Augmented Reality in Review Processes for Building Authorities: A Case Study in Vienna

Abstract

The digital transformation of the construction industry is still lagging due to its incomplete implementation throughout the entire building lifecycle. One stakeholder in particular has been largely overlooked thus far: public administration. This study explores the potential integration of augmented reality (AR) into the processes of building authorities, with a particular focus on the review part of the permissions process, taking the City of Vienna as an example. As part of the EU-funded BRISE-Vienna project, an AR platform was developed and tested and an AR application was designed to enhance the transparency, stakeholder communication, and efficiency throughout the process. This study compares the proposed AR-based review process with the traditional plan-based approach, assessing both hard and soft factors. To this end, the durations of the individual process steps were measured, with a particular focus on the time spent by the officers (as a hard factor). In addition, qualitative surveys were conducted to gather the subjective impressions of the test participants (as soft factors). The key findings were a reduction in the officers’ workloads and an improvement in spatial understanding. While the overall review time remained similar, the use of AR reduced officers’ workload by over 40%. Additionally, the test participants stated that AR improved their spatial understanding and alleviated the time pressure within the process. This case study demonstrates the potential of AR in the permissions process and could serve as a model for other cities and countries.

1. Introduction

The construction industry is one of the least digitalised industries at present, with productivity improvements lagging significantly behind other industries, such as manufacturing [1]. Progress in the digital transformation of the construction industry is still in its infancy, which is a key reason for the stagnating development of productivity in this industry. For example, analyses conducted over the past few decades in Germany and Austria have revealed productivity declines of 9.3% in Austria (1995–2018) [2] and 4.2% in Germany (1991–2021) [3]. In this context, switching to the Building Information Modelling (BIM) or openBIM way of working has the potential to increase productivity in the construction industry [4]. BIM serves as a foundational technology for Construction 4.0, enabling integration with advanced tools such as artificial intelligence (AI), 3D-printing, the Internet of Things (IoT), and augmented reality (AR), which can be used to streamline processes across the construction lifecycle. AR has a variety of potential applications, with digital building models being useful not only in the design phase but also throughout the entire lifecycle, as demonstrated by Bademosi and Issa in their study on the implementation of AR in construction [5]. Various research projects have already recognised the potential of AR in both the construction [6,7,8,9,10] and operation [11,12,13] phases. However, public administration is a stakeholder that has largely been ignored in the context of the construction industry’s digital transformation to date. Initial research projects have confirmed the suitability of AR tools as easily understandable modes of presentation [14,15].

In Vienna, the building authority is responsible for reviewing over 13,000 building applications each year [16]. The city can be considered a significant stakeholder, as it owns and operates about 220,000 apartments [17]. These processes are still largely analogue, and therefore, involve a high level of manual work, leading to longer processing times for building permits. The City of Vienna has therefore decided to further transform the digital building permit process; to this end, it launched the EU-funded research project “BRISE-Vienna”. As part of this digital transformation, new processes are being developed using BIM, AI, and AR technologies. The potential for partially automating checks using openBIM standards has already been demonstrated in [18], and in [19], the authors provided an overview of AR use-cases for city governments. The review process and construction negotiations are of particular relevance to the building authority of Vienna.

As in other European countries [20], the approval process in Vienna takes into account the interests of neighbours in addition to compliance with legal regulations (e.g., fire protection and zoning). For this purpose, neighbours can assess the documents during the review process and the subsequent building negotiations, providing these individuals the opportunity to familiarise themselves with the documents and to object to them. At present, this assessment is carried out on the basis of 2D plans; however, interpreting the information in a plan requires good spatial imagination and construction knowledge. For people outside the profession, attempting to understand the real 3D situation of the construction site using only 2D plans often leads to problems. This obstacle to understanding is often undesirable, leading people to describe the process as non-transparent. This barrier can be overcome through the use of three-dimensional visualisations instead of two-dimensional ones, the possibility for which is provided by augmented reality. In many sectors, such as medicine, construction, and education, augmented reality has already been used to make it easier to understand spatial problems, and AR has been the subject of much research [21,22,23,24,25,26]. Regarding the permission process, AR could facilitate communication between people inside and outside the profession; in this context, AR would be employed as an “educational tool”. For this reason, the authors examined the use of AR in education in more depth. The literature review revealed that AR has rarely been used in education to date [27,28,29]. Therefore, knowledge transfer is not a key focus of available AR applications, a key factor of which is the creation of suitable AR models with additional information. In order to reach a large audience, teaching materials must be made available quickly and without the need for programming skills. For this reason, a team led by the authors previously developed an AR platform, which was tested in trials [21]. Based on this AR platform, a specific AR application was developed explicitly for building authorities. This application not only allows for the visualisation of 3D models but also integrates document access and annotation features, enhancing the transparency and efficiency of the review process.

For this study, the authors developed an AR-based review process using this platform and app and then performed tests. These evaluations formed the basis for answering the following research questions: “Does AR positively support communication between stakeholders and authorities?” and “Does using AR reduce the burden on building authorities by speeding up processes?” Policymakers hope that better visualisation will lead to better communication with citizens. This is expected to create added value for stakeholders, speed up processes, and ultimately reduce the workload of building authority officers.

In contrast to previous research, this publication does not focus on AR-based participation in urban development projects but rather a process for AR-based (neighbour) review in the context of an openBIM building permission process. In addition, the necessary software was developed and evaluated as part of a small case study that included a quantitative assessment of the duration of the processes and a qualitative review of subjective perceptions.

Fauth et al. (2024) have shown the importance of public participation in permission processes across Europe [30]. This case study presents an evaluated process that can be adopted in other countries.

The remainder of this paper is structured as follows. Section 2 analyses the current process and, on this basis, the development of the AR-based process in the context of openBIM permission. Following the description of the test setup, Section 3 presents the results of the quantitative and qualitative analyses with regard to the duration of the process and the subjective impressions of the participants. Section 4 discusses the differences between AR- and plan-based processes, placing both in a broader context. Finally, Section 5 provides recommendations for possible further research.

2. Materials and Methods

This section describes the materials used and the methods applied for evaluating the AR-based openBIM approval process in the construction industry, particularly the comparison with the plan-based review process by the persons involved. The traditional, analogue 2D plan approach to the building approval process has evolved through the development of a digital, partially automated process based on openBIM and augmented reality (AR). The aims of this study are to analyse the influence of AR technology on the approval process and evaluate its efficiency, user-friendliness, and the quality of participation it enables.

The methodological approach is shown step-by-step in Figure 1. First, the traditional, plan-based review process is described to create a basis for comparison with the newly introduced AR-based method. Then, detailed descriptions of the AR platform, digital models, and specific functions developed for the review process are provided. Finally, the test setup is explained, including the selection of test participants, the construction projects used, and the process steps carried out.

This structured approach enables a comprehensive analysis that provides quantitative and qualitative data for evaluating the AR-based review process.

2.1. Traditional Review Process (Plan-Based)

To date, the compliance of projects in the building approval process has been checked in an entirely analogue manner in Vienna; in particular, the review is carried out manually based on 2D plans. In addition to the classic structural and legal check, the review also includes a social component as the Vienna Building Code defines so-called subjective public neighbouring rights.

These rights mainly concern compliance with building regulations, emissions, and immissions, and violations of these rights can be reported by authorised persons (e.g., neighbours). Although the building authority checks for compliance with these rights, the review by neighbours introduces a principle of dual control. In this way, the process becomes more transparent and neighbours are more involved. The current process of plan-based review has already been described in [31], and is divided into the following steps.

• The process begins with the notification of authorised persons for the review process. The responsible officer checks the ownership of the neighbouring properties, including objects. The contact details of these owners are then collected using the land and central registers. The next step is to determine the review deadlines and notify the persons involved personally. At present, information is usually sent by letter (analogue or digital), requiring confirmation of delivery.
• In the next step, the authorised persons can appear at the building authority’s office within the specified period of time. Once their identity has been verified, the review begins.
• During the review of documents and plans, authorised persons can inspect all documents relating to the current submission (e.g., submission plans, structural analyses, building physics reports, statements).
• The responsible officer begins with a general presentation of the building project, and relevant information is discussed based on the documents submitted for approval. The submission plans usually serve as the basis. The scope of the information presented is not generally regulated and is customised by the officers; for example, information from the zoning and development plan (e.g., building class, construction method, alignment lines, special building regulations), the planned number of flats or type of use, and general explanations for technical building terms may be clarified during the meeting.
• After the officer’s general presentation of the building project, the neighbours can ask questions about the documents/plans.
• Before the review period begins, the authority checks that the subjective public rights of neighbours are being observed. In the event of a breach, the officer requests the building applicant to revise the plans. Interested parties can observe their subjective public neighbouring rights during the review process and, if a person involved complains about a violation of the law in this step, they can object to the submission.
• The responsible officer documents the objections raised in order to verify whether a detailed check is required. Legally relevant objections are collected and prepared as part of the construction negotiation.
• Once the objections have been raised, the review ends and the person can leave the building authority’s office. It is still possible to object after leaving the building authority’s office, as long as the objection is issued within the review period.

Regarding the use of Construction 4.0 technologies, processes should not simply be digitised; they must be redesigned for efficient digital transformation. The next section therefore describes the implementation of AR in the context of a digital approval process.

2.2. The openBIM Approval Process as Basis for Construction 4.0 Applications

Although most phases within the building lifecycle have already been digitalised (e.g., design, construction, operation), thanks to the spread of the BIM methodology, the phase between design and construction, namely, the approval process, has not yet been taken into account. As part of the “BRISE-Vienna” research project, an openBIM-based process for semi-automated compliance checking was developed to close this gap within the lifecycle (shown in Figure 2). This process has already been published [32] and is summarised here in an abbreviated form.

The focus is on digital models in openBIM format (primarily IFC), rather than 2D plans and PDF documents. The openBIM approval process described by Urban et al. in [32] requires three IFC models:

• A building application model (BAM): This is created by the architect and maps the project at the LOI300 and LOG300 level of detail.
• A reference model (REM): This reflects the information in the zoning plan and the building regulations. The framework conditions of these documents are used to generate a digital permissible spatial envelope; all new buildings must be located within these digital, spatial envelopes.
• A service information model (SIM): This is created by the building authority and displays all alphanumeric information that cannot be mapped in the REM.

During the approval process, the building authority uses these three models for compliance checks. Not all provisions of the building regulations can currently be checked automatically; for example, although it is currently possible to check for compliance with escape routes [18], this is only a partially automated check due to the currently incomplete mapping of building regulations. At the end of the digital permission process, as in the analogue process, there is a review or building negotiation phase.

Previously, building application documents (e.g., building plans, proof of ownership, and expert reports) had to be submitted in triplicate in paper form. However, in 2020, Section 63a of the Building Regulations [33] introduced the option for digital submission. At present, this involves uploading the same documents in digital (PDF) form via a portal provided by the City of Vienna. The next step is to automate the permission process as much as possible using BIM. This has been developed and tested as part of the BRISE-Vienna research project, in which BIM-based permissions form the basis for integrating AR into review processes. Legally, however, this requires further revision of the Building Regulations, as “building plans” are currently explicitly required for review.

2.3. AR-Supported Review Process Based on a New AR Platform

Part of the digital transformation of the building permission process in Vienna involves the use of augmented reality with AR devices (e.g., smartphones or tablets), and the switch to AR requires a redesign of the plan-based review process. In [19,34], two options were considered: an on-site review process (using a mobile device) and a review process at the building authority office (using a device provided by the building authority). Both forms enable independent review through the digital retrieval of documents or models; however, the former case may create legal problems due to the use of private devices. In particular, it cannot be ruled out that the mobile device is passed on to unauthorised persons, as not everyone is authorised to check in Vienna. To avoid this problem, the process must take place at the building authorities office (i.e., the second option mentioned above).

The platform developed at TU Wien, called AR-supported Teaching [21], serves as the basis for this process. The platform consists of an editor (for desktop) and an AR viewer (i.e., an AR app for mobile devices). With the editor, 3D models can be enriched or customised with information and then viewed using the AR app. The basis for the AR model is the building application model (provided in IFC format). The app requires further development to include additional functions in the review context. The following functions have been implemented:

• Personalised login to the AR platform using a personal QR code;
• Digital viewing of project-relevant documents (e.g., survey plan, energy efficiency calculations) in PDF format;
• A teaching scene explaining relevant legal and construction terms;
• Digital submission of objections.

This platform was used as a basis for the AR application for the city of Vienna. To optimise the review process with the use of AR, the process has been redesigned and the authors have described the process in detail in [34]. Based on this redesign, a team of the Digital building process research unit at the TU Wien adapted the AR platform. The process and the app were continuously and iteratively optimised. The final AR-based review process comprises the following steps (see Figure 3) [35]:

• The authority checks the subjective public rights of neighbours in advance and, in the event of a breach, requests the building applicant to revise their plans. The review process begins with the notification of authorised persons, analogous to the plan-based process. Once the deadlines for review have been set, the persons involved are notified personally. In addition to the general information (e.g., deadlines, rights), the invitation to the AR-based review process contains a project- and person-specific QR code.
• In the next step, the authorised persons can appear at the building authority’s office within the specified period. In contrast to the plan-based review, an officer does not moderate the AR-based review process. For this reason, there is also no identity verification (which is guaranteed by the QR code).
• The largely independent implementation takes the time pressure off the participants. A poster with the process steps is provided to orient the individual during the process. People attending the review start with illustrated step-by-step instructions (shown in Figure 3, step 3).
• Depending on the individual’s age group or technical affinity, carrying out the review using a tablet requires a familiarisation period. The time required to use the tablet without reference to the review is summarised in the “Tablet essential functions” step.
• A person attending the review shows up at the building authority’s office with their personal summons. The QR code on the invitation is used to identify the person and assign them to the project. After calling up the AR app and scanning the QR code, the review begins.
• Technical terms are a frequent hurdle during reviews and construction negotiations. Officers need a lot of time to explain terms relating to building regulations or the zoning and development plan. For this reason, the AR-based review process begins with a so-called teaching scene. Using a test building project, relevant technical terms (e.g., alignment lines, building classes, types of zoning) are explained using text and images. This is intended to create a basis for efficiently discussing questions with the responsible officer later.
• The relevant building project is then called up. During the file review, authorised persons can digitally view all relevant documents in the app (e.g., submission plans, structural analysis, building physics, statements). The building project is then located on a table, with AR visualisation enabling a free spatial view of the project. This helps to avoid the misunderstanding of information presented on 2D plans.
• If questions arise during the review, the responsible officer can be called in. If the attending person wishes to report legal violations during this step, they can enter an objection digitally on the tablet.
• Legally relevant objections are collected and prepared for the construction negotiation phase.
• Once the objections have been raised, the review ends and the person can leave the building authority’s office. It is still possible to raise an objection after leaving the building authority’s office, as long as it is submitted within the review period.

Finally, tests were carried out to verify the redesigned process, as described in detail in the following section.

2.4. Experimental Setup

Tests were carried out to validate the newly developed AR-based review process. To this end, data were collected and compared for both the traditional plan-based review process and the proposed AR-based review process. The tests consisted of two parts: quantitative and qualitative analyses. As part of the tests, the role of neighbours was taken on by test participants and the role of the building authority’s consultants by the authors. The test setup is described below, including the test participants, used building projects, and process design.

2.4.1. Augmented Reality

The use of AR is an important part of the EU-funded BRISE-Vienna project. Augmented reality is a form of extended reality (XR). The different levels of XR have been defined by Milgram and Kishino [36], including virtual reality (VR), mixed reality (MR), augmented virtuality (AV), and augmented reality (AR). According to Azuma [37], AR has three characteristics: the combination of real and virtual environments, real-time interaction, and three-dimensional localisation. In particular, AR enhances the real environment with virtual elements. Basically, two approaches can be distinguished with regard to the application of AR: on one hand, AR can be implemented by projecting virtual elements into the field of vision (e.g., using a Microsoft HoloLens) or by recording the environment with a camera and combining it with a (virtual) display (e.g., tablets, Apple Visio Pro). In principle, there are many AR integration possibilities for public authorities [19]. The case study reported here focused on implementing AR using tablets, allowing a virtual building model to be placed in the room using a reference surface. Additional virtual volumes can be displayed to indicate permissible building areas. Further documents relating to the submission can be accessed in static PDF format. The AR app’s functionality could also be implemented using a 3D viewer on a tablet or computer. The complex implementation of the AR application in the course of our case study was intended to provide the authority with a proof-of-concept. In the medium-term, AR viewing is planned for implementation independently of the authority at the respective construction site. A detailed description of this AR variant can be found in Gerger et al. [19].

2.4.2. Test Participants

Following Vienna’s legal situation, “all parties involved in the proceedings” [33] are entitled to inspect the plans. This definition includes the developer, the owner or co-owner of the affected property, and neighbours (owners or co-owners of neighbouring properties). In exceptional cases, other parties may be included. Of the considered parties, neighbours are the most critical group. The Vienna building code defines neighbours as: “… those who have a common boundary with the property affected by the construction project or who are separated from this property up to a certain distance …”, where the distance is defined in more detail in the local building codes. Typically, the review process concerns a subset of the population, regardless of age and gender, without prior knowledge of civil engineering or architecture.

The tests were carried out with 12 participants. Tullis and Stetson [38] showed that in most cases (i.e., in 90% of cases), a sample size of 12–14 people is sufficient to achieve meaningful results. Nevertheless, it should be noted that the test group does not provide a representative cross-section of the Austrian population, and thus, further tests are required. The composition is shown in

Table 1. Composition of the test group.

Age group	≤40: 9	>40: 3
Gender	Male: 8	Female: 4
Construction knowledge	Yes: 6	No: 6
Experience with review processes	Yes: 0	No: 12

. The test group is overly representative of young people (up to 40 years of age, 75%), males (67%), and people with previous knowledge of building technology (50%), compared with the general population. In Austria as a whole, 19% of the population is under 20, 26% is between 20 and 40, and 55% is over 40. The general significance of the results is limited by the following bias: the high proportion of young, technology-aware individuals may lead to under-representation of problems in dealing with AR; furthermore, the high proportion of individuals from the construction industry may lead to over-representation of technical objections. None of the test participants had previously participated in a plan-based review process.

2.4.3. Building Projects

In the test process, real construction projects that had been previously checked by the building authority were used.

Table 2. Characteristics of the authorised building projects.

Project Name	Type	Building Project Type	Levels	Flats
Arndtstraße	Residential complex	A	7	26
Hochwaldweg	Residential complex	B	5	7

compares some characteristic values of the building projects used as a basis [35]. Pictures of the used BIM models are shown in Figure 4.

2.4.4. Process Design

The tests consisted of two stages: a quantitative survey consisting of the AR-based review (stage 1a) and the plan-based review (stage 1b); and responses to the questionnaires (qualitative survey, stage 2).

To ensure the validity of the data, two independent parameters (building project, review type)—each with two possible values—were varied during the tests (see Figure 5). This resulted in $2^2=4$ test setups. The varied parameters included:

• Variation of the review type (AR- or plan-based), setting a reference for the second test to be compared with;
• Variation of the construction projects, preventing falsification of the process durations due to prior knowledge.

Only the visualisation potential (2D plans compared with 3D representation with AR) within a given setup was not directly comparable, due to the use of different construction projects. The following section explains the individual process steps in detail. Several framework conditions were defined:

• The test participants acted as representatives of involved parties or neighbours;
• The authors acted as the building authority in the tests and were trained by the authority’s officers;
• Rooms at the TU Vienna served as the test location, instead of the building authorities office.

2.5. Quantitative Evaluation

In the quantitative evaluation, the process duration of the AR-based review was compared with that of the plan-based review; for this purpose, the authors measured the elapsed time for each process step. Figure 6 compares the evaluated processes for the plan- and AR-based reviews.

The dotted box indicates those steps that are directly part of the review, and therefore were considered as part of the tests; the person symbols indicate the steps that cannot be performed without an officer. As the processes are different, their individual steps are not directly comparable. For this reason, steps with the same content were clustered into four common process stages: process, presentation of the construction project/definitions, document review, and objections. In the test, the durations of the individual process steps were measured, then assigned to the appropriate cluster. It is clear that the process-oriented steps prior to the actual presentation of the construction project differ between the plan- and AR-based reviews; these individual steps have, therefore, been clustered into the “Process” category. While the plan-based review begins with authentication and the provision of documents, the AR-based review primarily explains the process and how to use the tablet. The initial information required in the traditional plan-based review process has already been described in detail in [31]. The data collected during the tests included the processes and process durations, which were determined based on expert surveys and further validated based on the tests carried out.

2.6. Qualitative Evaluation

The qualitative analysis was performed to evaluate the subjective impressions of the test participants during the tests. This evaluation aimed to comparatively assess the processes of AR- and plan-based review, based on two questionnaires (completed after each trial, respectively). The questionnaires consisted of two parts: the first part consisted of socio-demographic questions with the aim of assessing the test group’s composition (regarding gender, age, knowledge of civil engineering or architecture, and prior knowledge of the review process); while, in the second part, single-choice procedural questions were asked (e.g., about legal knowledge or changes resulting from the use of AR). These questions were answered using a five-point de-verbalised response scale (Likert scale; 1 = “fully applicable”, 5 = “not applicable”; with the optimum being 1 for positive and 5 for negative statements [39]). Finally, open-ended questions about the general process and any suggestions for improvement were included. The order of the test type (AR-based, plan-based) was varied between the test participants in order to obtain more valid results. The first questionnaire was answered after the first trial, and served as a reference regardless of the trial type. The subsequent second test allowed the test participants to make a direct comparison. For this reason, the questions in the second questionnaire were worded comparatively, such that the answers were always framed in comparison to the first test (e.g., “I could visualise the building project better”).

3. Results

This section presents the study’s results, comparing the traditional plan-based and novel AR-based review processes. It focuses on quantitative and qualitative assessments, which were performed to evaluate the efficiency, user experience, and overall impact of AR technology on the permission process.

First, the process durations of the two procedures were analysed, highlighting the differences in the individual process steps and the variations in processing times. Subsequently, the subjective perceptions of the test participants were analysed based on their questionnaire responses, focusing on factors such as time pressure, spatial understanding, and user experience. In addition, the impacts of the use of AR technology on the submission of objections and the overall transparency of the process were analysed. The results provide valuable insights for optimising the building permission process using digital technologies.

3.1. Quantitative Evaluation

Figure 7 and Figure 8 show the recorded process durations. The durations in minutes per process step are shown graphically on the left (expressing the arithmetic mean values (aM)), while the median (Md) and standard deviation (SD) are plotted on the right. The colour of each process step refers to those used in Figure 6.

The evaluation of the plan-based review in Figure 7 shows the distribution of the durations of its process steps. By far, the most time-consuming step was “review of documents” at 11:42 min, followed by general explanations by the consultant (02:44 min) and the raising of objections (02:35 min). For some of these values (e.g., “presentation of the project”), the standard deviation was relatively high when compared with the median and mean values, representing the high dispersion of the measured data.

The highly individualised test setup can explain this dispersion: due to differences in the prior knowledge, interests, and spatial imagination of the test participants, their explanations do not follow a standardised scheme and, instead, are very individual. This ultimately results in highly scattered measurement data.

Figure 8 shows the measured durations for the AR-based review process steps. The use of AR requires additional steps at the beginning (“Reading the poster”, “Tablet handling”, and “QR code scan”); in particular, the AR-based review process itself and the handling of the tablet in general must be explained before the review process can start. The “review of documents” also took the longest (at 06:53 min) in the AR-based review, followed by the explanation of definitions via a “teaching scene” (04:48 min) and “submission of objections” (02:29 min). This distribution of process durations reflects the authors’ expectations. One aim of implementing AR-based review is to reduce the time spent by the officers and, for this reason, most of the explanations of terms and general explanations were integrated into a poster and a teaching scene. This approach was intended to reduce the time that building authority officers spend explaining terms, in order to have more time for their original work, namely, compliance checking of the documents. Referring to Figure 6, in which the steps which need to be guided by an officer are indicated by person-signs, it can be confirmed that the new process not only reduces the total amount of steps which need to be guided, but also the time required for these steps. This change was confirmed by the shorter duration of the “review of documents” step in the AR-based review process (06:53 min), compared with that in the plan-based review (11:42 min). However, it should be noted that the standard deviation reached high values in some cases; for example, reviewing documents is a highly individualised process, with the fastest person completing this step in 04:28 min, while the slowest subject took 21:45 min. Some test participants ignored the poster with the step-by-step instructions and tried to go through the process unguided; therefore, for the “reading the poster” step, times ranging between 00:00 and 03:45 min were observed.

3.2. Qualitative Evaluation

Figure 9 summarises participants’ subjective evaluations of the review processes. The questionnaires addressed aspects such as time pressure, transparency, medium usability (plans vs. tablets), and procedural understanding (e.g., spatial comprehension, legal rights, and objections). The questions are shown on the left, the response results are presented in graphical form in the centre (expressed as the arithmetic mean (aM)), and the median (Md) and standard deviation (SD) are shown on the right. The questions were answered using a five-point de-verbalised Likert scale (1 = “fully applicable”; 5 = “not applicable”). In particular, Figure 9 shows the results of the AR-based review compared with the traditional plan-based review. The results were directly transferred to the overall evaluation when AR was used in the second test (see also Figure 5, Setup 3 and 4). In the case of the AR-based followed by plan-based test sequence, the results were transferred in an inverted manner to the overall evaluation (5 = “fully applicable”; 1 = “not applicable”; see also Figure 5, Setups 1 and 2).

As the plan-based review process needs to be changed to integrate AR tools, mental and physical effects were also surveyed. While using a tablet for viewing was moderately challenging for most respondents (aM 2.50 and Md 1.50), the strain of constantly holding a tablet was perceived as more strenuous (aM 3.92 and Md 4.50). The significant standard deviation (of 1.71) in this context results from the differentiated consideration of the age groups. People aged 40 and younger noted the tablet’s ease of use, while it posed challenges for those aged over 40. This issue can be addressed in two ways: the first solution could be a digitalised process without the use of AR. AR visualisation could be optional for viewing at the building authority’s office, as no reference to existing buildings is required; in this case, a 3D representation of the planned building project on a computer in the context of its neighbouring buildings may be sufficient, considering that using a mouse and keyboard is generally more intuitive for all age groups. There is also the added value of a simple spatial representation, compared with the language of plans. Another solution could be the use of head-mounted displays (HMDs; e.g., glasses), which allow the user’s hands to remain free and thus offer certain advantages, especially on construction sites [7,40]. However, due to their weight, they can be uncomfortable to wear for long periods; further technical development and commercialisation could make the use of AR glasses more attractive in this context.

Furthermore, the evaluation showed a slight improvement in understanding subjective public neighbouring rights. In the AR-based review, the test participants had to first complete a training scene with relevant legal and technical construction terms before viewing the project. The test participants rated their understanding of neighbouring rights in the AR-based review slightly better (aM 2.33, Md 2.00). Regarding the submission of objections related to subjective public neighbouring rights, no significant changes were observed in this area: the ease of digital submission of objections was rated similarly to the conventional form (aM 3.08, Md 3.00).

Both sides, that is, the participants and the officers, should benefit from the use of AR during the review process. When developing the AR-based review process and the AR app, the authors expected the following advantages in particular:

• Less time pressure for participants;
• Relief of the officers through more independent review by the participants;
• Better spatial understanding of construction projects through visualisation with AR.

The evaluation confirmed that the first objective had been achieved. The test participants stated that they were under less time pressure during the AR-based review, with an arithmetic mean of 2.33. Furthermore, the second objective—to reduce the time needed for explanations by officers—was also achieved. The main reason for both these achievements is the increased possibility of going through the review process independently, without time pressure due to the presence of the officers. While an officer is present from the beginning of the plan-based review and guides the participants through the process, a poster explains the individual steps in the AR-based review process; hence, the participants could take as much time as they want. However, the AR process itself was rated with an aM of 3.58 (Md 3.50), which is slightly worse than that of the traditional approach including the officer’s explanations. One factor affecting this result is the often insufficient attention paid to the step-by-step explanation; in particular, the quantitative evaluation showed that some people did not notice the poster or pay any attention to it (00:00 min). As a result, the test participants had avoidable questions about the procedure. In future, participants will be asked to confirm that they have read and understood the explanation poster. Finally, the tests also confirmed achievement of the third objective: a better spatial understanding of the construction projects. The evaluation showed the best results for this question; with an average of 1.50 and a median of 1.00, the use of AR showed the most significant potential in this regard.

In the open-ended questions, respondents highlighted a better ability to understand the AR visualisation. The paperless process using AR was also mentioned as a positive aspect. The negative feedback mainly concerned the incomplete information in the AR display and bugs affecting the AR app. Submission plans require linear dimensioning information. On one hand, this allows distances to be read directly and the relevant rules to be checked immediately while, on the other hand, linear dimensioning information allow areas to be calculated and thus provide a basis for checking rules regarding, for example, built-up areas. At the time of the performed tests, the AR app offered some ability to check distances and areas using maximum volumes, to some extent; however, the test participants had to rely on its accuracy and could not carry out an independent check (e.g., using coordinates).

Two problems were noted to have occurred: No reference surface was detected, such that the AR model could not be located; and the application occasionally crashed. The referencing problem did not occur during the initial localisation but happened during repeated attempts and could be resolved with a software update. The application crashes occurred after incorrect use of the AR application (e.g., attempting to zoom after fixing the AR model). The overall stability of the app was significantly improved using the feedback obtained through the case study.

The participants suggested the following improvements:

• Expand the teaching scene to include more legal and technical definitions;
• Enhance the AR app’s stability to address its occasional crashes and localisation issues;
• Include listings for submission plans to enable direct distance measurements and rule validation.

Despite these challenges, participants regarded AR as a valuable enhancement to the plan-based process, particularly due to its visualisation capabilities. The first two suggested improvements have already been implemented through software updates.

4. Discussion

This section summarises the questions and findings of the previous sections, with the resulting statements forming the basis for answering the posed research questions. Building on the questions addressed in this study, it is differentiated from previous research. Furthermore, some focal points which should serve as starting points for further research are briefly outlined.

4.1. Quantitative Evaluation

The quantitative evaluation involved comparing the process durations according to process clusters in plan- and AR-based review approaches (cf. Figure 6). Figure 10 shows the durations for each process and the shift in process durations. Although the use of AR led to a reduction in the time required to “review the submission documents”, it led to increases in the duration of upstream processes. However, the shortened time for the “review of documents” stage is significant, as it is in this stage that the person determines whether subjective public neighbouring rights are violated.

These rights relate, among other things, to exceptions regarding crossing alignment lines. These alignment lines mark the boundaries of the buildable area in the development plan for Vienna. These limits may only be exceeded in exceptional cases; for example, by bay windows or stairwells over a maximum of 1/3 of the building frontage by 1.50 m or by balconies over a maximum of half of the building frontage by 2.50 m.

This information can be easily labelled and measured in scale plans. The AR-based visualisation provided the same level of information, with volumes in the AR model representing the permissible areas for protruding building parts (Figure 11). Although this allows for easy visual verification, the scaleless representation in AR does not allow the model geometry to be checked via measurement.

The use of 3D visualised annotations (e.g., linear dimensioning information) could solve this problem in the future. The combined visualisation of the zoning plan model and the building application model is expected to offer the advantage of more straightforward comparability. At present, the zoning plan (permissible) and the submission plan (planned) can only be viewed side-by-side.

The lower information content, therefore, also affected the process duration. As noted above, the shortening of this process step was accompanied by an increase in upstream processes, with the switch to a non-moderated review leading to longer process durations. The test participants needed more time to find their way through the new process independently, as reading and understanding the step-by-step instructions required more time than the moderated guidance provided by the instructor. Handling the tablet and the AR app also required additional time compared with the plan-based review. The AR-based process also led to a longer duration for the “presentation of the construction project/definitions”. The introduction of a teaching scene for relevant legal and construction terms explains this increase. In contrast to a conventional plan-based review, the terms are not selected on a procedural basis but instead cover a broader, general context, resulting in a longer process. By independently working out the meanings of relevant building terms, the user can review the terms without time pressure; notably, this change was perceived positively by the participants in the qualitative evaluation. While the teaching scene led to a greater understanding of structural engineering topics among the test participants, it also created the basis for a more efficient discussion with the officers and longer process durations. One possible solution is to make the teaching scene publicly accessible, as the general, project-independent design of the teaching scene makes it possible to view it from home on one’s own device. In contrast to building application models, the teaching scene could be made available to everyone such that those involved could view the explanations at any time, regardless of their location and process. Those involved can obtain information in advance at home, further reducing the time spent at the building authority’s office.

In addition to the semi-automated review of building applications using openBIM, the review and building negotiation processes can also be made more efficient using AR. The evaluation of the openBIM process has already confirmed the potential for speeding up the overall process. Two indicators of particular interest for evaluating AR-based review processes in this context are the overall process duration and the time required by the building authority officers. The time the officer does not need to spend on explanations, as in the plan-based review process, can be used to check building applications, thus speeding up the building permission process. Figure 12 compares the total process duration and the groups of people involved in plan- and AR-based reviews. The total process durations determined during the tests did not differ significantly between the two forms of reviews (plan-based 19:11 min to AR-based 19:49 min). The longer overall process duration in the AR review is due to the longer duration of the process steps and the more extensive definitions. The use of AR nevertheless led to an improvement for the building authority, due to the division of the process into steps that the participants can carry out independently. Notably, the time required by the consultant was reduced by 45% to 10:35 min (shown in dark). Even if the AR model could be thoroughly tested, no significant shifts would be expected here, as this is time primarily independently spent by participants. This potential time-saving is primarily linked to a good understanding of tablet handling, and requires a process that is characterised by the abilities of the participants. If support is required from the outset, explaining or learning how to use a tablet could quickly offset the advantages of a more understandable visualisation.

4.2. Qualitative Evaluation

The qualitative evaluation results were presented in the Qualitative evaluation section. However, the composition of the test group may have led to a distorted picture, due to the disproportionately high number of participants with previous construction knowledge. Typically, the building authority’s plan-based review process (for neighbours) is almost exclusively attended by people without civil engineering or architectural knowledge. For this reason, Figure 13 divides the results according to the participants with and without prior construction knowledge.

It can be seen that there were significant differences in the participants’ assessment of mental and physical strain. People with a construction background are used to the “language of plans” and often have a good spatial imagination. For these reasons, the improvement in mental stress was only marginally improved with AR (aM 2.83, Md 2.50). For people without previous construction knowledge, the “language of plans” is often more challenging to understand, and these participants rated the use of AR via tablet as more mentally relieving (aM 2.17, Md 1.50) but more physically demanding (aM 3.33, Md 4.00). Both groups rated the support in visualising the construction project similarly (aM 1.67, Md 1.33). People with no previous construction experience rated the added value of AR slightly better. There were no significant differences in their understanding of the law and the possibility of raising objections.

In a final question, the respondents were allowed to choose between the forms of review processes: “Which method would you choose in future?” Figure 14 shows the results regarding this question, which also present apparent differences depending on previous construction knowledge. People with previous construction knowledge rated the lower information content of AR more negatively than the comparison group. For this reason, two-thirds of those with prior knowledge would continue to opt for conventional plan-based review process. Overall, AR was seen as having more of a supplementary role with respect to the plans. In contrast, the decision was 50:50 for those with no previous construction experience.

5. Conclusions

The progressive penetration of the construction industry with BIM methodologies serves as a starting point for Construction 4.0 technologies. Although the use of mixed realities in the design, construction, and maintenance phases has already been investigated in numerous projects [6,7,8,10], another suitable phase for the use of AR is the building permission process. AR has high potential as a universal visualisation method, especially in the building permission process, facilitating the communication of complex information to citizens. The efficacy of communication through plans in the construction industry has been proven over the decades; however, understanding plans requires technical knowledge and an excellent spatial imagination. Therefore, certain limits may be reached when attempting to communicate with citizens. Legal frameworks vary from city to city and country to country [30]. Nevertheless, the basic findings described here can be transferred to other cities and serve as a basis for the development of specific AR apps. This study focused on the restrictions of Austria and the City of Vienna. In particular, an AR-based review process for building authorities previously developed by the authors [19,34] was tested using an AR platform also developed by the authors. In addition to the AR-based review process with AR devices (tablets) serving as hardware, tests were carried out using the traditional plan-based review process for comparison. On one hand, the tests confirmed the expected advantages of AR visualisation when compared with 2D plans; on the other hand, the tests also revealed the differentiated use of tablets across age groups, with AR and tablets considered as an additional hurdle for older individuals. Using touch gestures and constantly holding the tablet are often unfamiliar and may be perceived as annoying. Therefore, special consideration should be given to usability for older people when developing AR applications. Derby et al. [41] have described how customised software development can help in this regard, where one possibility involves the use of HMDs (e.g., Apple Vision Pro, Microsoft HoloLens 2), which keep the user’s hands free. Controlling AR apps with gestures, eye-tracking, and voice commands could be perceived as more intuitive. Further tests are required to obtain useful information regarding this issue.

The test subjects made three suggestions for improving the AR app: eliminating bugs, expanding the teaching scene, and integrating measurement chains. The first two have already been implemented in the software, and work is currently underway regarding the last suggestion; in particular, the possibilities for integrating measurement chains within the IFC structure are being examined. Once this suggestion has been incorporated, the biggest disadvantage for construction professionals will have been eliminated, which is expected to further increase the approval rate of the AR process (see Figure 14).

In addition to the qualitative evaluation, an evaluation based on a quantitative assessment was also performed. The quantitative evaluation revealed that, even though the total review time was similar, the officers’ workload was reduced. Notably, the total in-office review time could be further reduced by allowing individuals to prepare for the review from home; for example, providing access to the project-independent teaching scene at home and as often as desired. The developed teaching scene is based on a general building model, making it easy to access this scene anywhere (e.g., at home), as there are no legal reasons preventing its general publication. The required terms can thus be studied without time pressure, and it can be assumed that individuals will be better prepared when they come to view the material. This would make the process even more convenient for neighbours and efficient for authorities. Employees are currently dealing with a high number of applications: 13,000 per year, to be precise. The time that is no longer needed for information services (neighbour reviews) can be used for process compliance checking, which is expected to speed up the overall permission process.

This case study demonstrated how AR can be used as a tool to strengthen citizen participation while reducing administrative burden. In particular, using AR and splitting the review process into independent and moderated processes reduced the time spent by officers by over 40%; this result should facilitate and encourage building authorities in different countries in their decision to integrate such solutions.