buildings

Article
Evaluating Modular House Construction Projects: A Delphi
Method Enhanced by Conversational AI
Augustinas Maceika 1,† , Andrej Bugajev 2, *,† and Olga R. Šostak 2,†

1 The Faculty of Mechanics, Vilnius Gediminas Technical University, Sauletekio Ave. 11,
LT-10223 Vilnius, Lithuania; [email protected]
2 The Faculty of Fundamental Sciences, Vilnius Gediminas Technical University, Sauletekio Ave. 11,
LT-10223 Vilnius, Lithuania; [email protected]
* Correspondence: [email protected]
† These authors contributed equally to this work.

Abstract: This study focuses on evaluating modular house construction projects, which is a critical
segment within sustainable building practices. Despite the significant advantages of modular con-
struction, such as enhanced resource efficiency and reduced environmental impact, existing research
often overlooks its unique attributes and constraints. Our objectives were to identify crucial parame-
ters for a comprehensive evaluation of modular construction, particularly emphasizing sustainability,
and to explore how an advanced conversational AI tool, ChatGPT, can assist in modular building
assessments. We employed the Delphi method to define these parameters and integrated ChatGPT to
develop a robust assessment methodology. This approach allowed us to harness AI-driven insights
to enrich the evaluation process. Our findings suggest that ChatGPT delivers high-quality results
comparable to those produced by experts in modular building assessments. ChatGPT formulated a
detailed description of the evaluation scale for each criterion, effectively outlining the guidelines for
evaluating modular house projects. To illustrate the effectiveness of our proposed methodology, we
applied it to a real-world modular house project in Lithuania, demonstrating how this approach can
significantly contribute to advancing sustainable construction practices.

Keywords: modular building; project evaluation; ChatGPT; conversational AI; generative AI; artificial
intelligence; Delphi method; MCDM; sustainability

Citation: Maceika, A.; Bugajev, A.;

Šostak, O.R. Evaluating Modular
House Construction Projects: A 1. Introduction
Delphi Method Enhanced by
The modular building [1] has become an increasingly popular construction method
Conversational AI. Buildings 2024, 14,
due to its numerous advantages such as cost savings, speed of construction, and flexibility
1696. https://doi.org/10.3390/
buildings14061696
in design. A modular building offers cost savings by reducing material waste, benefiting
from economies of scale, and optimizing labor, and its speed of construction stems from
Received: 10 April 2024 simultaneous processes and predictable timelines [2]. Flexibility in design comes from
Revised: 18 May 2024 customization, adaptability, and stringent quality control in a factory setting [3]. In the
Accepted: 4 June 2024
context of this research, we will refer to a modular building as a structure composed
Published: 6 June 2024
of standardized modules fabricated off-site in a factory or manufacturing facility and
assembled on-site. These modules, made from materials like metal, wood, or plastic, include
complete sections of a building, such as rooms with integrated floors, roofs, and walls.
Copyright: © 2024 by the authors.
Modular buildings provide quick and efficient assembly or disassembly, flexibility, and high-
Licensee MDPI, Basel, Switzerland. quality control. They can be used for both temporary and permanent structures, including
This article is an open access article residential housing, commercial buildings, mobile offices, and temporary housing. Modular
distributed under the terms and construction (MC) has been identified as a possible remedy that significantly enhances
conditions of the Creative Commons productivity, accelerates building timelines, and diminishes the substantial ecological
Attribution (CC BY) license (https:// footprint of the construction industry. Despite being embraced in several countries across
creativecommons.org/licenses/by/ the globe, its widespread use is still hindered by numerous interconnected obstacles. For
4.0/). instance, in Germany—a country with a strong off-site construction (OSC) market—MC

Buildings 2024, 14, 1696. https://doi.org/10.3390/buildings14061696 https://www.mdpi.com/journal/buildings

Buildings 2024, 14, 1696 2 of 36

is still not widely adopted by housing corporations. This is unexpected, especially given
its crucial role in rapidly providing housing in areas experiencing severe shortages, and
according to the research [4], results show that barriers concerning the market (supply
and demand), the industry’s attitude, and the process (B8) can be regarded as highly
relevant factors.
Exploring the customer decision process for modular building projects, especially
in the residential sector, reveals a complex network of decision-making factors. These
factors, ranging from design aesthetics and functionality to cost-effectiveness, sustainability,
construction timelines, and customization options, are pivotal in understanding customer
preferences in modular housing projects (for illustration, see Figure 1). Prior studies have
not fully captured or understood these intricacies.

Figure 1. Modular building example (generated by ChatGPT 4.0 [5]).

The significance of these factors varies for each customer, influenced by their individual
needs, preferences, and values. This variation highlights the diverse considerations in
modular housing selection, including sustainability. Sustainability, while one of many
factors, plays an increasingly important role in customer choices, reflecting a broader
societal shift toward environmentally conscious living. In this context, ’weights’ refer to the
importance or priority of each factor in decision-making process. For some, sustainability
and energy efficiency may be paramount, aligning with their commitment to eco-friendly
living. Others might prioritize different aspects such as budget or design.
Understanding these varying weights is crucial for professionals in the modular
housing industry, including developers, architects, and marketers. It enables them to tailor
their offerings to meet the diverse preferences of customers, while also considering the
environmental impacts of their projects. This approach helps bridge the gap between
individual customer needs and the collective goal of sustainable development.
As a key component of evaluation methodology, an evaluation scale must be estab-
lished to provide a standardized method of evaluating and comparing different residential
modular building projects. Such a scale enables customers to make informed decisions by
evaluating various projects against uniform criteria. This is particularly significant given
that traditional methods like the Analytic Hierarchy Process (AHP) and Data Envelopment
Analysis (DEA) can achieve better performances due to a less biased evaluation—some
criteria undergo direct evaluation using instructions accompanied with an evaluation scale.
The goals of this research were to identify and quantify the key factors influencing
the customer’s selection of residential modular building projects, determine their relative
importance, and establish the evaluation scale. Furthermore, in our study, we aimed to
demonstrate how to utilize generative AI tools such as ChatGPT [5] to cover the lack of
data from experts. While there are many well-known methods that are like the Delphi
Buildings 2024, 14, 1696 3 of 36

methods. The main task for ChatGPT is to complement experts output, which is needed for
both the research itself and the successful application of its outcome for practical cases in
the future.
The objective of this research was modular house projects in a project evaluation
context. The main scientific hypothesis of this research is that ChatGPT can be efficient
enough to provide high-quality results by creating a description of the evaluation scale that
complements and speeds up the expert’s work and reduces the description to corrections
using the limited one-shot learning technique, i.e., when the model is given a single example
from which to infer how to perform a task.

1.1. The Background of the Research

In our research on the assessment of modular construction projects, we placed a strong
emphasis on evaluating customer preferences and establishing evaluation scales. This
approach aligns with innovative methodologies used in various industries. For instance,
studies on mobile phone selection and the energy-sustainable renovation of building
envelopes employ the Analytic Hierarchy Process (AHP) for decision making, a method
that is also crucial in our assessments of modular buildings [6,7]. Further emphasizing
the importance of robust analytical frameworks, research on building energy modeling
at an urban scale resonates with our focus on sustainability and efficiency in modular
construction [8].
These shared methodologies manage complexities across different construction disci-
plines, including energy efficiency in building renovations and infrastructure management
based on multicriteria approaches [9,10]. The latter, employing a two-phase AHP and the
TOPSIS method, mirrors our approach to modular construction, where balancing various
stakeholder interests and efficient resource allocation is crucial. In our work, as an illustra-
tion, we used a weighted method, which also required determining the relative values of
the evaluated factors.
Further parallels are found in research focused on value management (VM) in con-
struction and the prioritization of social assistance candidates, emphasizing the need for
structured decision-making models [11,12]. A recent study on improving the seismic per-
formance in modular buildings, particularly focusing on prefabricated steel modules with
multi-tuned mass damper floor systems, directly complemented our research in modular
construction [13]. This study’s innovations in enhancing structural integrity and resilience
are integral to our goals of ensuring that modular buildings are not only efficient and
environmentally friendly but also structurally robust and safe.
The exploration of regenerative circularity in the built environment and asset manage-
ment in asset-intensive industries offers further insights. Studies focusing on sustainability
assessment tools and strategic plans for identifying critical assets mirror our efforts in incor-
porating sustainability and adaptability into modular construction [14,15]. Additionally, the
significant roles of knowledge, awareness, and environmental consciousness in adopting
green practices in construction organizations underline the importance of fostering such
attributes among stakeholders, which is a key aspect in advancing sustainable modular
building methods [16].
Technological integration also plays a pivotal role. Studies on the effective use of
Building Information Modelling (BIM) in the AECO industry, as well as structural design
optimization (SDO) using BIM and optimization algorithms highlight the integration
of technologies to enhance user experience and support sustainability [17,18]. These
principles are crucial not just in smart infrastructure but also in the evolving field of
modular construction.
Finally, the insights from sustainability transition studies in the construction industry
resonate with our findings in modular construction, emphasizing the importance of tech-
nological and organizational factors, the role of information sharing, and the influence of
social media and key opinion leaders in technology adoption [19].
Buildings 2024, 14, 1696 4 of 36

Collectively, these various studies underscore the importance of multi-criteria de-

cision making, stakeholder engagement, technological integration, and sustainability in
advancing the field of modular construction. The integration of modern methodologies
and innovations from related fields can significantly enhance modular building projects.
A comprehensive comparison between traditional on-site and modular off-site con-
struction methods was conducted to understand their environmental impact. Using three
representative single-family houses in Okanagan, British Columbia, Canada, a cradle-to-
gate life cycle assessment was performed. Data related to materials, transportation, and
energy were collected, and environmental impact measures were calculated. The com-
parisons indicated that neither method was definitively more environmentally friendly.
However, the study suggested that implementing certain strategies like optimal design
and reducing material and worker transportation could minimize environmental impacts.
The study provided a foundation for future life cycle assessment studies on new building
projects [20].
Recent research introduced a framework to assess the economic sustainability of
single-family modular houses over their life cycle. It involves selecting economic perfor-
mance criteria, developing indicators, and using multi-criteria decision analysis to create
economic sustainability indices. The framework was applied to two modular houses in
British Columbia, Canada, with results showing excellent overall economic sustainability.
However, certain areas required improvement. The framework can guide decisions on
construction methods and pinpoint areas needing enhancement in a modular building’s
life cycle [21].
Modular buildings feature streamlined construction processes, precise resource uti-
lization, predictable scheduling, adherence to building codes, rigorous quality control,
and sustainability measures like resource efficiency and energy efficiency. The higher
speed of construction, better productivity and workmanship, cost savings, higher safety,
higher product control and quality, and less negative environmental impacts are the most
noticeable advantages of modular construction [2]. These characteristics enhance efficiency,
compliance, and sustainability compared to traditional construction methods.
Integrating ChatGPT and similar large language models (LLMs) into construction
research offers new possibilities and challenges, a notion supported by a study that critically
assessed GPT models in the industry [22]. This research revealed the potential of GPT
models to enhance various project life cycle stages, identified challenges, and created a
prototype for material optimization. These findings underscore the relevance of our work
in applying ChatGPT for modular house project evaluation, highlighting the untapped
potential of LLMs to revolutionize construction management and sustainability efforts.
GPT models have also been proved applicable in construction scheduling based on specific
project requirements; however, the technology still has limitations, and further development
is needed before it can be widely adopted in the industry [23].
Summarizing everything said above, we showed the context and the background of
this research—there are several opportunities for pioneering the incorporation of genera-
tive AI technology into the construction field. Thus, the current research is dedicated to
continuing the topic that was extensively discussed in [22] by presenting a new concrete
way to incorporate generative AI technology into construction industry, more specifically
into the project planning phase when project assessment is performed, and the hard work
of a competent expert is required.

1.2. The Role of ChatGPT in This Research

Besides the utilization of ChatGPT to improve the text (with careful revision, without
the creation of new knowledge), which is an obvious direct application (for the current
text as well) of ChatGPT, in our study, we demonstrated how to utilize generative AI tools
such as ChatGPT to cover the lack of data from experts. Thus, the main task for ChatGPT
in the current research context was to complement experts output, which is needed for
both the research itself and the successful application of its outcome for practical cases in
Buildings 2024, 14, 1696 5 of 36

the future. Implementing an AI tool like ChatGPT as a knowledge expert is particularly

beneficial in navigating the challenges presented in [24]. Given the challenges in collecting
and interpreting data, especially from small surveys that may be subjective, an AI like
ChatGPT can provide a clear and unbiased perspective based on data.
It can process vast amounts of information from a wide range of sources to provide
valuable insights, mitigating some of the subjective bias inherent in human responses.
Moreover, in an industry where professionals are pressed for time, ChatGPT can handle
vast amounts of data swiftly and efficiently, relieving the burden from human experts. It
can effectively analyze information and provide comprehensive reviews in much shorter
time if we compare it to a human analyst. Importantly, while the framework proposed
in [24] relies on continuous benchmarking processes that may take years to develop a
comprehensive database, an AI like ChatGPT can expedite this process. It can swiftly
review, revise, and recalibrate benchmark values and tweak performance scales in real
time, adapting to site-specific socio-economic and geographical conditions. This offers a
more efficient, adaptable and dynamic approach to benchmarking in the emerging modular
construction industry. In conclusion, in our research, ChatGPT provided an efficient,
objective, and comprehensive analysis of data for benchmarking in the modular building
industry, making it an invaluable tool for such research.
The following are the progressive steps involved in the ChatGPT framework [5]:
1. Interactive Dialogue Structure: ChatGPT’s interactions resemble a structured discussion.
Using this dialogue framework, ChatGPT can respond to follow-up queries, correct
mistakes it has made, refute false assertions, and decline inappropriate requests.
2. Data Collection and Reward Model Training: ChatGPT is built on GPT-3.5, a language
model that employs deep learning to generate human-like writing. The process
includes collecting comparison data and training a reward model to evaluate the
generated content.
3. Sampling and Policy Initialization: The database is sampled to obtain a fresh prompt.
The supervised policy serves as the initialization for the proximal policy optimization
(PPO), guiding the policy to produce a result.
4. Reward Determination and System Update: A reward is determined by the reward
model for the generated output, utilizing proximal policy optimization. The reward is
then used to update the policy, contributing to the iterative improvement of the system.
Our research mirrors the complexities mentioned in [25] regarding selection and
evaluation by experts, the necessity of multi-aspect viewpoints, and the need to avoid
inconsistency in responses when developing sustainability-type questionnaires. To navigate
these challenges, we propose the use of ChatGPT. ChatGPT, a sophisticated AI developed
by OpenAI, was designed to generate human-like text based on the input it receives.
By integrating ChatGPT into our process, we were able to mitigate some of the issues
associated with human variability and bias. As an AI, ChatGPT is inherently neutral
and consistent, providing reliable input free from the subjectivity inherent to human
experts. This AI can also manage the high volume of respondents needed for a robust
study. Unlike human researchers, ChatGPT can swiftly process and interpret data from
many participants, reducing the time required for data analysis. Moreover, using linguistic
variables for assessment, as mentioned in [25], becomes more efficient with ChatGPT. The
AI can interpret, relate, and assign fuzzy numbers to linguistic terms, ensuring a more
precise and objective assessment. Lastly, the AI excels in questionnaire analysis. It can
rapidly analyze filled-out questionnaires, identify patterns, and provide insights, again
reducing the time required for this process and increasing the consistency of the results. In
summary, integrating ChatGPT into our research process provides an efficient, objective,
and reliable approach to overcoming the challenges associated with the evaluation and
analysis stages of the construction project.
The rapid advancements in artificial intelligence (AI), particularly in large-scale lan-
guage models like ChatGPT, have ushered in a new era of digital transformation. With its
extensive training data and sophisticated algorithms, ChatGPT has demonstrated remark-
Buildings 2024, 14, 1696 6 of 36

able abilities, especially in generating human-like text across diverse disciplines, facilitating
interdisciplinary research, and serving a plethora of use cases ranging from customer
service to research collaboration [26,27]. Its potential to democratize scientific research,
simplify tasks, and make academic pursuits more effective and time-saving underscores
the transformative nature of such AI models [28].
However, as with any powerful technology, the use of ChatGPT is not without its
challenges. Concerns about the reliability and accuracy of its outputs, biases inherent
in its training data, and its opaque decision-making process have been raised [26,28].
Additionally, there are ethical challenges associated with AI models in research, including
data privacy concerns and the potential for unintentional plagiarism.
Yet, the potential of ChatGPT to revolutionize various domains cannot be understated.
Its ability to comprehend conversational contexts, generate coherent and grammatically
correct language, and adapt across sectors and disciplines positions it as a game-changer
in the AI landscape [27]. Moreover, the concept of human–machine collaboration un-
derscores the dynamic interaction between humans and AI, highlighting the need for a
holistic understanding of their synergy across cognitive, metacognitive, affective, and social
domains [29].
We propose the use of the Multi-Criteria Decision Making (MCDM) Delphi method,
complemented by the capabilities of ChatGPT, to evaluate the projects of modular buildings.
The Delphi method is a Bayesian parameter inference MCDM method that allows for the
estimation of the posterior distributions of model parameters without relying on explicit
likelihood functions. This approach is particularly effective in handling the complexities
and uncertainties inherent in the evaluation of modular buildings, which can vary widely
in design, materials, and construction methods [30].
The Delphi method enables us to gather and refine experts’ opinions iteratively to
estimate parameters that significantly affect the performance of modular buildings, such as
materials used, construction methods, and design choices. Following this, we planned to
utilize ChatGPT to further analyze and improve the evaluation of individual criteria for
specific modular construction projects. ChatGPT’s advanced natural language processing
capabilities can assist in interpreting the Delphi method’s results, providing nuanced
insights and suggestions for each criterion. This can include generating detailed reports,
proposing optimization strategies, or even identifying potential areas of risk or innovation
based on the aggregated expert opinions [31].
Moreover, ChatGPT can facilitate dynamic interaction with the data, allowing stake-
holders to query specific aspects of the modular construction project and receive instant,
informed responses. This interactive approach can significantly enhance decision-making
processes, ensuring that all considerations are thoroughly evaluated.
The combination of the Delphi method and ChatGPT offers a comprehensive and
robust framework for modular building evaluation. While the Delphi method provides a
solid foundation through expert consensus, ChatGPT adds a layer of intelligent analysis and
interaction, making the evaluation process more thorough and adaptable. This integrated
approach not only accommodates the heterogeneity and complexity of modular buildings
but also embraces the variability in expert opinions, leading to a more refined and actionable
evaluation. Despite the fact that Delphi is quite an old method that was developed in
the 1950s by the RAND Corporation to forecast the impact of technology on warfare,
its age does not diminish its usefulness. Rather, the Delphi method has proven to be
adaptable and valuable across different disciplines, including business, education, health,
and public policy. It is particularly useful in situations where individual judgments need to
be synthesized into a group consensus and direct communication between experts is not
feasible or is undesirable due to potential bias or influence. In recent years, the popularity
of this method in scientific articles was significantly increased (Figure 2).
Buildings 2024, 14, 1696 7 of 36

1000

Number of Articles
800
600
400
200
0

13

14

15

16

17

18

19

20

21

22

23
20

20

20

20

20

20

20

20

20

20

20
Year

Figure 2. The number of articles by year with the Delphi method. This result was obtained from the
Web Of Science platform with an advanced search query, “AK = (Delphi OR Delphi method)”.

In conclusion, integrating the Delphi method with ChatGPT presents a powerful and
innovative tool for assessing the performance of modular buildings. This combination
allows for a detailed and interactive evaluation of various performance parameters, enhanc-
ing the quality and effectiveness of modular building projects through a more informed
and data-driven decision-making process.
This study made the following contributions:
• Conducted a comprehensive survey using the Delphi method to identify and select the
critical factors affecting customer selection in modular construction projects, demon-
strating the importance of expert and consumer input in refining research focus areas
within the modular construction domain.
• Introduced an innovative research design that strategically integrates ChatGPT to im-
plement the application of these selected factors by introducing a detailed evaluation
scale. This showcases the potential for AI to complement traditional research method-
ologies by enhancing the precision and relevance of data analysis in the modular
construction sector.
• Demonstrates the process of crafting detailed prompts to guide ChatGPT, ensuring
the generation of relevant and accurate information for the development of an evalua-
tion scale specifically tailored to modular construction projects. This highlights the
meticulous approach required to effectively utilize AI tools in academic research.
• Calls for the development of sophisticated AI-driven tools tailored to the construc-
tion industry, capable of navigating complex consumer decision-making processes
and sustainability considerations. This contribution emphasizes the need for further
exploration into the integration of AI in enhancing sustainable housing develop-
ment strategies, while also acknowledging the foundational work conducted without
AI assistance.
The remainder of this paper is structured as follows. Section 2 describes the method-
ology of the research. Section 3 discusses how the methodology was applied to the case
study. Section 4 presents the results of the proposed research methodology applied to the
real case in Lithuania. Finally, Section 5 presents and discusses the results from the case
study. Section 6 provides concluding remarks and proposes future research areas.

2. The Methodology of this Research

In this section, we present the methodology of this research, i.e., how we addressed
the problem of evaluation. More specifically, we address the evaluation of each criterion
for evaluation separately.

2.1. Factors for Evaluation

In the pursuit of evaluating the the multifaceted dimensions of modular housing
projects, our research identified a series of critical factors that influence their success and
Buildings 2024, 14, 1696 8 of 36

sustainability. To contextualize these factors within the broader academic and practical
discourse, we conducted a comprehensive review of the existing literature. Table 1 presents
the culmination of this effort, listing each identified factor alongside references to the
works of other researchers who have investigated similar themes. This alignment not only
validates our research approach but also situates our findings within the ongoing dialogue
on modular construction, bridging the gap between our study and the broader body of
knowledge in the field. These factors are the input criteria for the first round of the Delphi
method. The factors presented in Table 1 are defined in more detail in Sections 2.2.1–2.2.14.

Table 1. Factors and corresponding references.

Factor Number and Name Sources

B1.1. Construction time [21,32–34]
B1.2. Design costs [21,34]
B1.3. Construction costs [21,32,34,35]
B1.4. Building lifespan [21,32,33]
B1.5. Amount of investment [34,36]
B1.6. Building guarantee lengths [30]
B1.7. Exploitation costs [21,35]
B1.8. Modular building adaptivity [21]
B1.9. Traceability of products and works *
B1.10. Design time [21,34]
B1.11. Maintenance costs [21,32]
B1.12. Disposal costs [21]
B1.13. Investment risk [21]
B1.14. Compatibility of elements from different manufacturers *
B2.1. Geometry and configuration [30,33,37,38]
B2.2. Aesthetic design [30,32,35,37–39]
B2.3. Functionality [32,33]
B2.4. Energy efficiency class [30]
B2.5. Heating and ventilation [32,37,39]
B2.6. Insolation and light [37,39]
B2.7. Sound isolation [37]
B2.8. Amenities [37,39,40]
B2.9. The quality of the modular house, the completeness of the work, the
*
delivery and the preparation of the documentation
B2.10. The area [33,37,38]
B2.11. Number of floors [33,38]
B3.1. The usual (permanent residence) location [34]
B3.2. Protection against burglary and theft [30]
B3.3. A good view from the window, the presence of greenery around *
B3.4. The situation of the neighborhood (friendly, safe environment) [21,34,40]
B3.5. Compliance with the status of the resident [34]
B3.6. Adaptation to the lifestyle [32,34]
B4.1. Geographical location [33,35,37,40]
B4.2. Site availability [38]
Buildings 2024, 14, 1696 9 of 36

Table 1. Cont.

Factor Number and Name Sources

B4.3. Distance from the city center [36,38,40]
B4.4. Parking lot and other infrastructure [39]
B4.5. Swimming pool, garden [36,39]
B4.6. Public and private service facilities [36,38,39]
B4.7. City structure and development [36]
B5.1. Compliance with the strategy of renting, selling or living in [34]
B5.2. Compliance with the situation on the labor market and labor supply [21,40]
B6.1. Recyclability of elements [21,32,41]
B6.2. Absence of negative effects on humans [21,32,41]
B6.3. No negative impact on nature [21,41]
B6.4. Production from renewable sources [21,35,41]
B6.5. Eco-friendly operation [42]
B6.6. Ecological production [21,41]
B6.7. Environmental friendliness of transportation [21,43]
CB5. Preliminary price [30,34]
CB6. Compliance with norms and standards [30]
* These factors were obtained from respondents during the first round of the Delphi method.

2.2. Ranking Criteria

Here, we present the criteria for the ranking of different factors for the modular
building project evaluation. Every criteria is ranked with a number from 1 to 5 (also known
as the Likert scale). A higher numerical value corresponds to a more favorable project
attribute. Specifically, the scale is defined as follows:
1. Poor—The project significantly lacks the desirable attributes of the given criterion.
2. Fair—The project displays basic features of the criterion but falls short of a satisfac-
tory standard.
3. Good—The project meets the expected standards for the criterion, representing an
average performance.
4. Very Good—The project exceeds the average standards, showing enhanced attributes
of the criterion.
5. Excellent—The project excels in fulfilling the criterion, showcasing exemplary perfor-
mance and attributes.
This ranking system was designed to provide a quantitative measure of each project’s
characteristics, facilitating a comprehensive evaluation based on predefined criteria. It can
be used in two general ways to implement the evaluation of each factor of the project:
1. Evaluate each factor for each object using some expert opinion-based evaluation procedure.
2. Set up an evaluation scale with defined criteria for each rank of each factor.
The second way is not available for all factors, as it must involve the subjective opinion
based on the factor’s definition (e.g., “A good view from the window”). However, where
it is appropriate, the defined criteria for each rank (which lets us define the scale) could
improve the evaluation process in speed and robustness, eliminating the subjectivity. The
main barrier of such an approach is the need to perform the expensive work of an expert
with high competence, and, as it was mentioned before already, we intend to reinforce the
expert with ChatGPT.
The scale values represent a consensus-driven approach to appraising the multifaceted
aspects of modular construction projects, ensuring a standardized evaluation process. The
Buildings 2024, 14, 1696 10 of 36

ranking scale (along with detailed descriptions for each rank) was proposed by Chat-
GPT and was verified by the expert. According to the expert, no major changes were
required; however, some minor inconsistencies were identified. More specifically, there
were some overlapping ranges of parameters in the scale for evaluations. Thus, it was
decided to correct the output provided by ChatGPT without some major reiteration by
ChatGPT—according to the expert, the corrections finalized the result. Thus, in the points in
Sections 2.2.1–2.2.14, the results were obtained using appropriate prompts with ChatGPT,
and details with the original prompts and output are provided in Appendix A.

2.2.1. Construction Time

The ranking of construction time is based on the predicted time that is needed for
the construction of the building. Modular buildings generally have shorter construction
times compared to traditional or classical buildings. The precise construction time can vary
depending on several factors, including the complexity and size of the project, the manu-
facturing capabilities of the modular construction company, site preparation requirements,
and the level of customization desired. However, in general, modular buildings tend to be
constructed more quickly due to the following reasons:
• Off-Site Construction: Modular buildings are constructed in a factory-controlled envi-
ronment, allowing for simultaneous work on site preparation and building fabrication.
This off-site construction approach reduces construction time significantly.
• Parallel Work Processes: With modular construction, site preparation tasks, such as
foundation construction or utility connections, can proceed concurrently with the
fabrication of modules in the factory. This parallel work process accelerates the overall
construction schedule.
• Streamlined Production: Modular buildings follow standardized designs and construc-
tion processes. The manufacturing process in a factory environment allows for efficient
assembly line production, with skilled workers specializing in specific tasks. This
streamlined production process minimizes delays and improves construction speed.
• Reduced Weather Dependency: Since the majority of the modular building construc-
tion takes place in a controlled factory environment, it is less affected by adverse
weather conditions that can cause delays in traditional construction. This factor helps
maintain consistent construction schedules and reduces the overall project timeline.
As a result of these and some other factors, when compared to more conventional
construction methods, modular construction can result in savings of 50% (time), 30%
(materials), and 25% (waste), thus achieving significant economic savings [44]. However, it
is important to note that the exact building time can still vary depending on the specific
project requirements and circumstances.
On average, modular buildings can be constructed in a shorter timeframe compared
to traditional or classical buildings. While the exact duration can vary depending on the
size and complexity of the project, as well as other factors, here is a general comparison:
• Modular Buildings: The construction of modular buildings typically takes less time
compared to traditional construction methods. Depending on the size and complexity
of the project, modular buildings can often be completed in a range of a few weeks to
a few months. Smaller modular structures, such as single-family houses or small office
buildings, can often be completed in a matter of weeks. Larger modular buildings,
such as multi-story commercial structures, may take a few months to complete.
• Traditional/Classical Buildings: The construction timeline for traditional buildings is
typically longer due to the sequential nature of the construction process. It involves
various stages, including site preparation, foundation construction, framing, inte-
rior and exterior finishing, and other installations. The duration of construction for
traditional buildings can vary significantly depending on the size, complexity, and
construction methods used. On average, the construction of traditional buildings can
take several months to over a year, depending on the project scope.
Buildings 2024, 14, 1696 11 of 36

It is important to note that these estimates are general and can be influenced by
project-specific factors. Both modular and traditional construction methods have their own
advantages and considerations, and the choice between them should be based on factors
such as project requirements, budget, site constraints, and timeline preferences.
Construction times can vary widely based on the size and complexity of the building,
the number of workers, local regulations, and many other factors. Rough estimates are
provided in Table 2.

Table 2. Time ranges for construction.

Rank Traditional Construction Time (Months) Modular Construction Time (Months)

1 [24,36) [12,18)
2 [18,24) [9,12)
3 [12,18) [6,9)
4 [9,12) [4,6)
5 [6,9) [3,4)

2.2.2. Amount of Investment

The ranking of cost-related factors is based on the predicted costs that are needed
for the construction of a building. Projecting costs, construction costs, and the amount of
investment are related to the financial aspects of a construction project, but they represent
different aspects of the overall expenses. Here is a breakdown of each term:
Projecting Costs: Projecting costs, also known as design or planning costs, refer to the
expenses incurred during the initial stages of a construction project. These costs typically
include architectural and engineering fees, site surveys, soil testing, permit fees, and other
expenses related to the design and planning phase. Projecting costs are associated with
developing the project concept, conducting feasibility studies, creating design plans, and
obtaining necessary approvals and permits. These costs are incurred before the actual
construction begins and are part of the overall project budget.
Construction Costs: Also known as hard costs, these are the expenses directly associ-
ated with the physical construction of the building. These costs include materials, labor,
equipment, subcontractor fees, construction permits, site preparation, foundation work,
structural components, interior and exterior finishes, mechanical and electrical systems
installation, and any other construction-related expenses. Construction costs represent
the largest portion of the overall project budget and are directly related to the actual
construction work.
Amount of Investment: The amount of investment refers to the total capital or finan-
cial resources needed for the construction project. It includes both projecting costs and
construction costs, along with other related expenses. The amount of investment is the
total sum required to initiate and complete the project, covering all the planning, design,
and construction activities. It includes costs for land acquisition, design and planning fees,
construction costs, contingency funds, financing expenses, professional fees, marketing
expenses (if applicable), and any other costs associated with the project. The amount of
investment represents the overall financial commitment required to bring the construction
project to completion.
In summary, projecting costs are the expenses incurred during the design and planning
phase, construction costs are the expenses directly related to the physical construction work,
and the amount of investment is the total capital required for the entire project, encompass-
ing both projecting costs and construction costs, along with other associated expenses.
In addition to projecting costs and construction costs, the amount of investment for a
construction project may include several other related expenses. These expenses can vary
depending on the project scope, location, and specific requirements. Here are some common
examples of other related expenses that may be included in the amount of investment:
Buildings 2024, 14, 1696 12 of 36

Land Acquisition: the cost of purchasing or acquiring the land on which the con-
struction project will be built. This can include the purchase price of the land, legal fees,
surveying costs, title searches, and any necessary site preparations.
Financing Expenses: costs associated with obtaining financing for the project, such
as loan origination fees, interest payments, loan guarantee fees, and other financing-
related expenses.
Contingency Funds: an allowance for unforeseen circumstances or changes that may
arise during the construction process. Contingency funds are set aside as a buffer to cover
unexpected expenses or project modifications.
Professional Fees: fees paid to professionals involved in the project, such as architects,
engineers, project managers, consultants, and legal advisors. These fees can vary based on
the size and complexity of the project.
Insurance and Bonding: costs associated with insurance coverage for the project,
including builder’s risk insurance, liability insurance, and any necessary performance or
payment bonds.
Utilities and Infrastructure: expenses related to connecting the project to essential
utilities such as water, electricity, gas, and sewer systems. This can include fees for utility
hookups, the installation of infrastructure, and any necessary upgrades or extensions.
Permitting and Regulatory Costs: fees associated with obtaining permits, licenses, and
complying with local building codes and regulations. This can include building permits,
environmental impact assessments, zoning fees, and inspections.
Marketing and Sales Expenses (if applicable): costs associated with marketing and
promoting the project, such as advertising, sales commissions, model unit construction,
and staging.
Administrative and Overhead Costs: general administrative expenses related to project
management, office space, utilities, accounting, legal services, and other operational costs.
It is important to note that the specific expenses included in the amount of investment
can vary depending on the nature and scale of the project. It is recommended to consult with
financial advisors, project managers, or construction professionals to accurately assess and
estimate the full range of expenses that should be considered in the amount of investment
for a particular construction project.
Table 3 lists five intervals of price bounds for projecting costs, construction costs, and
the amount of investment for both traditional and residential modular buildings, ranked
from the most expensive to the cheapest.

Table 3. Different types of costs for differernt types of buildings ($/sq.m).

Traditional Modular Traditional Modular Total

Rank Total Modular
Projecting Projecting Construction Construction Traditional
1 [540,648) [486,594) [5400,6480) [4860,5940) [6480,7560) [5940,7020)
2 [432,540) [378,486) [4320,5400) [3780,4860) [5400,6480) [4860,5940)
3 [324,432) [270,378) [3240,4320) [2700,3780) [4320,5400) [3780,4860)
4 [216,324) [162,270) [2160,3240) [1620,2700) [3240,4320) [2700,3780)
5 <216 <162 <2160 <1620 <3240 <2700

2.2.3. Building Lifespan

The lifespan of a building is dependent on many factors including the quality of
construction, the materials used, and the ongoing maintenance. Here are five ranks from
the shortest to longest lifespan, considering both traditional and modular buildings:
1. Rank 1—Very Short Lifespan: temporary modular buildings such as event tents or
emergency shelters (1–5 years). These structures are designed for immediate use and
are not built to last. They serve their purpose in disaster-struck areas, temporary
events, or short-term accommodation needs.
Buildings 2024, 14, 1696 13 of 36

2. Rank 2—Short Lifespan: portable cabins or temporary classrooms (6–20 years).

These are often seen in construction sites, schools, or as temporary offices. They are
made more durable than the very-short-lifespan buildings but are not meant for very
long-term use.
3. Rank 3—Average Lifespan: prefabricated houses or manufactured houses (21–60 years).
These houses are built in factories and then assembled on-site. They are designed for
longer-term use and can withstand weather conditions better than temporary structures.
4. Rank 4—Long Lifespan: permanent modular houses with high-quality materials
(61–100 years). These houses are built to last, using quality materials and construc-
tion techniques that ensure their longevity. They are intended to serve as full-time
residences for decades.
5. Rank 5—Very Long Lifespan: High-end, architect-designed modular houses (101+ years).
These houses are designed and built with exceptional attention to detail, using high-quality,
durable materials. With proper care and maintenance, they can last for over a century,
similar to traditionally constructed houses.

2.2.4. Building Warranty Lengths

The warranty lengths of the buildings were ranked using ChatGPT. The obtained
ranking data for traditional and modular buildings are presented in Table 4.

Table 4. Rankings for modular building warranty lengths.

Rank Traditional Building Warranty (Years) Modular Building Warranty (Years)

1 ≤2 ≤2
2 [2,5) [2,5)
3 [5,10) [5,10)
4 [10,20) [10,20)
5 20+ 20+

Building warranty length refers to the period during which the builder or manufacturer
of a building is responsible for repairing or resolving certain defects and issues that may
arise after construction. Here is a detailed breakdown:
• Nature of Warranty: This covers defects in materials, workmanship, structural issues,
and major systems. Specific coverage varies by warranty.
• Duration: This ranges from a few years to several decades, depending on the type of
warranty and covered components.
• Types of Warranties: These include the following:
– Limited Warranties: These cover specific elements for a defined period.
– Structural Warranties: These focus on structural integrity, including foundations,
and load-bearing parts.
– Home Builder Warranties: These are for new residential buildings, varying in
length and coverage.
• Importance: This protects against costly repairs due to construction defects and serves
as a quality assurance measure.
• Transferability: Some warranties can be transferred to new owners, enhancing prop-
erty value.
• Limitations and Exclusions: Not all issues are covered; exclusions vary by warranty.
• Claim Process: This requires filing a claim within the warranty period for assessment
and resolution.
Buildings 2024, 14, 1696 14 of 36

2.2.5. Building Exploitation Costs

Modular building exploitation costs are provided in Table 5, they refer to the on-
going expenses associated with operating and maintaining a modular building after its
construction and installation. These costs can vary depending on factors such as the size
and purpose of the building, its location, the level of usage and occupancy, and the spe-
cific requirements for maintenance and utilities. Some common components of modular
building exploitation costs include the following:
Utilities: These include expenses for electricity, water, heating, cooling, and other
utilities necessary for the functioning of the modular building. The cost will depend on the
energy efficiency of the building and the local utility rates.
Maintenance and Repairs: Regular maintenance, inspections, and repairs are essential for
preserving the condition and functionality of the modular building. This includes tasks such
as cleaning, painting, HVAC system maintenance, plumbing repairs, and general upkeep.
Insurance: Modular buildings, like any other structures, require insurance coverage
to protect against potential risks such as property damage, liability claims, or accidents.
The insurance premiums will depend on factors such as the building’s value, location, and
intended use.
Property Taxes: Modular buildings are subject to property taxes, which are typically
based on the assessed value of the building and the local tax rates. The tax rates may vary
depending on the jurisdiction.
Security and Safety Measures: Expenses related to security systems, surveillance
equipment, fire alarms, and safety protocols are essential for ensuring the protection of the
modular building and its occupants. This includes costs for security personnel, monitoring
services, and periodic safety inspections.
Cleaning and Waste Management: Regular cleaning services and waste disposal are
necessary for maintaining cleanliness and hygiene within the modular building. The
costs will depend on the size of the building, the frequency of cleaning, and the waste
management services utilized.
Property Management: If the modular building is part of a larger property or complex,
there may be additional costs associated with property management services, such as
administrative fees, maintenance coordination, and tenant communication.
It is important to note that the specific exploitation costs for a modular building can
vary significantly depending on various factors mentioned earlier. It is recommended
to consult with professionals or property management experts who can provide a more
accurate estimate based on the specific circumstances of the modular building in question.
These costs include building maintenance, utilities, property taxes, and other ongoing
expenses associated with the building exploitation. Note that the actual cost can vary
greatly depending on the specific property, its location, usage, and many other factors.

Table 5. Ranking of exploitation costs.

Traditional Building Exploitation Modular Building Exploitation

Rank
(USD/sq.m/year) (USD/sq.m/year)
1 [65,86) [54,76)
2 [43,65) [32,54)
3 [22,43) [16,38)
4 [11,22) [5,16)
5 <11 <5

2.2.6. Modular Building Adaptivity

Modular building adaptivity refers to the flexibility of the building design to be
changed or adjusted over time based on varying needs. This could include the ability to
add or remove modules, rearrange internal layouts, or even relocate the entire building.
Buildings 2024, 14, 1696 15 of 36

Here are five ranks, from lowest to highest, considering aspects such as design flexibility,
ease of modification, and cost of changes:
1. Rank 1—Poor: The building design allows for very minimal or no changes in layout
or use over time. Major alterations would require significant structural changes, high
costs, and lengthy timeframes.
2. Rank 2—Fair: The design permits some minor changes, such as small interior mod-
ifications, without major structural alterations. However, these changes may still
involve considerable costs and time.
3. Rank 3—Good: The building design incorporates a moderate level of flexibility. It
allows for changes in room layouts or usage with relative ease and at moderate cost.
However, larger scale changes may still be challenging or costly.
4. Rank 4—Very Good: The design allows for significant flexibility. Large-scale changes
in layout or use can be achieved at moderate cost and in a reasonable timeframe. The
building may also have features that facilitate easy modifications, such as standardized
connections between modules.
5. Rank 5—Excellent: The building design provides for maximum adaptability. It allows
for significant changes, including the addition or removal of modules, changes in layout
or use, or even the relocation of the building with relative ease, minimal cost, and
within short timeframes. The design may incorporate specialized features that enhance
adaptability, such as quick-release connections or adjustable structural elements.

2.2.7. Traceability
Traceability in the context of modular building construction refers to the ability to
track the progress of a job or project, the parts used, and the workers involved, often for
the purpose of quality assurance, record keeping, and regulatory compliance.
Here are five ranks, from lowest to highest, considering the ability to track the various
stages of production, quality control measures, and labor tracking:
1. Rank 1—Poor: minimal or no systems in place for tracking job progress, components
used, or workers involved; no digital record keeping; and limited or no documentation
of job stages.
2. Rank 2—Fair: basic tracking systems in place, possibly manual record keeping or
rudimentary digital systems. Documentation may not be consistent or comprehensive,
and quality control measures may be lacking.
3. Rank 3—Good: digital tracking systems in use, with the regular documentation of
job progress, components, and labor. Quality control measures are employed, but
they may not cover all aspects of the project or all stages of construction.
4. Rank 4—Very Good: advanced digital tracking systems in use, with comprehensive
and regular updates on job progress, components, and labor. Quality control is
integrated into the tracking process, with most aspects of the project covered.
5. Rank 5—Excellent: state-of-the-art digital tracking systems in use, with real-time
updates on job progress, components, and labor. Quality control measures are fully
integrated into the tracking system, covering all aspects of the project at all stages of
construction. Traceability is considered in the design stage and is a key component of
the construction process.

2.2.8. Geometry and Configuration

Assessing building geometry and configuration objectively is a complex task that
involves several criteria, including spatial efficiency, accessibility, lighting, thermal comfort,
and acoustics. This method is based on the book of Markus and Morris [45]. Each of the
following five ranks is associated with an increasing level of efficiency and comfort:
1. Rank 1—Poor: The configuration and geometry of the building significantly reduce
the spatial efficiency. It is challenging to move around due to the poor layout. It
Buildings 2024, 14, 1696 16 of 36

has inadequate natural lighting and ventilation, resulting in poor thermal and acous-
tic comfort.
2. Rank 2—Fair: The building has some issues with spatial efficiency, layout, and
accessibility. While there is some allowance for natural light and ventilation, these are
not optimized, affecting thermal and acoustic comfort to a certain degree.
3. Rank 3—Good: The building offers satisfactory spatial efficiency with a layout that
facilitates ease of movement. There is a balance of natural and artificial light, and
sufficient ventilation to ensure adequate thermal and acoustic comfort.
4. Rank 4—Very Good: The building’s geometry and configuration contribute to high
spatial efficiency. The layout allows for ease of movement and good accessibility. Nat-
ural light and ventilation are effectively used to enhance thermal and acoustic comfort.
5. Rank 5—Excellent: The building’s geometry and configuration demonstrate optimal
use of space, with a layout that ensures easy movement and excellent accessibility.
Natural light and ventilation are optimized to provide excellent thermal and acoustic
comfort, meeting and exceeding all design standards.

2.2.9. Aesthetic Design

Assessing the aesthetic design of a building is a subjective task as it largely depends
on individual preferences. However, it is possible to use certain objective criteria based
on architectural theories and principles, such as proportion, harmony, rhythm, contrast,
and unity. In the following, each of the five ranks is associated with an increasing level of
aesthetic appeal [46]:
1. Rank 1—Poor: The design does not consider principles of proportion, harmony,
rhythm, contrast, and unity. The building does not evoke a positive response and
lacks visual appeal.
2. Rank 2—Fair: The design reflects a basic understanding and implementation of the
aesthetic principles but lacks cohesiveness. Some elements may appear disconnected
or disjointed.
3. Rank 3—Good: The design shows a balance of aesthetic principles with a noticeable
sense of proportion, harmony, and rhythm. The building’s design evokes a generally
positive response.
4. Rank 4—Very Good: The design effectively incorporates aesthetic principles, resulting
in a visually appealing structure. Proportion, harmony, rhythm, contrast, and unity
are clearly exhibited in the design, evoking a strong positive response.
5. Rank 5—Excellent: The design demonstrates an exceptional understanding and
application of aesthetic principles. The building’s design is highly appealing, evoking
a strong emotional response and effectively blending proportion, harmony, rhythm,
contrast, and unity to create a compelling and memorable aesthetic experience.
Please note that aesthetic appeal can be highly personal and vary greatly among
different people and cultures. Therefore, this is a generalized guide that may not apply to
every case.

2.2.10. Energy Efficiency Class

Usually, the following classes are used for standard construction projects: A++, A+, A,
B, C, D, E, F, and G. We split these into the following five groups, and since A++ is used as
an important requirement for many construction projects in the Euro zone, we put it into a
separate category:
1. Rank 1: G and F;
2. Rank 2: E and D;
3. Rank 3: C and B;
4. Rank 4: A and A+;
5. Rank 5: A++.
Buildings 2024, 14, 1696 17 of 36

2.2.11. Heating and Ventilation

The ranking of ventilation is based on the type of the ventilation being used in the project.
1. Natural Ventilation System: This system relies on passive air flow through doors,
windows, and other openings. It does not consume energy for operation but lacks
control over air quality and can result in energy losses due to unregulated thermal
exchange with the outdoor environment. A typical example might have manually
operated windows and vents without any mechanical ventilation components.
2. Standard Mechanical Ventilation System: This system mechanically brings fresh
air into the building and expels stale air, offering a more controlled ventilation than
natural ventilation. However, it lacks energy efficiency as it does not recover heat
from the outgoing air. The system might include fans for supply and removal, with
simple filters and basic control mechanisms.
3. Improved Standard Mechanical Ventilation System: This is an upgraded version of
the standard mechanical ventilation system that includes a heat recovery unit with a
moderately efficient heat exchanger, an electric heater, fans, and higher-grade filters.
This system improves energy efficiency by pre-heating the incoming air with the
outgoing air. It may also have a basic system for managing combustion air, with static
controls for various components.
4. Advanced Mechanical Ventilation System with Heat Recovery: This system includes
a high-efficiency heat recovery unit, more energy-efficient fans, and advanced filters.
It offers improved control mechanisms, such as demand-controlled ventilation based
on CO2 levels or occupancy sensors. It also provides more reliable combustion air
management, eliminating the need to manually open windows or other openings. This
system represents a significant upgrade in energy efficiency and control compared to
the previous system.
5. State-of-the-Art Mechanical Ventilation System with Energy Recovery: This system
not only recovers heat but also humidity, providing additional energy savings and
improved indoor air quality in climates where humidity control is a concern. It uses
high-performance components throughout and employs advanced control systems
that dynamically adjust ventilation based on various factors. For example, it may
include a very high-efficiency energy recovery unit, variable speed fans, superior
filters, and smart controls that adapt to changes in indoor and outdoor conditions.
The systems are ranked from 1 (least efficient) to 5 (most efficient). The third system is
a moderately upgraded version of a standard mechanical ventilation system and is roughly
comparable in efficiency to the system we provided, assuming some improvements in heat
recovery and control systems.

2.2.12. Insolation and Light

The ranks for insolation and light are the following:
1. Rank 1—Poor Quality:
• Traditional Building: Natural light ratio <1:10, Artificial lighting <50 lx.
• Modular Building: Natural light ratio <1:10, Artificial lighting <50 lx.
2. Rank 2—Basic Quality:
• Traditional Building: Natural light ratio 1:10, Artificial lighting 50–74 lx.
• Modular Building: Natural light ratio 1:10, Artificial lighting 50–74 lx.
3. Rank 3—Moderate Quality:
• Traditional Building: Natural light ratio 1:8, Artificial lighting 75–99 lx.
• Modular Building: Natural light ratio 1:8, Artificial lighting 75–99 lx.
4. Rank 4—Good Quality:
• Traditional Building: Natural light ratio 1:6, Artificial lighting 100–150 lx.
• Modular Building: Natural light ratio 1:6, Artificial lighting 100–150 lx.
Buildings 2024, 14, 1696 18 of 36

5. Rank 5—Excellent Quality:

• Traditional Building: Natural light ratio >1:6, Artificial lighting >150 lx.
• Modular Building: Natural light ratio >1:6, Artificial lighting >150 lx.

2.2.13. Sound Isolation

The rankings for sound isolation quality in buildings (for both traditional and modular
buildings) are the following:
1. Rank 1—Minimal Sound Isolation: Traditional/Modular Building: Sound Reduction
Index (SRI) <30 dB;
2. Rank 2—Basic Sound Isolation: Traditional/Modular Building: Sound Reduction
Index (SRI) 30–39 dB;
3. Rank 3—Moderate Sound Isolation: Traditional/Modular Building: Sound Reduc-
tion Index (SRI) 40–49 dB;
4. Rank 4—Good Sound Isolation: Traditional/Modular Building: Sound Reduction
Index (SRI) 50–60 dB;
5. Rank 5—Excellent Sound Isolation: Traditional/Modular Building: Sound Reduction
Index (SRI) >60 dB.

2.2.14. Security
In context of residential modular houses, security refers to protection against burglary
and theft. The system provided to ChatGPT as an example involved GSM/GPRS connec-
tivity, sound alarms, text notifications, app-based controls, zone-specific sensors, magnetic
contacts, and motion/glass impact detectors. It showed a high level of sophistication and
comprehensive coverage; therefore, the system was ranked as a 4 on a scale from 1 to 5.
Here is an overview of security systems for modular buildings from the simplest to the
most advanced:
1. Basic Security System: This system consists of minimal intrusion detection compo-
nents, which could include simple door and window contacts, along with a couple
of motion detectors. There is a basic control panel with a local siren for immediate
vicinity alerts, but the system lacks any form of remote monitoring or notification.
2. Intermediate Security System: This system upgrades the basic system by integrating
basic remote monitoring functionality through a wired telephone connection or simple
internet connection. This could include door and window contacts, multiple motion
detectors, and a control panel with a basic digital display. Notifications could be
achieved through dial-out phone calls or simple text messages to predefined numbers.
3. Advanced Security System: This system encompasses a wider range of detection
methods such as door and window contacts, multiple motion detectors, and glass
break detectors. It provides remote monitoring functionality through a Wi-Fi con-
nection, which allows the user to monitor and control the system with a smartphone
application. Automated notifications are provided through text and email alerts. The
system may also have some level of integration with other smart house systems.
4. Comprehensive Security System: Building upon the advanced system, this security
system includes additional intrusion detection components such as video surveillance
and biometric access controls. It provides remote monitoring through both Wi-Fi and
GSM/GPRS for redundancy and notifications, including voice calls, text messages,
and emails. The system has a high level of integration with a smart house automation
system and includes advanced features like timed arming and disarming, geofencing,
and customized notifications.
5. State-of-the-Art Security System: The most comprehensive security system includes
all features from previous systems but adds more sophisticated measures. It might
include facial recognition, artificial intelligence for threat detection, smart locks, and
full integration with a house automation system. The system may provide cloud
storage for video surveillance footage, advanced tampering detection measures, and a
Buildings 2024, 14, 1696 19 of 36

backup power supply. The notifications could include a real-time video feed to mobile
devices, and multi-factor authentication could be used for system access.

3. The Application of the Methodology

In this section, we describe the main methods that were applied to the case study of
a project evaluation in Lithuania. The methodology applied in this study consists of the
following main steps:
1. Establishment of Factors: In utilizing the Delphi method, factors related to the
successful implementation and efficient operation of modular houses were identified.
In using the Delphi method, the evaluation was usually performed using two rounds.
The list of factors used in different rounds is provided in Tables 6 and 7. During
the first round, an evaluated list of these factors, prioritized based on relevance and
impact, was compiled.
2. Selection of Evaluable Factors: From the compiled list, factors suitable for evaluation
using a scoring system were selected. We used a threshold value of 3.3 to separate
the representative factors for the second round according to the technique presented
in [47], i.e., all factors with scores (obtained according to [48]) that are greater or equal
than the threshold value 3.3 were passed to the second round (see Tables 6 and 7).
Thirty respondents took part in the first and second Delphi rounds, which is the
recommended number of respondents [49].
3. Determination and Interpretation of Scale Values: The interpretation of values for the
scale of factors was performed with the assistance of ChatGPT. The process involved
describing each factor using the prompt-and-response method, which facilitated the
assignment of a numerical value to represent its importance. This approach allowed
for consistent evaluation across different aspects of modular housing implementation.
Each value was evaluated to understand its significance in the context of modular
housing. The interpretations provided insights into their collective contribution to the
project’s success.
4. Analysis of ChatGPT Method: The application of the obtained scale to specific
example let us evaluate the advantages and disadvantages of using the ChatGPT
method. The conclusions highlight the method’s strengths in handling complex
evaluations and identify areas for potential improvement or caution. The findings are
succinctly presented, offering valuable insights into the application of ChatGPT in
research methodologies.
In Tables 6 and 7, the ‘+’ and ‘−’ symbols denote whether the according factor was
included or not, respectively, in the survey in the corresponding round of the Delphi method.
As it can be seen from these tables, many factors were excluded from the final round because
of the lack of interest from the respondents. As it can be clearly seen, the sustainability-
related factors were among the excluded ones; the users tended to seek the best ratio
of price and benefit for themselves. Perhaps including some information about some
attractive conditions, e.g., stimulating toward sustainable solutions, would make some
positive shifts in the attractiveness of such solutions. The inclusion of such information
was not considered in the current research; however, the current result reveals the problem
of the priorities of the respondents participating in the survey, i.e., the sustainability factors
have low importance. Another important note, sustainability increasing solution support is
often implemented through requirements for projecting companies rather than end users,
e.g., as it is performed with newer requirements for residential buildings in Lithuania. The
energy performance class of new buildings must be not less than A++ if the construction
works will start after the 2021 1st of January [50]. Additional statistics are presented—the
means of evaluations were calculated, and the standard deviation SD was determined.
Buildings 2024, 14, 1696 20 of 36

Table 6. Factor presence in different rounds part 1.

Round 1 Round 2
Factor Factor
Factor Code Mean SD Score Mean SD
Presence Presence
B1.1. 4.2 0.925 3.693 + 4.233 0.774 +
B1.2. 4.267 0.907 3.716 + 4.1 0.885 +
B1.3. 4.833 0.379 4.606 + 4.3 0.837 +
B1.4. 4.767 0.43 4.582 + 4 0.91 +
B1.5. 4.267 1.015 3.362 + 4.267 0.868 +
B1.6. 4.133 1.137 3.305 + 4.367 0.718 +
B1.7. 4.6 0.498 4.524 + 4.3 0.794 +
B1.8. 3.53 0.973 3.466 + 4.167 0.874 +
B1.9. - - - − 3.933 0.828 +
B1.10. 3.667 1.155 3.141 + - - −
B1.11. 4 0.983 3.28 + - - −
B1.12. 2.867 1.196 2.866 + - - −
B1.13. 4.067 1.048 3.291 + - - −
B1.14. - - - - - - −
B2.1. 3.733 0.944 3.536 + 3.833 0.986 +
B2.2. 3.933 1.048 3.591 + 3.533 1.167 +
B2.3. 4.633 0.615 4.196 + 4.133 0.9 +
B2.4. 4.5 0.938 3.446 + 4.567 0.817 +
B2.5. 4.433 0.858 3.431 + 4.4 0.86 +
B2.6. 4.067 0.868 3.656 + 4.1 0.96 +
B2.7. 4.333 0.922 4.073 + 3.867 1.196 +
B2.8. 4.4 0.77 3.773 + 3.967 1.033 +
B2.9. - - - − 4.3 0.837 +
B2.10. 3.867 0.937 3.237 + - - −
B2.11. 3.6 1.07 3.137 + - - −

Table 7. Factor presence in different rounds part 2.

Round 1 Round 2
Factor Factor
Factor Code Mean SD Score Mean SD
Presence Presence
B3.1. 3.933 0.907 3.609 + 3.667 1.269 +
B3.2. - - - − 4.367 1.066 +
B3.3. - - - − 3.8 1.157 +
B3.4. 3.867 1.008 3.232 + - - −
B3.5. 3 1.114 2.914 + - - −
B3.6. 3.6 0.969 3.144 + - - −
B4.1. 4.1 0.845 4.004 + 3.7 1.208 +
B4.2. 4.333 0.711 4.091 + 3.933 1.081 +
B4.3. 3.367 1.033 3.056 + - - −
B4.4. 4 1.145 3.25 + - - −
B4.5. 2.533 1.196 2.757 + - - −
B4.6. 3.6 1.07 3.137 + - - −
B4.7. 3.533 1.106 3.109 + - - −
B5.1. 3.733 1.173 3.161 + - - −
B5.2. 3.567 1.073 3.116 + - - −
B6.1. 2.833 1.289 2.829 + - - −
B6.2. 3.9 1.213 3.212 + - - −
B6.3. 3.733 1.081 3.172 + - - −
B6.4. 3.233 1.357 2.963 + - - −
B6.5. 3.6 1.249 3.1 + - - −
B6.6. 3.3 1.236 3.005 + - - −
B6.7. 3.033 1.326 2.893 + - - −
CB5. 4.433 0.679 4.126 + 4.433 0.728 +
CB6. 4.267 1.015 3.362 + 4.067 0.868 +
Buildings 2024, 14, 1696 21 of 36

Workflow Diagram
The whole evaluation process can be summarized in a workflow diagram. In this
section, we provide the workflow diagram of the proposed evaluation process. It is
presented in Figure 3. Detailed descriptions of the steps are as follows:
1. Start. The beginning of the evaluation process.
2. Initial Data Collection. Collect practical examples of modular buildings to serve as
the initial dataset.
3. Application of the Delphi Method. Engage a panel of experts to apply the Delphi
method, iterating through rounds to refine and determine key criteria and their
relative weights for evaluating modular buildings.
4. ChatGPT Involvement for Data Expansion and Analysis.
• Provide ChatGPT with a representative set of sample data from the collected
examples.
• ChatGPT uses this dataset to generate additional contextual information and
data aligned with the refined criteria from the Delphi method.
5. Evaluation Scale Development Using ChatGPT. Develop an evaluation scale based
on the information generated by ChatGPT, in alignment with the criteria and weights
established through the Delphi method.
6. Expert Verification of the Evaluation Scale.
• An expert verifies the evaluation scale to ensure its accuracy and validity.
• If serious flaws are detected, the process returns to the ChatGPT Data Expansion
and Analysis stage (Step 3), including the description of the flaws for ChatGPT
to address.
7. Final evaluation. Conduct the final evaluation of modular buildings using the devel-
Buildings 2024, 14, 1696 oped and verified evaluation scale. 22 of 37
8. End. The conclusion of the evaluation process.

Start

Initial Data
Collection

Delphi Method
Application

ChatGPT Data
Expansion Address Flaws

Evaluation Scale
Development

Expert Flaws found

Verifica-
tion

No flaws

Final evaluation

End

3.Workflow
Figure 3.
Figure Workflowdiagram illustrating
diagram the evaluation
illustrating process. process.
the evaluation
It is important to note that, as it was mentioned before, the initial scale for ranking,
suggested by ChatGPT, was confirmed by the expert without any modifications. It was
acknowledged as adequate and accurate for fulfilling the goals of this study.

4. Case Study: The Modelling of Investor’s Behaviour

The Description of the Case
Buildings 2024, 14, 1696 22 of 36

It is important to note that, as it was mentioned before, the initial scale for ranking,
suggested by ChatGPT, was confirmed by the expert without any modifications. It was
acknowledged as adequate and accurate for fulfilling the goals of this study.

4. Case Study: The Modelling of Investor’s Behaviour

The Description of the Case
In this section, we briefly introduce the case of a project evaluation in Lithuania and
the application of the previously described methodology to this case.
It depends on concrete method and objectives how these estimates will be used. The
simple approach is to compute the sum of the estimates in order to obtained the project
estimation; in the provided case, the sum is equal to 403. However, the interpretation of
this result is out of the scope of the current research; it usually depends on the context of
the data being analyzed.
In using the Delphi method, certain factors were evaluated based on the ranking
description provided by ChatGPT, while others were assessed based on the opinions of
customers and experts. Table 8 presents the distribution of factor evaluators; most factors
were evaluated using the scale that was provided by ChatGPT, and another approach was
used in cases where objectiveness of opinion was necessary. In other words, the whole
evaluation procedure is simplified, leaving only necessary factors for the user to evaluate.
It is important to note that the data were limited to a single detailed project example; thus,
the one-shot learning technique was applied.

Table 8. The approach for the evaluation of different factors.

Factor Group Number and Name Factor ID The Approach for Factor Evaluation
B1.1 ChatGPT
B1.2 ChatGPT
B1.3 ChatGPT
B1.4 ChatGPT
B1. Economic factors that are related to the construction and efficient operation of a modular
house B1.5 ChatGPT
B1.6 ChatGPT
B1.7 ChatGPT
B1.8 ChatGPT
B1.9 ChatGPT
B2.1 Customer’s opinion
B2.2 Customer’s opinion
B2.3 Customer’s opinion
B2.4 ChatGPT
B2. Factors that are related to the architectural implementation and efficient operation of a
modular house B2.5 ChatGPT
B2.6 ChatGPT
B2.7 ChatGPT
B2.8 Customer’s opinion
B2.9 Customer’s opinion
B3.1 Customer’s opinion
B3. Factors that are related to the psychological and social needs of a modular house resident B3.2 ChatGPT
B3.3 Customer’s opinion
B4.1 Customer’s opinion
B4. Factors that are related to the location of a modular house
B4.2 Customer’s opinion
CB5. Preliminary price of a modular house CB5 Customer’s opinion
CB6. Compliance of the modular house with norms and standards CB6 Expert’s opinion
Buildings 2024, 14, 1696 23 of 36

A one-story, 60-square-meter modular house project was selected as the case study.
This project accounts for investments in not only the house, its construction, and furnishing
but also the purchase of the plot of land for construction.
The evaluation results for this illustrative project are shown in Table 9. The score
of each factor is multiplied by its weight. The weighted scores are provided in the last
column of Table 9. This evaluation method is beneficial for comparing individual projects
to determine the best option for the customer.

Table 9. The factor evaluation.

Estimation
Weight
Factor

Score
Data for Evaluation
B1.1 Up to 8 weeks 4 4.23 16.93
B1.2 EUR 1000 5 4.10 20.50
B1.3 EUR 12,000 4 4.30 17.20
B1.4 100 years 4 4.00 16.00
B1.5 Total USD 169,430, including additional investment in the amount of USD 117,590 5 4.27 21.33
B1.6 5 years 2 4.37 8.73
B1.7 USD 3240 per year (including the depreciation of the house, land taxes, and utility bills) 2 4.30 8.60
B1.8 It is possible to adapt the house to the most diverse needs of the residents 3 4.17 12.50
B1.9 The house is carefully designed, manufactured by certified professionals using certified materials 3 3.93 11.80
B2.1 One-story house, 60 m², with external dimensions 9735 × 6080 m and 2480 mm high 4 3.83 15.33
B2.2 The house is finished in natural wood, and the roof is covered with profiled sheet 4 3.53 14.13
B2.3 The functionality of the house is good 4 4.13 16.53
B2.4 A++ class 5 4.57 22.83
B2.5 A heat pump with a boiler and house design ensures even heating and efficient ventilation 4 4.40 17.60
B2.6 Up to 150 lx lighting in the living rooms and kitchen, up to 300 lx in the bathroom, corridors, and auxiliary rooms 4 4.10 16.40
B2.7 C class (55–60 dB.) 4 3.87 15.47
B2.8 All the necessary amenities and communications are connected 5 3.97 19.83
B2.9 The price includes the elements of the house, the value-added tax, delivery to the location, assembly, and connections 4 4.30 17.20
B3.1 The location of the house is familiar to the customer, and the environment is suitable for living 5 3.67 18.33
B3.2 Traditional protection with sensors 1 4.37 4.37
B3.3 The plot is located in an array of gardens, surrounded by greenery, and the plot is landscaped 5 3.80 19.00
B4.1 The plot of the house is located in an array of gardens, the environment is clean, and the neighbors are familiar 5 3.70 18.50
B4.2 The place is easily accessible by car, but the public transport connection is poor 4 3.93 15.73
USD 51,840 (The price includes the elements of the house, the value-added tax, delivery to the location, assembly, and
CB5 4 4.43 17.73
connections)
The evaluated house was designed and built by certified professionals, in accordance with the requirements of the
CB6 5 4.07 20.33
Eurocodes, using certified materials

Several key considerations are highlighted in this evaluation as follows:

1. The scores for certain factors were assigned based on the subjective judgment of the
evaluator, as the evaluation of these factors is inherently subjective. Specifically, these
factors include B2.1, B2.2, B2.8, B2.9, B3.1, B3.3, B4.1, B4.2, CB5, and CB6.
2. Evaluating a single project in isolation does not suffice for a comprehensive final
evaluation. A comparative analysis involving multiple projects is required for such
an evaluation. Consequently, the final evaluation should be contextualized within
a decision-making framework, although this aspect falls beyond the scope of the
current research.
3. The evaluation of factor CB6 demands specific expertise, necessitating the involvement
of an expert in some cases. In this particular instance, the evaluator possessed the
necessary competence to assess this factor independently.
4. For the remaining factors, evaluator bias is significantly reduced, thanks to the imple-
mentation of a standardized scale, which is an important achievement and a notable
contribution of our publication. This scale, meticulously developed with the assis-
tance of ChatGPT, ensures a more objective and consistent evaluation process across
different evaluators and contexts.
Buildings 2024, 14, 1696 24 of 36

5. Discussion
In examining the customer priorities for modular houses, it is evident that sustainability-
related factors that directly impact the house’s operating economics and living conditions,
such as energy efficiency class, are given precedence. This focus reflects a growing con-
sumer awareness of the immediate benefits of sustainable features, particularly in terms of
cost savings and enhanced living comfort. As our research emphasizes the importance of
sustainability and efficiency in modular construction, the findings from a study on Building
Information Modeling (BIM) [51] become particularly relevant. This study identified key
drivers for BIM adoption that parallel our identified factors for modular building selec-
tion, highlighting innovation in construction practices as critical for both fields (Modular
Construction and BIM).
Our findings on modular construction, emphasizing efficiency and sustainability,
complement recent research that identifies critical challenges in BIM implementation. The
solutions and methodologies we propose could be instrumental in addressing these BIM-
related challenges, offering valuable insights for their resolution [52]. However, it is crucial
to recognize that factors indirectly related to customer needs, like carbon emissions during
the production, transportation, and construction phases, tend to be less emphasized. This
oversight presents a significant challenge in the sustainable development of the modular
housing sector. This suggests a need for manufacturers, transport companies, and builders
to proactively address these broader environmental impacts, even if they are not top-of-
mind for consumers. Their involvement is vital in ensuring the overall sustainability of the
sector, extending beyond immediate customer concerns.
The divergent views of experts, customers, and other stakeholders in modular housing
also pose challenges. Resolving these discrepancies requires a strategic approach to find
common ground and compromises. In this context, tools like ChatGPT can be instrumental.
By providing a wide array of information and facilitating dialogue, ChatGPT can aid in
resolving conflicts and support more sustainable decision-making processes.
In addressing consumer behavior, it is observed that while consumers recognize the
importance of sustainable practices and often expect them to be standard, this does not
necessarily translate into brand loyalty or a preference for companies with the lowest envi-
ronmental impact. This contradiction, highlighted in [53], reveals a gap between consumer
ideals and their willingness to incur additional costs for sustainability. This paradox in
consumer behavior underscores the complexity of integrating sustainability into business
models and the need for innovative strategies that align consumer expectations with sus-
tainable practices without imposing prohibitive costs. Addressing this paradox is beyond
the scope of our current study but represents a critical avenue for future research. Ex-
panding on strategies to bridge this gap—whether through targeted educational programs,
incentives for sustainable living, or enhanced marketing strategies that better communicate
the long-term benefits and cost savings of sustainable features—could provide valuable in-
sights. This gap underscores the need for a multidisciplinary approach, combining insights
from environmental psychology, marketing, and construction management to develop
effective strategies that align consumer behavior with sustainable construction practices.
We leave this as an important area for further exploration, urging researchers to delve
into the complexities of consumer decision-making processes to foster a more sustainable
modular housing market.
In our investigation into modular construction, we encountered the need for op-
timization in project management, a theme also central to a recent study on time–cost
optimization in construction projects. This study introduced a modified dynamic opposi-
tional learning TLBO (MDOLTLBO) algorithm to enhance the efficiency of finding optimal
solutions [54]. The innovative approach taken to refine the population initialization and
search space in this study parallels our efforts in developing an evaluation scale for modular
building projects, where optimizing decision-making processes is key. The application of
such advanced optimization techniques in both contexts underscores their importance in
streamlining complex construction projects.
Buildings 2024, 14, 1696 25 of 36

In this paper, we aim to provide a comprehensive overview of the advantages and dis-
advantages of using ChatGPT, drawing from the recent literature to highlight its potential
benefits, challenges, and areas for future exploration.
Advantages of using ChatGPT for building project evaluation include the following:
1. Broad Information Access: ChatGPT can draw upon a vast range of information
pertaining to the factors being evaluated. This allows for comprehensive insights
and analyses.
2. Scaled Evaluation: ChatGPT can categorize the values of the evaluated factors ac-
cording to a predefined scale, providing a detailed description for each point on the
scale. This covers the need for the expert to create a described scale suitable as a part
of the evaluation instructions.
3. Rapid Responses: Queries posed to ChatGPT are answered swiftly, streamlining the
analysis process and saving time. Thus, a similar approach might be efficient in other
similar studies as well.
4. Quality Language Output: Responses from ChatGPT are articulated in coherent
and professional language, making the information easily digestible. In this sense,
ChatGPT being a language model means that there is no drawbacks in the language
quality context.
5. Consistency: Unlike human experts, ChatGPT provides consistent answers over
time, ensuring that evaluations are not influenced by factors like fatigue, mood, or
personal bias.
Disadvantages of using ChatGPT for building project evaluation include the following:
1. Precise Prompting Required: Formulating precise prompts is essential when interact-
ing with ChatGPT. This often necessitates collecting detailed information about each
subject under investigation, i.e., expertise for correct prompts is needed, and it can be
seen as a partial replacement of expertise in the field in which the evaluation is being
conducted.
2. Accuracy Concerns: While ChatGPT is informed by a vast amount of data, its re-
sponses may occasionally be imprecise or not entirely relevant. Users may need to
spend time verifying and, if necessary, correcting these responses. An expert verified
the ChatGPT output in our case.
3. Subjective Limitations: ChatGPT is not equipped to provide answers that rely on
the subjective opinions of specific individuals, be it the customer, expert, or any
other stakeholder. For criteria that implies a subjective assessment by its nature, the
assessment must still be conducted by the customer.
4. Knowledge Cutoff: ChatGPT’s knowledge is not continuously updated. Therefore,
it may not be aware of very recent trends, studies, or best practices in the building
industry post its last training cut-off. In particular, this is important in the case of data
that are very sensitive to changes over time; thus, this did not have a big negative
impact in our case.
5. Lack of Context Awareness: Unlike human experts who can adjust their assessments
based on nuanced understanding and context, ChatGPT might not always catch subtle
project-specific intricacies. While this can be mitigated to some extent with precise
and detailed prompting, it does not entirely replace the depth of understanding that a
human expert might have.
6. Unavailability of references to the source of knowledge: It is impossible to track
the original source of different parts of ChatGPT-generated knowledge; moreover,
ChatGPT does not take responsibility for its output.
It is important to note that the knowledge that was generated by ChatGPT is not con-
sidered a unique output that is needed for research; we used it for demonstration purposes.
The main result is the proven fact of the ChatGPT-generated knowledge’s suitability to
cover (in the target domain) the needs of a complex professional evaluation procedure.
Buildings 2024, 14, 1696 26 of 36

6. Conclusions
In this research, we explored the feasibility and effectiveness of employing ChatGPT
in the evaluation of factors critical to the implementation and operation of modular houses.
Our findings reveal several key insights:
1. ChatGPT as a Tool for Scale Description: ChatGPT has proven to be an efficient tool for
articulating detailed descriptions for each score on a rating scale, significantly facilitat-
ing the scoring process for standardized evaluations. However, its utility is somewhat
limited for factors requiring subjective judgments from experts or customers.
2. This research showed the consumer behavior paradox toward sustainability, highlight-
ing the discrepancy between proclaimed sustainable values and actual purchasing
behaviors—it was clearly observed after the Delphi method application in current re-
search.
3. Precision in Prompting and Verification: The success of utilizing ChatGPT hinges on
the precision of the prompts provided and the subsequent verification of its outputs
against established knowledge. Despite ChatGPT’s capacity to provide insightful
responses, these may not fully capture the intricacies of consumer preferences and
market dynamics. Adjusting prompts to better reflect the nuanced considerations of
sustainability and cost implications in consumer choices is essential.
4. Potential for Broader Application in Factor Generation: Beyond evaluating predefined
factors, there exists an opportunity to leverage ChatGPT in generating and refining
the list of factors itself. Such an approach could offer novel insights into the paradox of
consumer behavior toward sustainability, harnessing ChatGPT’s ability to amalgamate
diverse perspectives and identify emerging trends in sustainable living practices. In
current research, the factors were curated using the Delphi method and interviews
with potential modular house users. In a broader sense, this research contributes to
the possible incorporation of modern large language models into the field of MCDM.
5. Enhancing Decision-Making Sustainability: ChatGPT’s role extends to supporting
more sustainable decision-making processes by providing stakeholders with access to
a broad spectrum of information, reflecting current opinions and trends. Its applica-
tion can facilitate the identification of optimal compromises, help bridge information
gaps, and reduce uncertainties, particularly in reconciling cost, sustainability, and con-
sumer preferences. This capability is invaluable in addressing the consumer behavior
paradox, underscoring the importance of informed and balanced decision making in
promoting sustainable modular housing solutions.
6. The evaluation scale provided by ChatGPT was based on detailed single-example data
(one-shot learning) that the large language model was fed. Unfortunately, there is no
convenient way for the user to access the exact information about data sources used
by ChatGPT. However, the validation of results by an expert in the current research
showed that the knowledge behind the output was sufficient, and only minor flaws
needed to be addressed. Thus, the main research hypothesis about ChatGPT’s ability
to efficiently solve the evaluation scale establishment problem was confirmed. The
information generated by ChatGPT itself should not be viewed as an original part of
the knowledge in this research; however, it is both for demonstration purposes and as
proof of possible suitability in the context of the methodology. In other words, it was
used to demonstrate how to use ChatGPT to simplify expertise tasks for an expert
who only needed to validate and correct the output.
7. The other important output of this research is the transparent demonstration of
prompts that led to satisfactory results.
Future Directions: The outcomes of this study highlight the significant potential for
leveraging advanced AI tools like ChatGPT in the modular construction sector and beyond.
These tools not only enhance assessment methodologies but also provide deeper insights
into consumer behaviors and sustainability preferences. Future research should aim to
develop AI-driven tools that are specifically designed for the broader construction industry.
Buildings 2024, 14, 1696 27 of 36

These tools should be adept at navigating the intricate dynamics of consumer decision
making and the multifaceted aspects of sustainability. Pursuing such advancements is
crucial for closing the current gaps and promoting a more sustainable framework in
modular housing development and other construction domains.

Author Contributions: Conceptualization, A.M. and O.R.Š.; data curation, A.M. and A.B.; formal
analysis, A.B.; investigation, O.R.Š.; methodology, A.M., A.B. and O.R.Š.; resources, A.M. and A.B.;
software, A.B.; supervision, O.R.Š.; validation, A.M. and A.B.; visualization, O.R.Š.; writing—original
draft, A.M.; writing—review and editing, A.B. All authors have read and agreed to the published
version of the manuscript.
Funding: This research received no external funding.
Data Availability Statement: Data were obtained through paper questionnaires.
Acknowledgments: Authors acknowledge the use of ChatGPT 4 (Open AI, https://chat.openai.com)
to generate prompts which were provided in the article, to generate an illustrative figure and to
proofread the final draft.
Conflicts of Interest: The authors declare no conflicts of interest.

Appendix A. The Conversation with ChatGPT

Appendix A.1
Here, we provide an exported conversation with ChatGPT [5] as an important part of
our investigation.
Prompt:
This is a ventilation system for modular building: “A separate mechanical ventilation
system is designed in the house. Ventilation unit with heat recovery (rotary heat exchanger
ef.85%), electric heater (1.0 kW 1), fans (2 × 0.2 kW 1), filters (ePM2.5 65%) for supply
(ePM10 50%) for removal and control automation kit. The ventilation unit is provided in the
technical room of the existing building. In the rooms, the air is distributed through round
galvanized sheet ducts and is supplied to the rooms and removed through the supply and
removal ceilings. Air flow control flaps are installed in the required places. Duct tightness
class not lower than B. Air intake and exhaust ducts are insulated with thermal insulation.
During the use of the fireplace, when additional air supply is required for combustion,
the vent chamber reduces to compensate for the air supply air flow 80% (up to 384 m3 /h),
exhaust air flow 20% (up to 96 m3 /h), 288 m3 /h remains for air compensation. If the
amount of air for combustion support is insufficient, the window is opened. For the hood
in the kitchen, the duct from the hood goes to the exhaust zone of the vent chamber AHU-1,
bypassing the chamber elements, air removal 450 m3 /h, without speed adjustment. An air
flow closing-opening valve is installed on the duct in the hood, which must open when the
hood is running and close when the hood is off”, please rank its’ efficiency from 1 to 5 (from
worst to best) and provide alternative systems with the rest of 4 ranks. The answer’s result
should be the list of 5 ranked and named systems along with their description. Provide the
answer in LaTeX.
ChatGPT:
Given that this model has a knowledge cutoff in September 2021 and the performance
of ventilation systems greatly depends on factors such as the exact layout and usage of
the building, this will be a hypothetical ranking based on generalized understanding of
such systems and their efficiency as of the last update. However, before the ranking, let’s
analyse the provided system.
The provided system includes a heat recovery ventilation (HRV) unit, which is ex-
cellent for energy efficiency, and the 85% efficiency of the rotary heat exchanger is quite
good. However, the reliance on opening a window if the combustion support is insufficient
may lead to energy losses and impact overall performance. Similarly, the absence of speed
adjustment for the kitchen hood exhaust could limit the system’s adaptability to different
conditions. Let’s assign this system a rank of 3 for now.
Buildings 2024, 14, 1696 28 of 36

Now, let’s fill the other four ranks with some generalized system types. Please note,
this is a simplified representation and the efficiency of actual systems would depend on
numerous specific factors, including but not limited to their precise configuration, usage
patterns, the specific building in question, and local climate conditions.
1. Natural Ventilation System: This is the most basic form of ventilation that relies on the
natural flow of air through windows, doors, vents, and other openings. While it has
no energy costs for operation and can be adequate for some settings, its performance
can be inconsistent and it is generally less efficient than mechanical systems in terms
of air quality control and energy management.
2. Standard Mechanical Ventilation System (without heat recovery): This system
mechanically brings fresh air into a building and expels stale air. While it provides
more consistent air exchange than a natural ventilation system, it can lead to significant
energy losses as the incoming air must be heated or cooled to match the desired
indoor temperature.
3. Mechanical Ventilation System with Heat Recovery (as provided): The system in
question fits here. The addition of a heat recovery unit greatly improves energy
efficiency by pre-heating (or pre-cooling) the incoming air with the outgoing air.
However, the specific design provided has a few limitations as mentioned above.
4. Improved Mechanical Ventilation System with Heat Recovery: This is similar to
the provided system but with a few improvements. It could include speed adjust-
ment for the kitchen hood exhaust, a more sophisticated system for managing com-
bustion air that doesn’t rely on opening a window, and perhaps some additional
energy-saving measures such as demand-controlled ventilation based on CO2 levels
or occupancy sensors.
5. State-of-the-art Mechanical Ventilation System with Energy Recovery: This goes
one step beyond the previous category by not only recovering heat, but also humidity.
An energy recovery ventilation (ERV) system can transfer moisture as well as heat
between the outgoing and incoming air streams, which can provide additional en-
ergy savings and improve indoor air quality in climates where humidity control is
a concern.
Again, please note that these are very generalized descriptions and real-world perfor-
mance can vary significantly based on many factors.
Prompt:
You said “(as provided): The system in question fits here”. In previous answer,
however, I need full answer with all 5 ranks and full description of third rank in a style as
it was not mentioned by me before.
ChatGPT:
I see, you want a reworked version of the given system ranked at 3, with new descrip-
tions for all five systems. Please note that the specific efficiency of each system can vary
widely based on the exact configuration and use conditions. However, here’s a hypothetical
ranking of ventilation systems for modular buildings:
1. Natural Ventilation System: This system relies on passive air flow through doors,
windows, and other openings. It does not consume energy for operation but lacks
control over air quality and can result in energy losses due to unregulated thermal
exchange with the outdoor environment. A typical example might have manually-
operated windows and vents without any mechanical ventilation components.
2. Standard Mechanical Ventilation System: This system mechanically brings fresh
air into the building and expels stale air, offering a more controlled ventilation than
natural ventilation. However, it lacks energy efficiency as it does not recover heat
from the outgoing air. The system might include fans for supply and removal, with
simple filters and basic control mechanisms.
3. Improved Standard Mechanical Ventilation System: An upgraded version of the
standard mechanical ventilation system that includes a heat recovery unit with a
Buildings 2024, 14, 1696 29 of 36

moderately efficient heat exchanger, an electric heater, fans, and higher-grade filters.
This system improves energy efficiency by pre-heating the incoming air with the
outgoing air. It may also have a basic system for managing combustion air, with static
controls for various components.
4. Advanced Mechanical Ventilation System with Heat Recovery: This system includes
a high-efficiency heat recovery unit, more energy-efficient fans, and advanced filters.
It offers improved control mechanisms, such as demand-controlled ventilation based
on CO2 levels or occupancy sensors. It also provides more reliable combustion air
management, eliminating the need to manually open windows or other openings. This
system represents a significant upgrade in energy efficiency and control compared to
the previous system.
5. State-of-the-Art Mechanical Ventilation System with Energy Recovery: This system
not only recovers heat but also humidity, providing additional energy savings and
improved indoor air quality in climates where humidity control is a concern. It uses
high-performance components throughout and employs advanced control systems
that dynamically adjust ventilation based on various factors. For example, it may
include a very high-efficiency energy recovery unit, variable speed fans, superior
filters, and smart controls that adapt to changes in indoor and outdoor conditions.
The systems are ranked from 1 (least efficient) to 5 (most efficient). The third system is
a moderately upgraded version of a standard mechanical ventilation system and is roughly
comparable in efficiency to the system you provided, assuming some improvements in
heat recovery and control systems.
Prompt:
Consider this modular building security system:“The security alarm system is de-
signed to transmit burglar alarm, fire alarm and system technical signals to the keyboard
and/or to the user’s phone with the help of the GSM/GPSRS module and/or to the manu-
facturer’s software installed on the PC. In any case, a warning about unauthorized access
to the protected premises or a violation of the system is announced by sound sirens and
text messages to the liquid crystal display on the control panel - keyboard. The location
of the control panel is selected in the technical room and protected so that it cannot be
accessed without triggering an alarm signal when the security system is in storage mode.
To control the system (turn it on/off), a control panel—keyboard—is provided at the main
entrance to the premises. Information about the protected zones and the current state of the
system can be seen on the display screen of the remote control and in the manufacturer’s
app, which is available on both mobile devices and PCs. System elements are protected
against unauthorized opening/removal 24 h a day. Security alarm sensors are designed in
such a way that there are only sensors of the same room in one zone. Magnetic contacts are
designed for window and door protection. Combined motion/glass impact sensors are
designed to protect the volume of the premises and to protect the windows from breaking.
An outdoor siren is provided for the security system in the facade part of the building, and
an indoor siren is provided for the keyboard. Wireless motion detectors compatible (of
the same manufacturer) with the ESIM384 centrale are designed in the auxiliary building”,
please rank its’ efficiency from 1 to 5 (from worst to best) and provide alternative systems
with the rest of 4 ranks. The results should be the list of 5 ranked and named systems along
with their description. I need full answer with all 5 ranks and full description in a style as
it was not mentioned by me before. Provide the answer in LaTeX.
ChatGPT:
Given that I don’t have access to real-time data, this will be a hypothetical ranking
based on generalized understanding of such systems as of the last update. The ranking will
be focused on factors such as security, redundancy, control, automation, and connectivity.
Your provided system involves GSM/GPRS connectivity, sound alarms, text notifi-
cations, app-based controls, zone-specific sensors, magnetic contacts, and motion/glass
impact detectors. It shows a high level of sophistication and comprehensive coverage,
Buildings 2024, 14, 1696 30 of 36

therefore, I’d rank this system as 4 on a scale of 1 to 5. Let’s create an overview of security
systems for modular buildings from the simplest to the most advanced.
1. Basic Security System: This system features minimal intrusion detection components,
such as simple door/window contacts and a couple of motion detectors. It lacks
advanced monitoring, notification, and control capabilities. It might include a simple
siren for local alerts but does not provide remote monitoring or control functionality.
2. Standard Security System with Basic Connectivity: An upgrade to the basic system,
this includes more sensors and introduces basic remote monitoring functionality, pos-
sibly via a phone line or a simple Internet connection. It might include door/window
contacts, several motion detectors, and a control panel with basic display capabilities.
Notifications could be limited to phone calls or simple text messages.
3. Advanced Security System with Smart Capabilities: This is a more comprehensive
system that includes a wider range of sensors (door/window, motion, glass break)
and introduces smart capabilities. It might feature Wi-Fi connectivity for remote
monitoring/control via a smartphone app, automated notifications (text/email), and
some level of integration with other smart home systems.
4. Comprehensive Security System with GSM/GPRS Connectivity (as provided):
The provided system fits this category. It includes a wide range of sensors (mag-
netic contacts, combined motion/glass impact), offers remote monitoring/control via
GSM/GPRS, and provides alerts via multiple channels (siren, text notifications). It
offers app-based control and displays system state information on a remote control. It
also features protective measures against unauthorized tampering.
5. State-of-the-Art Security System with Advanced Automation and Integration: This
system builds on the previous category by incorporating additional advanced fea-
tures. It could include advanced intrusion detection components (video surveillance,
biometric access controls), AI-based threat detection, and full integration with a smart
home automation system. It might also feature advanced tamper detection measures,
backup power supply, and multiple communication channels for redundancy (Wi-Fi,
GSM/GPRS, landline). Notifications could include voice calls, texts, emails, and
app notifications.
The systems are ranked from 1 (least efficient) to 5 (most efficient). The fourth system is
the one you provided, demonstrating a comprehensive security setup with a good balance
of intrusion detection components and communication channels.
Prompt:
Again, you said “(as provided): The provided system fits this category”. In previous
answer, however, I need full answer with all 5 ranks and full description in a style as it was
not mentioned by me before.
ChatGPT:
My apologies for misunderstanding your request earlier. Now, I understand you want
a complete ranking, where the system you described is not directly used as one of the
points of comparison but could be equivalent to one of the ranked systems in terms of
functionality. Here is a hypothetical ranking of security systems:
1. Basic Local Security System: This is a simple system with only local alarm capabilities.
It includes basic intrusion detection components like door/window contacts and a
few motion detectors. However, it lacks advanced monitoring, remote notification,
and control capabilities. The system has a local siren for alerts but does not offer
remote monitoring or control functionality.
2. Intermediate Security System with Connectivity: This upgraded system includes
more sensors and offers basic remote monitoring, likely through a phone line or
simple internet connection. It comprises door/window contacts, motion detectors,
and a control panel with basic display functions. Notifications are delivered via phone
calls or basic text messages.
Buildings 2024, 14, 1696 31 of 36

3. Advanced Security System with Internet Connectivity: A more comprehensive

system, this includes a variety of sensors, such as door/window contacts, motion
detectors, and glass break detectors. It introduces internet connectivity, allowing
remote monitoring and control through a smartphone app. Notifications are delivered
through automated text and email alerts.
4. Comprehensive Security System with GSM Connectivity: This sophisticated system
includes an array of sensors, such as magnetic contacts, combined motion/glass break
detectors, and advanced monitoring capabilities. It offers remote monitoring and
control via GSM/GPRS, with alerts delivered through multiple channels such as sirens
and text notifications. The system features app-based control and displays the system
state information on a remote screen.
5. State-of-the-Art Security System with Advanced Features: Building upon the com-
prehensive system, this state-of-the-art security setup includes additional advanced
features. These could include video surveillance, biometric access controls, artificial
intelligence-based threat detection, and integration with other smart home systems.
The system includes advanced tamper-detection measures, a backup power supply,
and redundant communication channels (Wi-Fi, GSM/GPRS, landline). Notifications
are delivered through a variety of channels, including voice calls, texts, emails, and
app alerts.
In this revised ranking, your described system would likely fall around the “Com-
prehensive Security System with GSM Connectivity” level, although the specific features
might not exactly match.
Prompt:
I need 5 intervals of building time bounds for 5 ranks (from longest to shortest) for
both: traditional and modular buildings. Provide the answer in LaTeX.
ChatGPT:
Sure, let’s discuss some general time frames for the construction of traditional and
modular buildings, ranked from longest to shortest. Keep in mind that construction times
can vary widely based on the size and complexity of the building, the number of workers,
local regulations, and many other factors. These are rough estimates and should not be
used for precise planning.

Rank Traditional Building Time (Months) Modular Building Time (Months)

1 24–36 12–18
2 18–24 9–12
3 12–18 6–9
4 9–12 4–6
5 6–9 3–4

In this table, Rank 1 represents the longest building times, while Rank 5 represents the
shortest. For both traditional and modular buildings, the time frames are reduced as we
move from Rank 1 to Rank 5. Generally speaking, modular buildings can be constructed in
a shorter time frame compared to traditional buildings due to the advantages of factory
production, concurrent site work, and reduced weather-related delays.
Prompt:
I need 5 intervals of building costs bounds for 5 ranks (from the most expensive to
cheapest) for both: traditional and modular buildings. Provide the answer in LaTeX.
ChatGPT:
Costs in construction can vary considerably depending on factors such as location,
design complexity, labor costs, materials used, the local real estate market, and more.
As such, providing specific cost ranges can be challenging and potentially misleading.
However, for the purpose of this exercise, let’s discuss relative cost levels in very general
terms for both traditional and modular buildings.
Buildings 2024, 14, 1696 32 of 36

It’s also important to note that while modular construction can offer some cost savings
due to efficiencies in the building process, the cost of the finished building can be similar to
traditional construction depending on the final design, finish choices, site work required,
and other factors. The figures below are for illustrative purposes only and represent a range
of potential costs per square foot in a hypothetical scenario.

Rank Traditional Building Cost ($/sq.ft.) Modular Building Cost ($/sq.ft.)

1 $400–$500 $350–$450
2 $300–$400 $250–$350
3 $200–$300 $150–$250
4 $100–$200 $100–$150
5 Below $100 Below $100

Please remember, the values are purely for illustrative purposes. Real costs may vary
dramatically based on the factors mentioned earlier. Modular construction can be cheaper
in some scenarios but not in all cases. It is always advisable to consult with construction
professionals to get accurate estimates for your specific project.
Prompt:
Please, consider the Euro zone, use Euros and International System of Units.
ChatGPT:
Thank you for the clarification. As per your request, here’s a hypothetical break-
down of costs in the Eurozone, in Euros per square meter. This ranking is from the most
expensive to the least. Note that these costs can vary widely depending on the factors
mentioned previously.

Rank Traditional Building Cost (€/sq.m) Modular Building Cost (€/sq.m)

1 5000–6000 4500–5500
2 4000–5000 3500–4500
3 3000–4000 2500–3500
4 2000–3000 1500–2500
5 Below 2000 Below 1500

Again, these figures are hypothetical and for illustrative purposes only, and actual
costs can vary greatly. It’s always advisable to consult with construction professionals for
precise estimates based on your specific project.
Prompt:
I need 5 intervals of price bounds for 5 ranks (from the most expensive to cheapest)
for both: traditional and residential modular buildings. Please do separate ranking for:
projecting costs, building costs, amount of investment. Consider the Euro zone, use Euros
and International System of Units. Provide the answer in LaTeX.
ChatGPT:
Given the wide range of factors that can affect the cost of construction, it’s challenging
to provide precise costs. Variables such as location, materials, labor, design complexity,
local regulations, and market conditions can significantly impact the overall cost.
However, for the sake of this exercise, let’s discuss very general potential costs for both
traditional and modular residential buildings in the Euro zone, broken down by project
costs, building costs, and total investment. These figures are hypothetical and meant for
illustrative purposes.
Please note that “projecting costs” often refers to the cost of architectural design and
engineering work, while “building costs” generally refers to the actual construction costs.
Buildings 2024, 14, 1696 33 of 36

Traditional Projecting Modular Projecting Traditional Building Modular Building Total Investment
Rank
(€/sq.m) (€/sq.m) (€/sq.m) (€/sq.m) (€/sq.m)
1 500–600 450–550 5000–6000 4500–5500 5500–6600
2 400–500 350–450 4000–5000 3500–4500 4400–5500
3 300–400 250–350 3000–4000 2500–3500 3300–4400
4 200–300 150–250 2000–3000 1500–2500 2200–3300
5 Below 200 Below 150 Below 2000 Below 1500 Below 2200

Please remember that these figures are highly estimated, and actual costs can vary
significantly. To get accurate estimates for a specific project, it’s always advisable to consult
with construction professionals.
Prompt:
Correct your answer considering this definition of amount of Investment: The amount
of investment refers to the total capital or financial resources needed for the construction
project. It includes both projecting costs and building costs, along with other related
expenses. The amount of investment is the total sum required to initiate and complete
the project, covering all the planning, design, and construction activities. It includes costs
for land acquisition, design and planning fees, construction costs, contingency funds,
financing expenses, professional fees, marketing expenses (if applicable), and any other
costs associated with the project. The amount of investment represents the overall financial
commitment required to bring the construction project to completion. Also, provide
different data for modular and traditional buildings.
ChatGPT:
Thank you for providing a more detailed definition. Based on this, let’s adjust the
potential costs for both traditional and modular residential buildings. The projections still
account for variables such as location, design complexity, materials, labor, and other costs.
The final “Amount of Investment” column now also includes hypothetical costs related to
land acquisition, professional fees, contingency funds, and other associated expenses.
Please note that these figures are hypothetical, and real costs can vary widely.

Traditional Projecting Modular Projecting Traditional Building Modular Building Total Investment
Rank
(€/sq.m) (€/sq.m) (€/sq.m) (€/sq.m) (€/sq.m)
1 500–600 450–550 5000–6000 4500–5500 6000–7000
2 400–500 350–450 4000–5000 3500–4500 5000–6000
3 300–400 250–350 3000–4000 2500–3500 4000–5000
4 200–300 150–250 2000–3000 1500–2500 3000–4000
5 Below 200 Below 150 Below 2000 Below 1500 Below 3000

As always, consult with construction professionals for precise estimates based on your
specific project.
Prompt:
Total Investment must be provided for modular and traditional buildings separately,
i.e., two columns instead of single.
ChatGPT:
I apologize for misunderstanding your request earlier. Here is the corrected table with
separate “Total Investment” columns for both traditional and modular residential buildings.
Buildings 2024, 14, 1696 34 of 36

Total Investment
Traditional Modular Projecting Traditional Modular Building Total Investment
Rank (Traditional)
Projecting (€/sq.m) (€/sq.m) Building (€/sq.m) (€/sq.m) (Modular) (€/sq.m)
(€/sq.m)
1 500–600 450–550 5000–6000 4500–5500 6000–7000 5500–6500
2 400–500 350–450 4000–5000 3500–4500 5000–6000 4500–5500
3 300–400 250–350 3000–4000 2500–3500 4000–5000 3500–4500
4 200–300 150–250 2000–3000 1500–2500 3000–4000 2500–3500
5 Below 200 Below 150 Below 2000 Below 1500 Below 3000 Below 2500

References
1. Rajanayagam, H.; Poologanathan, K.; Gatheeshgar, P.; Varelis, G.E.; Sherlock, P.; Nagaratnam, B.; Hackney, P. A-State-of-the-Art
review on modular building connections. In Structures; Elsevier: Amsterdam, The Netherlands, 2021; Volume 34, pp. 1903–1922.
[CrossRef]
2. Kamali, M.; Hewage, K. Life cycle performance of modular buildings: A critical review. Renew. Sustain. Energy Rev. 2016,
62, 1171–1183. [CrossRef]
3. Gan, V.J. BIM-based graph data model for automatic generative design of modular buildings. Autom. Constr. 2022, 134, 104062.
[CrossRef]
4. Feldmann, F.G.; Birkel, H.; Hartmann, E. Exploring barriers towards modular construction–A developer perspective using fuzzy
DEMATEL. J. Clean. Prod. 2022, 367, 133023. [CrossRef]
5. OpenAI. ChatGPT-4: Optimizing Language Models for Dialogue. Available online: https://openai.com/chatgpt/ (accessed on
31 October 2023).
6. Nallakaruppan, M.K.; Johri, I.; Somayaji, S.; Bhatia, S.; Malibari, A.A.; Alabdali, A.M. Secured MCDM Model for Crowdsource
Business Intelligence. Appl. Sci. 2023, 13, 1511. [CrossRef]
7. Golić, K.; Kosić, T.; Kosorić, V. AHP-Based Model for Energy-Sustainable Renovation of Building Envelopes: A Case Study.
Sustainability 2023, 15, 8384. [CrossRef]
8. Mutani, G.; Alehasin, M.; Usta, Y.; Fiermonte, F.; Mariano, A. Statistical Building Energy Model from Data Collection, Place-Based
Assessment to Sustainable Scenarios for the City of Milan. Sustainability 2023, 15, 14921. [CrossRef]
9. Gogate, N.G.; Shelake, A.G.; Band, P. Selection of most significant risk factors for Indian tunnel projects: An integrated
fuzzy-based MCDM approach. Int. J. Constr. Manag. 2023, 24, 161–176. [CrossRef]
10. Bošnjak, A.; Jajac, N. Determining Priorities in Infrastructure Management Using Multicriteria Decision Analysis. Sustainability
2023, 15, 14953. [CrossRef]
11. Kineber, A.F.; Ali, A.H.; Elshaboury, N.; Oke, A.E.; Arashpour, M. A multi-criteria evaluation and stationary analysis of value
management implementation barriers for sustainable residential building projects. Int. J. Constr. Manag. 2023, 24, 199–212.
[CrossRef]
12. Diwandari, S.; Sela, E.I.; Syahputra, B.E.; Parama, N.A.; Septiarini, A. The Utility of Decision Tree and Analytics Hierarchy Process
in Prioritizing of Social Aid Distribution due to Covid-19 Pandemic in Indonesia. J. ICT Res. Appl. 2023, 17, 82–97. [CrossRef]
13. He, Q.; Zhang, S.; Shang, J. Dynamic Characteristic and Parameter Analysis of a Modular Building with Suspended Floors.
Buildings 2023, 13, 7. [CrossRef]
14. Sala Benites, H.; Osmond, P.; Prasad, D. Inquiry on Perceptions and Practices of Built Environment Professionals Regarding
Regenerative and Circular Approaches. Buildings 2023, 13, 63. [CrossRef]
15. Aghabegloo, M.; Rezaie, K.; Torabi, S.A.; Khalili, S.M. A BIA-Based Quantitative Framework for Built Physical Asset Criticality
Analysis under Sustainability and Resilience. Buildings 2023, 13, 264. [CrossRef]
16. Maqsoom, A.; Umer, M.; Alaloul, W.S.; Salman, A.; Ullah, F.; Ashraf, H.; Musarat, M.A. Adopting Green Behaviors in the
Construction Sector: The Role of Behavioral Intention, Motivation, and Environmental Consciousness. Buildings 2023, 13, 1036.
[CrossRef]
17. Ma, X.; Li, X.; Yuan, H.; Huang, Z.; Zhang, T. Justifying the Effective Use of Building Information Modelling (BIM) with Business
Intelligence. Buildings 2023, 13, 87. [CrossRef]
18. Afzal, M.; Li, R.Y.M.; Ayyub, M.F.; Shoaib, M.; Bilal, M. Towards BIM-Based Sustainable Structural Design Optimization: A
Systematic Review and Industry Perspective. Sustainability 2023, 15, 15117. [CrossRef]
19. Cândido, L.F.; Lazaro, J.C.; Freitas e Silva, A.O.d.; Barros Neto, J.d.P. Sustainability Transitions in the Construction Sector: A
Bibliometric Review. Sustainability 2023, 15, 12814. [CrossRef]
20. Kamali, M.; Hewage, K.; Sadiq, R. Conventional versus modular construction methods: A comparative cradle-to-gate LCA for
residential buildings. Energy Build. 2019, 204, 109479. [CrossRef]
21. Kamali, M.; Hewage, K.; Sadiq, R. Economic sustainability benchmarking of modular homes: A life cycle thinking approach. J.
Clean. Prod. 2022, 348, 131290. [CrossRef]
22. Saka, A.; Taiwo, R.; Saka, N.; Salami, B.A.; Ajayi, S.; Akande, K.; Kazemi, H. GPT models in construction industry: Opportunities,
limitations, and a use case validation. Dev. Built Environ. 2024, 17, 100300. [CrossRef]
Buildings 2024, 14, 1696 35 of 36

23. Prieto, S.A.; Mengiste, E.T.; García de Soto, B. Investigating the Use of ChatGPT for the Scheduling of Construction Projects.
Buildings 2023, 13, 857. [CrossRef]
24. Kamali, M.; Hewage, K.; Milani, A.S. Life cycle sustainability performance assessment framework for residential modular
buildings: Aggregated sustainability indices. Build. Environ. 2018, 138, 21–41. [CrossRef]
25. Sadeghi, M.; Naghedi, R.; Behzadian, K.; Shamshirgaran, A.; Tabrizi, M.R.; Maknoon, R. Customisation of green buildings
assessment tools based on climatic zoning and experts judgement using K-means clustering and fuzzy AHP. Build. Environ. 2022,
223, 109473. [CrossRef]
26. Dalalah, D.; Dalalah, O.M. The false positives and false negatives of generative AI detection tools in education and academic
research: The case of ChatGPT. Int. J. Manag. Educ. 2023, 21, 100822. [CrossRef]
27. Gill, S.S.; Kaur, R. ChatGPT: Vision and challenges. Internet Things -Cyber-Phys. Syst. 2023, 3, 262–271. [CrossRef]
28. Ray, P.P. ChatGPT: A comprehensive review on background, applications, key challenges, bias, ethics, limitations and future
scope. Internet Things -Cyber-Phys. Syst. 2023, 3, 121–154. [CrossRef]
29. Joksimovic, S.; Ifenthaler, D.; Marrone, R.; De Laat, M.; Siemens, G. Opportunities of artificial intelligence for supporting complex
problem-solving: Findings from a scoping review. Comput. Educ. Artif. Intell. 2023, 4, 100138. [CrossRef]
30. Makuteniene, D.; Sostak, O.R.; Maceika, A. Decision Support System for Roof Installation. Stud. Inform. Control 2016, 25, 163–172.
[CrossRef]
31. FasterCapital. Chatgpt and Construction Industry Improving Project Management and Efficiency. Available online: https:
//fastercapital.com/content/Chatgpt-and-construction-industry-improving-project-management-and-efficiency.html (accessed
on 2 February 2024).
32. Kuzman, M.K.; Grošelj, P.; Ayrilmis, N.; Zbašnik-Senegačnik, M. Comparison of passive house construction types using analytic
hierarchy process. Energy Build. 2013, 64, 258–263. [CrossRef]
33. Bhandari, S.; Riggio, M.; Jahedi, S.; Fischer, E.C.; Muszynski, L.; Luo, Z. A review of modular cross laminated timber construction:
Implications for temporary housing in seismic areas. J. Build. Eng. 2023, 63, 105485. [CrossRef]
34. Maceika, A.; Bugajev, A.; Šostak, O.R. The Modelling of Roof Installation Projects Using Decision Trees and the AHP Method.
Sustainability 2020, 12, 59. [CrossRef]
35. Ahmad, T.; Aibinu, A.A.; Stephan, A.; Chan, A.P. Investigating associations among performance criteria in Green Building
projects. J. Clean. Prod. 2019, 232, 1348–1370. [CrossRef]
36. D’Acci, L. Quality of urban area, distance from city centre, and housing value. Case study on real estate values in Turin. Cities
2019, 91, 71–92. [CrossRef]
37. Woo, J. A Post-occupancy Evaluation of a Modular Multi-residential Development in Melbourne, Australia. Procedia Eng. 2017,
180, 365–372. [CrossRef]
38. Li, H.; Wei, Y.D.; Wu, Y.; Tian, G. Analyzing housing prices in Shanghai with open data: Amenity, accessibility and urban
structure. Cities 2019, 91, 165–179. [CrossRef]
39. Fuerst, F.; Warren-Myers, G. Does voluntary disclosure create a green lemon problem? Energy-efficiency ratings and house prices.
Energy Econ. 2018, 74, 1–12. [CrossRef]
40. Picarelli, N. There Is No Free House. J. Urban Econ. 2019, 111, 35–52. [CrossRef]
41. Amarasinghe, I.; Hong, Y.; Stewart, R.A. Development of a material circularity evaluation framework for building construction
projects. J. Clean. Prod. 2024, 436, 140562. [CrossRef]
42. Altarawneh, N.A.; Murtadha, T.K. Managing engineering challenges in the design and implementation of eco-friendly residential
structures. Results Eng. 2023, 19, 101363. [CrossRef]
43. Fredriksson, A.; Huge-Brodin, M. Green construction logistics—A multi-actor challenge. Res. Transp. Bus. Manag. 2022, 45, 100830.
[CrossRef]
44. Torres, J.; Garay-Martinez, R.; Oregi, X.; Torrens-Galdiz, J.I.; Uriarte-Arrien, A.; Pracucci, A.; Casadei, O.; Magnani, S.; Arroyo, N.;
Cea, A.M. Plug and Play Modular Façade Construction System for Renovation for Residential Buildings. Buildings 2021, 11, 419.
[CrossRef]
45. Markus, T.A.; Morris, E.N. Buildings, Climate, and Energy; Pitman Publishing Limited: London, UK, 1980.
46. Ching, F.D. Architecture: Form, Space, and Order; John Wiley & Sons: Hoboken, NJ, USA, 2023.
47. Zhang, J. Evaluating regional low-carbon tourism strategies using the fuzzy Delphi-analytic network process approach. J. Clean.
Prod. 2017, 141, 409–419. [CrossRef]
48. Padilla-Rivera, A.; do Carmo, B.B.T.; Arcese, G.; Merveille, N. Social circular economy indicators: Selection through fuzzy delphi
method. Sustain. Prod. Consum. 2021, 26, 101–110. [CrossRef]
49. Neupane, S.M.; Bhattarai, P.C. Constructing the scale to measure entrepreneurial traits by using the modified delphi method.
Heliyon 2024, 10, e28410. [CrossRef]
50. Kumžienė, J. Buildings Classification in Lithuania after 2021 1st of January. Available online: https://smart-rehabilitation.eu/
building-classification-lithuania/ (accessed on 31 October 2023).
51. Famakin, I.O.; Othman, I.; Kineber, A.F.; Oke, A.E.; Olanrewaju, O.I.; Hamed, M.M.; Olayemi, T.M. Building Information
Modeling Execution Drivers for Sustainable Building Developments. Sustainability 2023, 15, 3445. [CrossRef]
52. Kineber, A.F.; Othman, I.; Famakin, I.O.; Oke, A.E.; Hamed, M.M.; Olayemi, T.M. Challenges to the Implementation of Building
Information Modeling (BIM) for Sustainable Construction Projects. Appl. Sci. 2023, 13, 3426. [CrossRef]
Buildings 2024, 14, 1696 36 of 36

53. de Oliveira, L.G.; Miranda, F.G.; de Paula Dias, M.A. Sustainable practices in slow and fast fashion stores: What does the customer
perceive? Clean. Eng. Technol. 2022, 6, 100413. [CrossRef]
54. Eirgash, M.A.; Togan, V.; Dede, T.; Basaga, H.B. Modified dynamic opposite learning assisted TLBO for solving Time-Cost
optimization in generalized construction projects. Structures 2023, 53, 806–821. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.