e.g., [60][61][62][63][64][65]). They also contain other digital, scientific databases such as IEEE Xplore17 so that researchers can access related papers in a topic without the need for searching different databases.It is important to note that these databases cover scientific papers from a wide range of publishers such as SAGE, Elsevier, MDPI, and Taylor & Francis.Advanced search option that considers title, abstract, and keyword was used to find the relevant scientific contributions.Journal articles, book chapters, and conference proceedings were included in the search.
In order to integrate the scientific state-of-the-art, it was necessary to consider other kinds of works, that remain outside the indexed scientific databases, but building up towards a progress on the topic as well.Among these we can list: non-indexed journal and conference papers; technical reports; MSc theses; tools and codes.
Further integration came from the experience of the authors and of other DBP experts, contacted both through the EUnet4DBP and in the context of the First EUnet4DBP workshop (1EUnet4DBPws) 18 by means of a questionnaire.The relevant contributions submitted to the 1EUnet4DBPws [66] were also associated to the workflow's steps to integrate the picture of the progress related to each step.The scope of the questionnaire was to collect the existing experiences, known by the members of the EUnet4DBP or attendants of the 1EUnet4DBPws (25-26 March 2021), on the digitization of the building permit process.
After a short introduction about the scope of the questionnaire, the participants were asked to report initiatives or documents by classifying them according to the addressed steps, shortly describe the contribution, provide, whether available, a reference and/or a link and to answer six multi-choice questions in order to enabling a further classification of these experiences.A printable version of the questionnaire is available at EUnet4DBP [67].The questionnaire was pre-tested by all authors of this paper providing comments and suggestions to improve the quality and clarity of the language, of the structure and of the contents.The participants in the 1EUnet4DBPws filled the questionnaire when submitting their contributions to the event, while it has been shared with all the 48 members of the EUnet4DBP via email, as a web link, starting from 22nd March 2021 until 12th March 2021.We collected 21 answers.
By doing this, we integrated the list related to each step with both the relevant contributions submitted to the 1EUnet4DBPws, with the results coming from the questionnaire with resources less related to the DBPs per se, but no less relevant for the specific step (for example about MVDs and IFC validation in Step 4).

Coding schema for the classification of the collected contributions
The scientific papers retrieved as described in Section 3.1 were distributed among the authors, who assessed their relevance for the scope of this paper, and classified them according to their contents, which is considered at different levels: the intended high-level goals (EUnet4DBP ambitions), intermediate objectives (EUnet4DBP requirements) and the concrete reference to the DBP issuing workflow steps.Moreover, the level of progress of each work and the kind of contributions are noted in order to support the reflection about the work available in more detail, quantitatively.Fig. 4 summarizes the framework used for the classification.
The steps of the workflow are introduced in Section 2.1.The focus of each study, or part of it, was classified according to them: Step 1 Rule interpretation and digitalization of city and building regulations; Step 2 3D city models and useful related geospatial data preparation; Step 3 Pre-consultation and preliminary analysis (i.e., reading and analysis of the city and building data as references and constraints to the design; selection of the necessary information);

Table 2
Re-phrased requirements of the EUnet4DBP.

N. Requirement r1
Digitalize the mindset of public officers.r2 A roadmap and a change framework towards a fully DBP process.r3 Normative text should be interpretable.r4 Machine readable building codes.r5 Several kinds of data can be involved in the process (GIS, BIM, etc.), with related clear information requirements.r6 Understanding the necessary process steps r7 Alignment across Europe and beyond in Scope and Ambition r8 Standardization r9 Common dictionaries r10 Modelling conventions and guidelines r11 Interoperability and APIs Step 4 BIM and export to IFC (including IFC readiness and possible MVDs); Step 5 Application review (data quality check) -i.e., IFC validation; Step 6 Conversion to and integration with 3D city model or geospatial data; Step 7 Application review (content) -i.e., regulations checks and reporting; Step 8 Completion of the works and building authorities notification (delivery of as-built BIM and further information useful for following building and city life).
This classification allowed building an overview of the current progress related to each step of the workflow, which is useful to detect research and development gaps.
Second, tags about the ambitions and DBP requirements addressed in each work were added, according to the definitions of the EU-net4DBP (Section 2.3).The original EUnet4DBP ambitions and related DBP requirements were discussed by the authors within the context of this paper.It was an opportunity to rephrase them more clearly and removing residual ambiguity.The resulting list is reported and explained in Table 1 and Table 2.
Such tags will allow discussing a different kind of gap, related to the single ambitions and requirements, which could be in some cases more typical of some step, but they are often transversal throughout the whole workflow.
Each paper can be focused one or more steps as well as one or more ambitions and requirements.For this reason, in the cases in which it occurs that more than one step/ambition/requirement is addressed, the paper is repeated in each of the concerned classifications, in order to make a comprehensive analysis of the contents spread across the reviewed contributions.
Moreover, kind of contribution, level of progress, and country have been used also as coding tags.The authors noted the kind of contribution according to the following classification: A Research (A1) Literature review, evaluation of software or procedure and data review; (A2) Solid research or application experiment report, possibly supported by data; B Implementation developments (B1) Demonstrator or early-stage experiments, preliminary to tools implementation, including studies implemented in a proof of concepts; (B2) Tools tested with many data, potentially useable/used in practice; C Initiatives in operational environments -Inclusive and comprehensive works, likely applied within working environments.
Finally, the authors assessed the level of progress of each contribution according to the discipline of Project Management [68]: Within such a work, the risk of bias in assigning the classification was very high, due to personal field of expertise and interests and subjective judgement or interpretations.Therefore, in order to limit such a risk, after a first screening and review round in which all the papers retrieved from the scientific databases were distributed equally among the authors, we redistributed the ones judged relevant for the DBP use case for going through a second round of review, by someone having different background than the first reviewer.This step was very useful for pointing out all the relevant aspects of the papers, since people with different backgrounds can often appreciate different faces of the same work.

Interpretation of results
As a last step in our methodology, we tried to draw up relevant points by interpreting and summarizing the analysed works.Again, we used both a quantitative method, by making graphs of the obtained results and a more qualitative summary of the contents of the papers related to each step.Due to the high level of interdisciplinarity of the topic, we could have great advantage by the multidisciplinary composition of the authoring team of this paper in assessing the completeness and internal consistency of such a complex and multi-faceted overview.

Results
In total, we considered relevant for our classification 111 papers and works.Fig. 5 illustrates the number of contributions by year that are considered in this study.It can be seen from this figure that although there were at most five contributions until 2013, last seven years except 2017 have at least 10 contributions.Fig. 5 thus shows the ever-increasing interest for DBP.
Table 3 itemizes the publication locations that have at least 5 contributions.It can be seen from this table that Automation in Construction is the prominent journal followed by the Journal of Information Technology in Construction.Table 3 also shows that a notable number of contributions come from the proceedings of important organizations related to different fields of study such as civil engineering and geomatics engineering.
It is important to note that some contributions that were classified as relevant are not assignable to a certain step.These are the following 8 contributions: Plazza et al. [14], Ponnewitz and Bargstaedt [26], Messaoudi and Nawari [27],Messaoudi et al. [28], Ponnewitz [29], Noardo et al. [32], Noardo et al. [59], and Zhong et al. [69].Contently, these contributions are integrated in the previous sections of this paper since they mainly consider the overall building permit process.They are not included in the further analysis.

State of the art and highlights for each of the steps of the reference workflow
In this section, the main contents and relevant topics of the papers, classified according to each step of the DBP process, are summarized.Moreover, heatmaps depict the evolution of the investigation related to each step with respect to: kind of contribution, level of progress and publication years intervals.They support the interpretation of the work done for each step.For example, the steps with highest priority were likely addressed earlier.The level of challenge can also be grasped: the most challenging steps could be addressed for longer, but reach lower progress.The nature of the challenge can emerge: would it be a more theoretic framework issue, it would be more addressed by types A; would it be a practice-related issue, it would be more towards type C; would it be an implementation issue, it is in type B. The change through the time in the nature of challenge addressed can mean that basic premises have been solved; for example, new technology has unblocked the implementation works, or new laws enforced applications.The remaining gaps can be visualized more effectively with such representations: for example, some applied work could be done in the past, but could not be related to the most advanced progress in the theoretical framework and implementations.The description and full classification of each contribution is available in the Appendix.

Step 1: Rule interpretation and digitalization of city and building regulations
This step is related to rule interpretation, the process of conversion of the natural language of city and building regulations into computable parameters and constraints for automating and digitalizing the building permitting process.It is important to highlight that there exist a wide range of rule types that should be considered in building permit issuing, for example, geometric properties, land use type and development, rights of way, building loads, and neighbourhood law.The interpretation of rules into a machine processable format is one of the major issues in automated rule-checking and it is a fundamental and challenging step in the design compliance checking [84,92].Moreover, the relevant information in documents such as public laws, codes and   regulative standards need to be captured in a time and cost effective way [93].
Several selection criteria and rule interpretation processes are described in literature.Malsane et al. [20] describes how knowledge formalization of building codes could provide "suitable, significant and required data for the development of the Building Regulation-specific object modelling".In particular, they claimed how the formalization of building regulations should include the classification of regulation clauses into "those which are computer-interpretable (declarative) and those which are not (informative)".The former provide a direct meaning to be interpreted (e.g., simple geometrical rules which when applied to an element can return true or false), while the latter contain data only partially suitable for interpretation into computer rules that can be processed (e.g., information is not obvious as checkable, needs human interpretation to understand the exact content and meaning).Finally, a remaining category of clauses exist that can be considered as unsuitable for automated compliance checking.
The interpretation process could rely on the programmer's interpretation and translation of the written rules into computer code [12], but in most cases the logic of the human language statements is formally interpreted and then translated.In fact, it is important to acknowledge that building regulations are complex and at times subjective in nature and therefore building regulation experts need to be involved in their conversion to computer interpretable rules to ensure the correct interpretations for the code checking [20].The Requirement, Applicability, Selection, and Exception (RASE) methodology provides an easy to understand, simple methodology for deconstructing rule sentences and to extract semantics from building regulations for compliance checking [73,77].RASE, as well as multiple hybrid approaches based on that methodology [71], have been demonstrated to operate on a different types of normative documents with a trustworthy results [93].Moreover, Natural Language Processing (NLP) and deep learning-based analysis of building regulations and their translation into a computerreadable format have been also investigated to support automated rule checking systems (e.g., [75,81]).Zhang and El-Gohary [74] for example proposed an NLP-based method to extend the IFC schema to incorporate compliance checking-related information in an objective and semi-automated manner.Zhang and El-Gohary [82] proposes a machine learning-based approach to automatically match the building code concepts and relations to their equivalent ones in the IFC.Almost all studies, in fact, aim at converting regulations to a machine-readable format to check the rules using IFC data of the buildings.
Among the reviewed papers that have been classified as related to Step 1 (full list in Table A.4), an important research project is represented by the development of KBimCode, a software-independent and standardized script language that was developed to represent the Korean Building Act as explicit computable rules [13].Park and Lee [79] demonstrated the applicability of KBimCode for checking regulations on the installation of an elevator and the installation of escape stairs.Several other studies focused on improving the classifications of KBim-Code [13,72,76,[78][79][80].Fire safety [20,94], accessibility [95], space requirements [92,96] are among the most implemented requirements in semi-automated code checking.Moreover, Noardo et al. [83] explored the building permit use case in collaboration with the municipality of Rotterdam (NL).The interpretation and formalization of regulation for building height, overhang and tower ratio is proposed as preliminary results.In the context of city-level checks, also Van Berlo et al. [70] proposes the storage of spatial planing information in 3D based on CityGML and the Dutch zoning data.They also proposed the conversion of such a dataset to IFC.Fig. 6 shows that an advanced but isolated work was done in the years interval 2006-2010, but the investigation is developing more in the following years, and especially in the last five years, with many contribution to early stage implementations (B1), reaching the three most advanced levels of progress.

Step 2: 3D city models and useful related geospatial data preparation
Five works were found, detailing the use of geospatial data and 3D city models as input for building permitting (Table A.5).Among the available standards, reviewed by Guler and Yomralioglu [34], the most considered Open data model for representing geospatial data and 3D city models as a base for the automatic analysis to check the regulations (mainly zoning and urbanistic rules) is CityGML, which is extended and adapted to the national standards in some cases, such as by Eriksson  [86].In some works, the requirements of geoinformation are specified (e.g., [43]), which is a relevant step towards interoperability.
Other papers [88][89][90] deal with the conversions of 3D city models to BIM, either considering the respective standards CityGML and IFC or more general national or proprietary widely used formats.Conversion is an essential step for allowing the designers to consider geoinformation as a suitable reference.
Besides the studies listed in Table A.5, other pilots are being implemented in which the geoinformation component plays a very relevant role and will likely publish their solutions in the next future, such as in the State of Geneva and Dubai [66].
In Fig. 7, it is visible how an initial experiment was done almost twenty years ago, but most of the evolution had place in the last years, in an intermediate level of implementation between research and early-stage implementations and overall reaching and intermediate progress.

Step 3: Pre-consultation and preliminary analysis
Step 3 (Table A.6) describes pre-consultation and preliminary analysis in advance of the submission of a building permit application which is proceeded by the applicant together with the authority.In some cases either the applicant or the authority is in charge of the preliminary analysis.This steps includes, among others, reading and analysis of the city and building data as reference and constrain to the design as well as selection of the necessary information.In conclusion, Step 3 papers consider early design stage approaches regarding the analysis of building codes [70,94,[97][98][99] and the development of frameworks [100,101].Within the review process, eleven papers were classified as a Step 3 F. Noardo et al.

Table A.5
Classification of contributions related to the Step 2, i.e., the modelling and use of 3d city models and geospatial data as input for DBPs.

Entry
Description AR AA KoC Progress Country Trdla [85] (2021) The Central Repository of the Liberec Region.The building will be as a storage of museum collections ensuring optimal conditions.The building holds up to hundreds of thousands of subjects, which are needed thousands of square metres.contribution.Among these, three are contributions from the practical field [77,102,103].The practical approaches focus on tool assistance of the early design phase and design check.
Step 3 started being investigated from a fifteen years ago, but became mostly interesting in the last five years, as most of the other steps (Fig. 8).The progress reached by the different contributions is various.

Step 4: Building information modelling and export to IFC
Step 4 refers to the research, development and applications related to the necessary IFC-related requirements to ensure the necessary information for enabling DBP assessment.A part of the literature found in this concern is still strongly focused on specific applications of IFCbased code checking, which naturally includes the necessary modelling rules in regard to both geometric and non-geometric information in IFC or proprietary formats.This can be seen in the examples of Malsane et al. [20] and Preidel and Borrmann [115] for fire safety checks, or the example of Zhou and El-Gohary [112] for energy analysis/checking.Focus has also been given to approaches that include the establishment of MVDs towards the proper setting/delivery of information requirements in IFCs for code checking and ultimately for building permits (e.g., [105,108]).
The use of IFC BIM models towards the process of DBPs has been active in Singapore for several years, through an initial IFC-to-FORNAX translation, and then applying proprietary rule-checking [106].More recent efforts have been gaining traction in IFC use towards DBPs namely in Estonia [110] and Czech Republic [109].
The contributions related to step 4 have been occurring since the early 2000's, as shown in Fig. 9 (and provided in detail in Table A.7).However, most contributions have been found to concentrate on the last five years (2017-2021).In the last years, we notice an increase in number of items and also a progress in terms of relation to implementation (including B and C contributions) at an executing level.
A general overview of technologies for requirement checking on building (IFC) models.Describes how one example requirement can be formalized and queries an IFC-model as a demonstration.

Step 5: Application reviewed (data quality check)
Step 5 describes the implementation of the ''validation checking'' concept as described in Hjelseth [6].Validation checking (e.g., BIM/IFC validation) validates the informative content of BIM models and checks if they embed the data set required.Therefore, it ensures quality and internal consistency of a BIM model, which is crucial in an interoperable BIM environment based on neutral data formats, such as IFC, when it is necessary to clearly formalize information exchange procedures.Moreover, the DBP use case requires a larger dataset including the integration between BIM and GIS for a wider application in the assessment procedure.For example, van Berlo and Papadonikolaki [107] proposes the GeoBIM integration and conversion of regulation data into spatial format (IFC).Moreover, the relationship between Geodata and BIM is discussed in Johansson [116], which evaluates how BIM data in the IFC format could be converted to 3D geodata in order to provide specifications for supporting an unbroken digital data flow.Papers classified in Step 5 are mainly focused on methods to check BIM contents in terms of required data for enabling code checking.Zhang and El-Gohary [117,118] proposed an automated method for extracting design information from BIM models in the IFC format into a semantic logic-based representation that is aligned with a matching semantic logic-based representation of regulatory information.Ciribini et al. [96] adopted an established model checker, Solibri Model Checker (SMC), 19 to validate the informative content of BIM models prior to proceed with code checking.Choi and Kim [119] suggested an open BIM-based design quality checking process according to which architects create OpenBIM data using specific guidelines and BIM libraries previously developed.Lee et al. [108] proposes a 19 https://www.solibri.com/method to validate IFC models according to MVDs and implements a demonstrator based on the Building Smart IfcDoc tool.
Step 5 also contains references to the development of checking tools adopting several technologies, including linked data [120,121], addon in BIM authoring tools [7], open software toolset [122] and the adoption of visual programming language for quality checks of the information conveyed in a given building model [92,115].
For Step 5, literature was produced only in the last five years, with an increasing interest on data quality and interoperable workflow.On the other hand, in some cases research works could reach a high level of implementation (B2) and progress (Fig. 10).

Step 6: Conversion to and integration with 3D city model or geospatial data models
This step concerns the conversion of BIM data to geospatial data and integration with a 3D city model which is a research topic that has   attracted increasing interest in the recent years and is often referred to as GeoBIM (See 2.2.3).In the building permit workflow the integration is performed to place the (planned) designed building (BIM) in context (GIS).The step is crucial for checking regulations that require information about the surroundings of the planned building and it enables analysis of how the planned building will influence already existing buildings in the area.Seven publications were related to this step with three of them being scientific contributions [86,88,123], one report [116], one presentation from the Eunet4DBP workshop [124] and two web resources describing pilot studies [125,126].In most studies, IFC models were converted to CityGML (incl.national ADEs), and the most common tool for performing the conversion was the extract, transform, and load tool Feature Manipulation Engine (FME) from SAFE Software. 20 It is also important to note that in addition to integrating BIM data and geospatial data to check that a planned building conforms to the regulations before the building permit is approved, the methods described under Step 3 were also applied to update an existing 3D city model with an as-built BIM model delivered in Step 8 of the workflow.
As for Step 5, also for Step 6 the contributions were developed only in the last five years and reach an intermediate level of progress for early-stage implementations and a case of application in operational environment (Fig. 11).

Step 7: Application review (content) -i.e., regulations checks and reporting
This step is related to the automatic checking of the content in a BIM model for a specific use.In building permitting, this means that the BIM model has to be examined against specific requirements defined in Codes and Regulations (i.e., Building Code, Urban Plan, Fire Code, Health and Safety code, etc.) to obtain a building permit approval.
As, along the design and construction process, there are many Codes and Regulations to be in compliance with, there are multiples uses for checking BIM models as well.Many studies focus on developing 20 https://www.safe.com/Accessed 17th January 2022.tools and prototypes for checking Urban Plan, in a GeoBIM perspective.These studies are localized mainly in The Netherlands [70,83,127] and in Sweden [86,116,123,128], while, at commercial level, Spain stands out with the development of the tool Cype Urban [129].In South Korea, a BIM-based e-submission and automated code compliance checking system for building permitting (KBIM) has been developed with the support of the Ministry of Land, Infrastructure, and Transport [33,98,130].Other studies intend to prototype applications for specific uses without considering country-related constraints.In these cases, the code checking uses are for: energy performance evaluations [113], acoustic performance checking [131] and sustainability regulations [132], fire codes [94,133,134], deep foundation construction [135], spatial program [136] and building code [7,100,110,137] compliance.
As stated by Hjelseth [138], the main challenges in achieving a BIMbased code compliance checking are: the complexity of Codes and the impracticability of existing automated checking approaches.The key problems are: rules interpretation, semantic matching of requirements in Codes and in BIM models and the standardization of information requirements [135].
Commercial software are not supportive in overcoming these challenges since their functionalities are based on the use of simple rules and unspecified content of information in the BIM model [6].For this reason, almost all studies and experiences in Table A.10 focus on developing bespoke prototypes and tools rather than on improving existing commercial solutions: only few studies [94,100] develop a code checking system based on Autodesk Revit and its VPL tool Dynamo, 21  while [70,96,139]     the studies focus on converting information extracted from the BIM model into ontology instances, ready to be automatically checked against logic-represented regulatory rules using NLP techniques [117], Java environment [141] and SPARQL Protocol and RDF Query Language [145].NLP techniques are used also to read and transform requirements of Codes into machine readable rules [81,114,131].Other very common techniques are: VPL, used by [92,130], and Visual Code Checking Language [142].
At the end of the content review process, the reporting of checking results should include information about all the checking performed   with relevant codes listed for reference.Results are displayed in a table format or in a PDF format, which should include the list of checks performed, engineering parameters (extracted from model), the name of checking rules, the corresponding code articles, and checking conclusions [135].
Step 7 is where most of research and literature has focused, with relevant effort from more than ten years ago.We can see from Fig. 12 that often an intermediate-to-high level of progress is reached, although limited to research and early-stages implementations (B1) until five years ago.In the last years, also the implementation level has raised and there are already some examples of B2 works at a level executing, plus two cases of application in operational environment, one of them at progress 'Closing'.

Step 8: Completion of the works and building authorities notification (delivery of as-built BIM and further information useful for following building and city life)
This step includes the completion of the works done previously as well as notification issues.After the application has been reviewed in terms of content, the responsible building authority must transfer a formal notification, often accompanied by notification of charges.In order to execute this part of the process in an aligned legal framework, building codes and potentially general administrative law need to be adapted.For example the building code of the state of Lower Saxony (Germany) [150] requires written form and is currently in progress towards a DBP paperless process.With respect to the use of mature building application platforms, not only communication between building authorities and applicants during the application process but also final notifications to complete these administrative acts are supposed to be an integrated module in alignment with the underlying law.Furthermore, the use of existing digital data, e.g., for CAFM or further city planning, are part of Step 8. Two publications were related to Step 8. Chognard et al. [88] describe a DBP procedure for a Swiss municipality.City of Sant Feliu [149] describes a development towards a smart city, offering public online services including participation in city planning.
The two contributions about Step 8 were developed in the last five years and are early-stage implementations, at intermediate-high level of progress (Fig. 13).

Excluded but related papers
We included in the review only the documents directly relevant to the scope of building permits digitalization, therefore, some other studies that contain relevant contents, but are originally intended for other use cases were excluded.However, for the sake of completeness, we report here some of those papers, which were selected by the algorithms in the bibliometric search and were assessed by the authors as somehow relevant although not directly related to the building permits themselves.
Soliman-Junior et al. [151] discusses available commercial tools (dRofus and SMC) to assist in the data requirements management and check during the design phase, in order to improve the final design quality.Zhang and Beetz [152] extend SPARQL to query IFCmodels for code compliance checking in a linked data environment and implements a prototype for a case study.Zhang and El-Gohary [153] propose a method to extend the IFC-schema to facilitate automatic code compliance checking by matching concepts in regulations with concepts in IFC and classify relationships with machine learning methods.Fan et al. [154] creates a framework for rule checking of BIM models with the focus on a user-friendly and flexible interface that enables a designer to include the rules valid for the model to check via a visual programming interface.Métral et al. [155] presents a model based on ontologies for compliance check of rules related to subsurface objects in 3D city models.Lee et al. [156] presents a method for automated rule checking of BIM models that also guides the   designer in the design process by automatically suggesting changes to comply to the rules.Guedes et al. [157] suggests to develop BIM information modelling guidelines to ensure that the necessary information are included in the IFC model to facilitate automated code checking.The study is focusing on airport design.Qi et al. [158] proposes the use of the Solibry Model Checker to check the rules related to the workers' safety.Trebbi et al. [159] considers available tools to be used for the checks.Dimyadi et al. [160] demonstrates how a subset of the New Zealand Building Code can be encoded into a computer executable format with the open standard Business Process Model and Notation.Kim et al. [161] presents a method, based on visual language programming, to translate natural language into computer executable code.Evans and Counsell [162] develops course design in architecture school that contains the BIM-based automated code checking using SMC to enhance the ability and awareness of the students.Mena et al. [163] proposes a new XML-based standard that facilitates the project information flow required for different phases, including building permit, in the life cycle of buildings.Nawari and Alsaffar [164] seeks to examine the role of BIM in improving the permitting procedure and proposes a framework for simplifying the permitting procedure for residential housing with a focus on architectural and structural design.Choi et al. [165] proposes an approach that allows to extract required information from IFC data by adding extra attributes in Revit to assess building design in terms of energy performance.Fahad and Bus [166] investigates the geolocation-based relationships of the objects (e.g., fire fighting device) in the IFC file using Well-Known Text (WKT) and graph databases, namely Stardog and GraphDB, for compliance checking.Belliard [167] examines the use of BIM in plan review process within building permitting to support traditional 2D approach by surveying with local architects.Strobbe et al. [168], Shi and Roman [95], and De Vos et al. [169] consider assistance of the design phase according to building regulations prior to the submission of the building permit application.Zdravkovic and Ostman [170] deal with presenting GEOinfo in public services.Lee et al. [171] investigate data exchange standards.Karim et al. [172] examine the dissemination of spatial information in the case of Malaysia.

Discussion
The classification of the reviewed contributions allowed the authors to draw the general picture about the available works for each step of the DBP process.Moreover, previous studies have been classified according to several aspects: • their overall distribution among the DBP steps to highlight where previous research efforts have been concentrated to identify any gaps (Section 5.1); • reference country, to evaluate if a geographical leadership exists in DBP-related research activities (Section 5.2); • kind of contribution and level of progress (Section 5.3); • main EUnet4DBP ambitions and related requirements addressed (Section 5.4).
In the following subsections such results are represented and discussed.

How much is each step investigated
As visible in the distribution in Fig. 14, the focus of science and other investigations in the last twenty years was mainly on Step 7, i.e., application review with respect to the compliance of the content of the model to building and city regulations.Indeed, the automation of checks is generally seen as the first condition to digitalize the building permitting process.In addition, the software available on the market to support this step (such as the one developed by companies like Solibri, CYPE, Xinaps, ACCA software, among others) are not all counted in this paper, but increases even more the effort directed at solving the regulations checking step.
Step 4, building model preparation, and Step 1, provision of digital regulations, clearly show how the use case was first investigated in the field of BIM, rather than in its connection to the geoinformation field.
Step 5, related mainly to BIM and IFC validation as a main step to allow interoperability between authoring plaftorms and checking tools, comes later, with a 10% of contributions and starting only from 2015 (Fig. 15), a few years after the IFC became an ISO standard in 2013.
Step 3 is probably less addressed because it is considered a minor issue with respect to other enabling phases of the digital workflow.
Finally, Step 2 and Step 6 deal with the connection between 3D city models and BIM and, more in general, with the use of 3D city models to check the regulations and support the DBP process.Besides one very early study from 2005 [43], the interests in those steps arises only from 2017.We expect it to grow in future (Fig. 15).
Finally, Step 8 was addressed earlier, but with an overall low effort, probably due to the higher priority of previous steps.

What countries are working on digital building permit process
Fig. 16 and Fig. 17 show the distribution of research and development efforts for digitalizing the building permitting process globally with the largest number of contributions in North America and Europe.Some countries in Asia and Middle East, with South Korea as the second major actor in the world, have also produced a relevant number of works and studies on the topic.
Noteworthy to mention is that the number of works do not represent the level of developments of the countries regarding digitalization of building permit issuing.For example, although Estonia and Norway have noteworthy experiences for the digital building permit process, the reviewed items are not many.

Kind of contribution and progress
Fig. 18 shows that the majority of contributions as of now are base studies on applications (A2), most likely implemented within demonstrators and proof of concepts (B1).Those two categories aim to address the most complex issues involved in the digital building permit process, often strictly related to specific use cases.In these, A1 works (literature etc.) are less frequent and need to address basic research questions.As of now, early implementations seldom reach a higher maturity (B2).Purely commercial approaches are not considered in this review.Category C is the least represented one.
Fig. 19 displays the distribution of items according to the kind of contribution throughout the steps.Step 1 (digitalization of regulations) presents a rather consistent distribution of items per kind, passing through basic research and proof of concepts to go towards more mature implementations.Step 2 (preparation of 3D city models as input) and Step 3 (pre-consultation and preliminary analysis) as well as Step 6 (GeoBIM conversions) have a similar distribution as Step 1, although no A1 items are reported, while there is research in operational environment.Step 4 (BIM modelling and IFC export) emerges in the graph in Fig. 19 with mainly on A2 works, whilst the implementation is still weak.[173,174].At the same time, Step 5 also contains references to the development of checking tools adopting various technologies.However, no applications (category C) are reported, indication that a validation on the field could still be missing.Step 7 (Content checking against regulations) counts on the higher effort both in terms of research and early implementations and in attempts in operational environment (C).In this case the two aspects research and application are tackled in parallel.There are only two items for Step 8 (Completion and notification) which are both located in the early implementation category (B1).
In Fig. 20 it is shown how most of the works are in an intermediate 'Executing' phase, even though some have also reached a higher progress.7 out of 111 are in 'Conception and initiation' phase and 13 in 'Definition and planning'.Considering project development, the majority of works are rather in work than being planned.
Fig. 21 shows that steps 2 and 4 are mainly in 'Executing' level of progress.For the others, works are present in all the levels of progress.Step 7 contains a significant amount of works in the 'Closing' phase.Step 8 includes the least amount of works with no significant difference between the progress levels.Being the most advanced step, this is a rather expected drop with respect to the previous steps for this sequential process.

Which are the EUnet4DBP ambitions and related requirements addressed
Fig. 22 shows how the T3 ambition, i.e., 'Technologies for data visualization, data analysis and data manipulation', is the most addressed one, especially, but not limited to, by Step 7.This is unsurprising because the role of technology in regulation checking and reporting is evident.
The other ambitions related to the technology, T1 -'Interoperable scalable systems useful at different levels (European, national, municipality) and in different Countries' -and T2 -'Platforms allowing a data-centric approach across the whole facility life-cycle by means of a central management of accesses' are instead less addressed.This might be because creating and operating the comprehensive system that enables the data flow for DBP is difficult since the various stakeholders and organizations get involved in the workflow.Some work is available for T2, especially related to Steps 2, about the provision of 3D city model and suitable geospatial data, and Step 3, about the pre-consultation phase.But in very few cases T1 is addressed.The ambitions related to rules and requirements, R1 -'Unambiguous regulations interpretable as computational parameters, algorithms, clear constraints and criteria' -and R2 -'Explicit specification of data requirements' are similarly addressed mainly for the Step 1, about digitalization of regulations, and Step 7 about automation of regulations checks.Moreover, R2 is also significantly addressed for Step 4, about BIM and IFC preparation, which should be compliant to the established requirements.
The ambitions about the process are slightly less addressed, especially P2 -'Align the process at EU level' -is almost not addressed at all in the current state of the art.It is probably because of the differences in the countries' legislative documents regarding building permitting.P1 -'Simplify the building permit process as much as possible' -is instead addressed again mainly by Step 4 and Step 7.
There are some cases in which the ambition E1 -'Mindset change of public officers' -is addressed, but still in weaker terms than for the others.
Another interesting finding, shown in Fig. 22 is how each step often addresses ambitions belonging to the different EUnet4DBP pillars process, rules and requirements and technology.It makes it clear how the three aspects need to be considered in all the parts of the workflow.
The most addressed requirements (Fig. 23) are r4 -'Machine readable building codes', especially in association to Steps 1 and 7, followed by r11 -'Interoperability and APIs', mainly associated to Step 7.Then, r10 -'Modelling conventions and guidelines' comes, which is especially associated to Step 4 (BIM modelling).However, both r10 and r11 are quite addressed within all the steps.
As for ambitions, it is interesting to see how the colour in the heatmap (Fig. 23) is quite spread throughout the table, showing how many requirements are related to many of the steps and how it is necessary to consider many aspects for achieving successful results.

Developments gaps and direction for DBP
This was the first time that such a critical literature review on the DBP was conducted, by classifying the reviewed contributions according to specific criteria and a pre-defined coding schema (DBP process step, kind of contribution, level of progress, addressed EUnet4DBP ambitions and related requirements) allowing a more accurate interpretation of the state of the art.Moreover, the focus on each step of the reference workflow can clearly point out what the possible unbalances and gaps are with respect to their investigation and provision of specific solutions.In fact, a step which is not suitably addressed could mean that the workflow would have issues at that point instead of getting smoothly to the end.
First evidence demonstrated by this paper is the distribution of the efforts along the DBP workflow.While the regulations checking (i.e., Step 7) is the most investigated topic and the DBP process step which can reach a higher maturity, we can see that the major gaps are currently related to the use of geoinformation (Step 2), especially in association with the BIM and reciprocal conversions (Step 6).Some further GeoBIM studies are being developed in the last years, but not directly related to DBP.The Pre-consultation phase (Step 3) is also addressed in few studies at the moment, probably because requiring different issues to be solved preliminarily, such as interpretation and digitalization of regulations and the setting of a proper platform including and based on geospatial data.Moreover, it could be seen as low priority, at the moment, for enabling the DBPs, and could still provisionally remain partially manual and human-based.A similar reasoning could be done for Step 8. Step 4 and Step 5, about the modelling and export of the BIM and its validation is also still little investigated and do not often reach a high level of maturity, which is instead an essential enabling condition for DBP and even for a successful application of Step 7. To summarize, admitting Step 3 and Step 8 as of secondary importance, the priority in research and development should be given, at the moment, to the data and interoperability issues involved in the Steps 2, 4, 5, 6.
Second, looking at the addressed EUnet4DBP ambitions, we see that many are still neglected, in particular: T1 -Interoperable scalable systems useful at different levels (European, national, municipality) and in different countries; T3 -Platforms allowing a data-centric approach across the whole facility life-cycle by means of a central management of accesses; P2 -Align the process at EU level; E1 -Mindset change of public officers.Moreover, we could notice, from processing this review, that one further ambition should be added to the list, namely: X1 -Common understanding of DBP-related concepts.Being a very multidisciplinary topic, the diverse stakeholders and researchers, from different fields, should have the same understanding of the useful terms and concepts related to the building permit digitalization.
In addition, many requirements are not yet at the centre of current investigations in literature either, especially: r1 -Digitalize the mindset of public officers; r7 -Alignment across Europe and beyond, in Scope and Ambition (having connection with ambitions T1 and P2); r9 -Common Dictionaries.Some others are more addressed, although still marginal, i.e.: r2 -A roadmap and a change framework towards a fully digital building permit process; r3 -Normative text should be interpretable; r6 -Understanding the necessary process steps.Furthermore, r5 -Several kinds of data can be involved in the process (GIS, BIM, etc.), with related clear information requirementsis becoming more interesting for the DBP community in the last years, but still do not reach the amount of investigations reserved to r4, r10, and r11.
Finally, we could see that in general the reviewed contributions seldom arrive at a higher implementation level than B1 (demonstrators and proof-of-concepts) and most of the times have a low from intermediate progress (Executing or Validating).Moreover, also fundamental basic research is often missing: while we have a lot of applied scientific approaches, we see that there are low contributions to A1.

Conclusion
In this paper we have investigated the current state of the art related to the DBP use case, starting from a literature review the items of which have been classified by a multidisciplinary team according to the level of implementation, level of progress, addressed ambitions as defined by the EUnet4DBP and related requirements.Moreover, the items were grouped according to the specific steps of a reference workflow in which they could bring relevant contribution.In this way, we could The e-permitting framework consisting of code checking, submission, pre-checking, and automated rule-making modules is developed based on the Korea BIM (KBIM), and the developed system is tested using an office building in South Korea.The Netherlands Zhou and El-Gohary [113] (2019) The BIM of an educational building was checked for compliance with three energy codes r3, r10 R2 A2 Executing USA Song et al. [114] (2020) Proposed an approach that contributes to broadening the cope of BIM-enabled rule checking to any natural language-based design requirements r4 R2 A2 Executing South Korea Zhang and El-Gohary [117] (2015) Propose a BIM IE method to automatically extract project information from IFC-based BIMs and transform it into a logic format (logic facts) that is ready to be automatically checked against logic-represented regulatory rules (logic rules).outline with high detail how much each aspect of the very complex topic of digitalization of building permitting process has been currently tackled.
The investigation pointed out how major efforts are currently done for the regulations digitalization and the technological aspects, mainly for automating the compliance checks to regulations.
All other important subjects are instead still behind, including the mindset change of public officers; scalability of the solutions (process and technology); interoperability-related topics such as the IFC data validation and insufficient efficiency of the systems that jointly exploit BIM and geospatial domains; development of platforms allowing the management of many involved processing in a unique environment.
It is important to note that the digitalization of building permit issuing is a complex task because it affects a broad range of sectors.A successful and efficient transition can be put into practice if a great number of sectors participate in this transition and evolve based on the needs with respect to prospective procedural and legislative changes.Interoperability between different organizations is of significance in terms of data exchange regarding DBP.Integrating the standards into processes is considered the most efficient way.At this point, it can be highlighted that the countries can use these standards as a basis to create national standards that meet countries' specific requirements within DBP.
This study is mainly based on literature and, although integrated on the base of the knowledge of the multidisciplinary authors and a questionnaire to externals, there could be initiatives not reported in literature which are however valuable for some of the investigated aspects.For example, the software which are already on the market offer valuable solutions for regulation checking, at a quite advanced progress, although limitations of software are often with the alignment to the specific checking needs in practice and with the other steps.Moreover, even if regulations checking tools are available, a major issue is the need to provide valid and suitable IFC models as input, or GeoBIM models, which are not ready yet.
The classification was difficult in some cases, and there was the risk that either the categories could be ambiguous, or the authors could be biased with respect to some of them.To limit this we revised the classifications and discussed in several meetings in order to agree on a common understanding, and decided to make the classification of each item by at least two people, having different background.In this way we could limit the differences in interpretation.
This work was very relevant to analyse the state of the art in the DBP use case because only by decomposing this complex topic in its parts it is possible to understand the current progress and proceed further to fill the gaps with a specific focus.
Future work could integrate the review with new contributions, and especially it would be interesting to repeat the investigation after some years to include the many developments which are currently ongoing.Also other documents or experiences than the ones documented in literature should be integrated with a more systematic work concerning them, including the documents available in national languages.Besides this, future steps should be directed at filling the gaps pointed out by this study and reach higher progress and implementation in all the parts of DBP use case.

Fig. 2 .
Fig. 2. Schematic overview of considered workflow steps assigned to involved parties.

F. Noardo et al.Fig. 4 .
Fig. 4. Schema and relationships among the classifications used in the paper.

F
.Noardo et al.

Fig. 6 .
Fig. 6.Heatmap showing the evolution of implementation and progress through the years for Step 1.

Fig. 7 .
Fig. 7. Heatmap showing the evolution of implementation and progress through the years for Step 2.

Fig. 8 .
Fig. 8. Heatmap showing the evolution of implementation and progress through the years for Step 3.

Fig. 9 .
Fig. 9. Heatmap showing the evolution of implementation and progress through the years for Step 4.

F
.Noardo et al.

F
.Noardo et al.

Fig. 10 .
Fig. 10.Heatmap showing the evolution of implementation and progress through the years for Step 5.

Fig. 11 .
Fig. 11.Heatmap showing the evolution of implementation and progress through the years for Step 6.
are based on SMC.All the other studies find their fondants in developing open-source and open-format solutions.In particular, almost all the research are based on the use of open standards, i.e., IFC and CityGML, for model representation.Focusing on the use of IFC, the data contained into the file have to be extracted in order to further being represented in a semantic-based logic and matched with regulatory information.To do that, most of

Fig. 12 .
Fig. 12. Heatmap showing the evolution of implementation and progress through the years for Step 7.

Fig. 13 .
Fig. 13.Heatmap showing the evolution of implementation and progress through the years for Step 8.

Fig. 14 .
Fig. 14.Histogram representing the contributions related to each step.

F
.Noardo et al.

Fig. 15 .
Fig. 15.Distribution per year of the efforts related to each step.

Fig. 16 .
Fig. 16.Map showing the provenance of the reviewed contributions.

F
.Noardo et al.

F
.Noardo et al.

Fig. 19 .
Fig. 19.Number of reviewed contributions per kind and step.

Fig. 20 .
Fig. 20.Distribution of reviewed contributions along the development phase.

F
.Noardo et al.

Fig. 21 .
Fig. 21.Number of reviewed contributions per each level of progress in each step.

Fig. 23 .
Fig. 23.Heatmap showing the more addressed requirements to reach the EUnet4DBP ambitions in each step.

F
.Noardo et al.

Table 1
Re-phrased ambitions of the EUnet4DBP.

Table A . 4
Classification of contributions related to the Step 1, i.e., rule interpretation and digitalization of city and building regulations.(AR: Addressed Requirement, AA: Addressed Ambition, KoC: Kind of Contribution).

Table A . 6
Classification of contributions related to the Step 3, i.e., 'Pre-consultation and preliminary analysis'.

Table A . 7
Classification of contributions related to the Step 4, i.e., BIM and export to IFC.

Table A . 8
Classification of contributions related to the Step 5, i.e., Application reviewed (formal requirements -data validation).

Table A . 9
Classification of contributions related to the Step 6, i.e., conversion to and integration with a 3D city model or geospatial data models.

Table A .
10Classification of contributions related to the Step 7, i.e., application reviewed (content -regulations checks).

Table A .
10 (continued).Proposes conceptual graph to represent the rule requirements such that they can be easily understood by rule experts and validates the proposed approach by using previously applied compliance checking with respect to visibility of patient rooms from nurse station.CYPEURBAN is a software developed to verify the compliance of BIM Models in IFC format against the Municipality Urban Plan.CYPEURBAN has been developed by CYPE with the support of the Association of Real Estate in Madrid (ASPRIMA).Development of a prototype to check if a building follows a few building permit regulations based on IFC-model and geodata (detailed development plan) imported to an FME environment where the checks are performed.Describes the development KBimLogic, a rule-based mechanism designed for the building permit related rules in Korea Building Act sentences.As a computer-readable definition of a rule, KBimCode has been developed to be executed in actual rule-checking software.Describes rule checking method, classification and its demonstration with actual requirement sentences from the Korea Building Permit as part of KBimLogic, a software that translates the Korea Building Permit requirement into computer-executable format.

Table A .
[148]ontinued).Description on how to use SMC to verify the compliance with the Building permit Code.MSc thesis proposing the checking of parking regulations by means of the 3D city model, in CityJSON.An improved and extended version of this study was published in[148].
a b Technical contribution (see Section 3).

Table A .
11 Classification of contributions related to the Step 8, i.e., completion of the work and building authorities notification.Development of a web-based platform to support the City of Sant Feliu in becoming a Smart City.Services under developments include also application for actively involving the citizen in the process of city planning.