MAPPING THE LANDSCAPE: A SYSTEMATIC LITERATURE REVIEW ON

AUTOMATED VALUATION MODELS AND STRATEGIC APPLICATIONS IN REAL

ESTATE

Abstract. In the rapidly evolving real estate industry, integrating automated valuation models (AVMs) has become critical for
improving property assessment accuracy and transparency. Although there is some research on the subject, no thorough qualitative
systematic review has been done in this field. This paper aims to provide an up-to-date and systematic understanding of the strategic
applications of AVMs across various real estate subsectors (i.e., real estate development, real estate investment, land administration,
and taxation), shedding light on their broad contributions to value enhancement, decision-making, and market insights. The
systematic review is based on 97 papers selected out of 652 search results with an application of the PRISMA-based method. The
findings highlight the transformative role of AVMs approaches in streamlining valuation processes, enhancing market efficiency,
and supporting data-driven decision-making in the real estate industry, along with developing an original conceptual framework.
Key areas of future research, including data integration, ethical implications, and the development of hybrid AVMs approaches are
identified to advance the field and address emerging challenges. Ultimately, stakeholders can create new avenues for real estate
valuation efficiency, accuracy, and transparency by judiciously utilizing AVMs approaches, leading to more educated real estate
investment decisions.

Keywords: real estate, automated valuation models, strategic applications, systematic literature review, PRISMA, conceptual
framework.

1. Introduction

The contemporary population growth requires the acquisition of land and the ensuing construction of buildings. The
rising population expansion is also creating a demand for real estate, which is making it more challenging to assess its
market value (Baur et al., 2023). Accurate real estate valuations are necessary for a fair and effective real estate market
(Jafary et al., 2024b). Due to its application in many areas of the economy, such as taxation, mortgage transactions,
expropriations, buying, and selling, the valuation process has been one of the most important factors in the economy
of the country (Reite, 2023). However, accurate real estate valuation is time-consuming and requires expert knowledge
(Glumac & Des Rosiers, 2020). As a result, automated solutions that enable quick and accurate real estate estimation
are beneficial to real estate stakeholders (Jafary et al., 2024a). Since inaccurate valuations can have a substantial
negative impact on a variety of stakeholders, including governments, policymakers, investors, real estate developers,
banks, insurance agencies, and consumers, it is imperative that the valuers accurately estimate the market value
(Swietek, 2024). Further, since most clients want to pay less for valuation services, the valuation should rapidly
embrace automation to offer affordable services (Trojanek et al., 2024). Key aspects of real estate valuation, like
durability, immovability, and a strong reliance on location and property attributes, inherently make valuations more
difficult (C. L. Lee et al., 2024). Furthermore, economic characteristics, heterogeneity, and stakeholder behaviors are
among the other significant aspects that might impact real estate values (Despotovic et al., 2023). Thus, while working
with a large number of properties with various features, valuation professionals are unable to handle such volumes of
data and considerations accurately and efficiently (H. Lee et al., 2024). Hence, to address such issues, smart,
trustworthy, and efficient automated valuation models (AVMs) should be used (Krämer et al., 2023; Matysiak, 2023).
AVMs have generated a buzz in real estate because they can be used for several purposes. It is found that AVMs
have higher precision than traditional appraisal methods, particularly for commercial real estate (Renigier-Biłozor et
al., 2019). Glumac & Des Rosiers (2020) exemplified a range of applications deemed most critical for AVMs,
including iBuyer platforms, mortgage financing, and mortgage-backed securities as well as property tax assessment.
Other applications may include land readjustment, portfolio risk assessment, real estate investment, insurance risk
assessment, lending risk evaluation, and negotiation margin (Renigier-Biłozor et al., 2022). Marketing real estate, in
addition, the intricate process of valuing real estate requires the consideration of numerous value-related criteria that
vary by type, origin degree of impact, and unit measurement (Su et al., 2021). In this context, AVMs presume that the
value of a property is predicted by various independent factors and estimate the value as an outcome variable in
parametric or non-parametric regression models over different dimensions of value (Renigier-Biłozor et al., 2022).
Apart from these essential features, some researchers further leverage artificial intelligence (AI), computer vision, and
deep learning to extract certain visual characteristics from image data, such as interior, external, and remote sensing
images, to improve the AVMs' predictive capabilities (Baur et al., 2023; Wan & Lindenthal, 2023). Although
contemporary valuation techniques have been shown to reduce uncertainty in the real estate valuation field, their actual
application has lagged behind expectations, and one of the main reasons may be the inability to obtain reliable and
accurate datasets (Batista & Marques, 2021). Besides, AVMs require clear, structured, precise, and classified data
about different property features (Yalpir et al., 2021). In addition, unbiased, objective, systematic assessments of real
estate have always been essential (Oliveira et al., 2021). The need for a fair market value appraisal is urgent now
because banks need to know the value of a property before approving a loan and the government needs to know the
value to calculate property taxes accordingly (Droj et al., 2024).
AVMs are currently in high demand in the real estate market because they can deliver property values faster,
better, and cheaper (Matysiak, 2023). Also, the deep involvement of big data and technology in the real estate sector
has boosted the development of AVMs into becoming an important tool for property complementary (Despotovic et
al., 2023). A summary of earlier literature review research regarding the application of AVMs in real estate contexts
can be found in Table 1. Therefore, this paper aims to review the strategic use of AVMs for real estate valuation, shed
light on their shortfalls and limitations in practice, and explore their potential impact and advantages they present to
the industry stakeholders based on the following research questions (RQs):

RQ1. In what ways have AVMs influenced strategic practices in the real estate sector?
This research question investigates how AVMs have a revolutionary impact on strategic decision-making in the
real estate industry.
RQ2. What are the main AVMs opportunities and challenges in the real estate sector?
Given the strategic role of AVMs identified in RQ1, this research question addresses the opportunities and
challenges that AVMs bring, which are crucial in comprehending the practical adoption and success of AVMs in the
real estate industry.

To this end, a systematic review of the literature was carried out with an emphasis on the strategic applications
for real estate valuation to contribute to research on the development of modern AVMs. The structure of the remaining
paper is described herein. Section 2 describes the methodology adopted for the systematic literature review, explaining
the research questions, scope, database, and criteria for the search and selection of the final pool of papers. Next,
Section 3 delves deeper into the content analysis outcomes, presenting the significant findings from influential
scholarly articles. Building upon this, Section 4 encompasses an extensive discussion along with future research
perspectives, and the main conclusions of the review are presented in Section 5.

Table 1. Main literature review studies on automated valuation models and real estate
Author(s) Keywords Number of Focus
articles
(Tekouabou et al., 2024) Artificial intelligence; machine 70 A systematic review of machine learning-
learning; systematic survey; based artificial intelligence in urban real
urban real estate; prediction estate prediction
(Wei et al., 2022) Real estate appraisal; big data; 124 Reviews literature on hedonic price
hedonic price model; valuation model-based real estate appraisal in the
method era of big data
(Bilge & Yaman, 2021) Integrated project delivery; 45 Review of literature on information
building information modeling; management roles in
project management; real estate development lifecycle
information management;
project lifecycle;
real estate development lifecycle
(D. Wang & Li, 2019) Mass appraisal; artificial 104 Review of literature on mass appraisal
intelligence; real-estate; geo- models of real estate
information systems; mixed
models
 Our study systematic literature review; 97 A systematic review of the strategic
automated valuation models; applications of automated valuation
real estate models in the real estate

2. Methodology

The study methodology used to investigate the literature on AVMs and real estate, as well as the interactions between
both of these topics, was a systematic literature review (SLR). An SLR is a significant research procedure that goes
beyond a simple review of prior studies and employs a pre-planned research strategy. The purpose of this kind of
review is to address particular research questions, expand upon prior research, choose and assess contributions,
analyze and synthesize data, and present evidence in a way that enables the researcher to make their own decisions
regarding what is known and unknown (Denyer & Tranfield, 2009; Malek & Desai, 2020). Compiling the findings
from various studies on a particular subject leads to a deeper level of conceptual or theoretical growth and a better
grasp of the subject than would be possible with a single investigation (Thomé et al., 2016).

2.1. Systematic literature review process

The present study reported the information of quality and relevance for the chosen publication by using Preferred
Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. To move away from the subjective
processes that produce biased research, the PRISMA group developed these standards in 2009 to report methodically
and statistically to ensure complete transparency (Liberati et al., 2009). Finding a relevant journal article about AVMs
in the real estate industry is the first step, which prompts a topic-based search for SLR in two licensed databases
(Scopus + WoS). According to several researchers, these are the most extensive relevant databases for our research
issue (Kipper et al., 2020; Vieira & Gomes, 2009).
Using the query string in the advanced search, the databases have revealed over 652 available articles. During the
penetrating keywords phase, the keywords and queries string information method was employed to locate related
material. The query string search for the Scopus database consists of TITLE-ABS-KEY (“automated valuation
models” OR “automated property appraisal”) AND “real estate” OR “property” OR “dwelling” OR “house” OR
“housing”). Meanwhile, the WoS query string search is TITLE-ABS-KEY “automated valuation models” OR
automated property appraisal AND “real estate” OR “property” OR “dwelling” OR “house” OR “housing”). The
second step involves going through the listed literature and selecting the appropriate articles related to the study's
topic.
The third step consists of the eligibility and exclusion process. The authors have determined a number of eligibility
and exclusion criteria (All peer-reviewed journal articles published in the English language between January 2000 and
July 2024 are among the inclusion criteria that are applied. Similarly, only articles that directly addressed the strategic
use of AVMs in the real estate industry were taken into account. Under the exclusion criteria, editorials, PhD and
master's theses, book chapters, conference papers, and proceedings were excluded. Moreover, publications that were
accessible before the year 2000 were disregarded to allow space for the analysis of more current research on the
selected topic. Lastly, publications that the authors were unable to access were also excluded). Subsequently, the
publications were examined comprehensively to verify their alignment with the research goal. Following this last
evaluation, 97 papers were remained. The fourth phase of data analysis and synthesis involved classifying the selected
articles according to their title, author, year of publication, kind of publishing, type of research, and theme of approach.
The results were categorized to provide a clearer picture of the previous research works in this field. An overview of
the strategic applications of AVMs in real estate was provided in the fifth phase through discussions regarding the
content analysis and the research's results.

Following the search process, 97 peer-reviewed scientific articles were found to be the primary sources discussing
or concentrating on the strategic applications of AVMs in the real estate industry.

• Year-wise publications: The publication window was unrestricted and extended until July 2024 with a total of
9 articles. Between 2020 and 2023, there was a notable increase in the number of publications (total 36 articles).
Researchers' interest in the contemporary and highly significant construct of automated valuation, its consequent
innovations, and their strategic implications is evident from the increasing trend in the number of publications.

• Journal-wise publications: The research is broadly dispersed throughout numerous journals. Academic papers
have covered AVMs and real estate in great detail. There is decentralization across thematic areas – engineering,
 urban studies, technological and social change, operations management, and strategy. Concerning the numbers,
leading journals included the Journal of Property Investment and Finance, Journal of Property Research, Journal
of Real Estate Finance and Economics (n = 5 each), and Journal of Property Tax Assessment and Administration,
and Sustainability (n = 4 each), followed by Property Management (n = 3).

• Contributions by Country This section covers the contribution of different countries to the literature on real
estate and AVMs. Aside from China, developed countries in America, Europe, and Asia account for the majority
of publications. The analysis indicates the topmost productive countries. With 13 publications - 13.4% of the pool
of articles—China is the most productive country. Poland, the USA, Australia, and Italy produced 8.2%, 7.2%,
6.2%, and 6.1% of the total literature, respectively. The main factor contributing to higher productivity in
countries like China is the easily accessible and well-documented data. The availability and dependability of
property transaction data are major issues in the majority of developing countries, which limits the scope of AVMs
research.

• Research method: We examined the trend of different research methods used to investigate the strategic
applications of AVMs for the real estate sector, taking into account the importance of the research methodologies
utilized in the articles. It is evident from our findings that mixed studies (n = 25) make up the majority of the
literature, followed by empirical/case studies (n = 19), survey (n = 17), modeling (n = 15), literature reviews
including conceptual papers (n = 12) and experimentations (n = 6) and simulation articles (n = 3). The results
indicate that both qualitative and quantitative academic views have drawn attention to the current study issue. The
case studies included an overall validation of the theoretical principles coupled with practical applications, while
the conceptual and review papers explained how AVMs may affect various real estate markets. Prototypes,
simulation techniques, and testing are nevertheless necessary to evaluate design issues and suggest feasible
solutions.

3. Data analysis and findings

Following a comprehensive review of 97 carefully selected papers, we conducted a content-based analysis of the
literature. To make informed decisions, content analysis aims to organize and evaluate the information acquired (Gaur
& Kumar, 2018). The topics and keywords found in each pertinent article were identified using a content analysis of
the articles. To aid in the discussion that follows, the primary real estate sub-sectors, AVM approaches, and strategic
applications of AVMs in the real estate industry are defined in the following subsections.

3.1. AVMs approaches

In terms of AVM approaches, the majority of articles have taken a broad perspective. More than one approach was
discussed in some papers, which resulted in 136 approach references from 97 papers. During the analysis, we found
that not all AVM approaches have received the same amount of attention in the literature. The most singly addressed
AVM approaches were hedonic pricing models (HPMs), machine learning (ML) models, artificial neural networks
(ANN), and hybrid models (HM). Some of the approaches are at more advanced stages of maturity (e.g., HPM, ML,
and ANN) in research and real estate property applications, while others such as genetic algorithm (GA) (Cardone et
al., 2024) are still in the early stages of development. Remarkably, although Regression Analysis (RA) applications
are only starting to be implemented in real estate property (Foryś, 2022), RA itself has already attracted considerable
academic attention.

Automated valuation models may produce a report in a matter of seconds and utilize a variety of data sources,
including location, property age, and condition (Krämer et al., 2023). AVMs are practical for real estate buyers, as
they reduce the need for prospective buyers to personally check every property on the market (Jafary et al., 2024a).
The primary AVMs approaches that are intended to give an accurate picture of AVM performance are covered in the
following subsections.

3.1.1. Hedonic pricing models

The unique features of the housing market have led to the widespread application of hedonic pricing in real estate
assessment research. This approach evaluates real estate pricing by considering the core attributes of the asset (Zaki
et al., 2022). The hedonic approach would assert the wisdom of real estate as an asset that can be considerably
conformed by many in terms of location, number of rooms, or living area (Potrawa & Tetereva, 2022). Property
attributes in this sense are often grouped according to their hedonic characteristics. Gröbel & Thomschke (2018)
characterized the three fundamental groups: locational group referring to real estate attributes like view type, distance
to central business district (CBD), structural attributes delineating the housing space, such as number of bedrooms, or
building age; and neighborhood features as characterizing a real estate.
Until recently, the hedonic pricing model (HPM) has been widely used to estimate the market value of real estate
utilizing multiple regression analysis (Rey-Blanco et al., 2024; Zaki et al., 2022). The model presupposes that real
estate is a differentiated commodity with various characteristics defining utility (Potrawa & Tetereva, 2022). In other
words, the buyer of a property is buying not this or that characteristic of the property, but rather a set of them that can
be realized in specific value to utilities. Multi-family residences are different in that parking issues will drive up the
price of the actual property (Valdez Gómez de la Torre & Chen, 2024). This can be attributed to a generic instance of
the model, criticized in some literature for over-simplifying the nonlinearity or complexity of the real world (Rey-
Blanco et al. 2024; R Wang & Rasouli, 2022). Following (Schirripa Spagnolo et al., 2024), the functional form of the
standard HPM is informed by robust priors on housing and household preferences. The model assumes that every
attribute is continuous and discrete which means perfect competition, market equilibrium, discrete preference, and a
union of the market.

3.1.2. Machine learning models

Machine learning (ML) is changing the real estate valuation sector. ML is a state-of-the-art artificial intelligence
technique that learns from the provided input data being used to give the right predictions (Potrawa & Tetereva, 2022).
Moreover, ML can be more disruptive in terms of changing existing valuation methodologies (Zaki et al., 2022). More
and more are leveraging these technical advancements in their work, which is clear, especially with the increase of
papers around the topic: The body of literature on utilizing Machine Learning models for real estate valuation is
rapidly expanding. Several studies show that ML-based AVMs outperform human pattern recognition-based models
(Steurer et al., 2021; Su et al., 2021). In addition, in real estate forecasting, ML techniques such as decision trees,
random forests, and gradient boosting are becoming more popular (Doumpos et al., 2021; Sing et al., 2022). It also
constructs an advanced model that allows accessing large complex data sets, and inferring patterns to forecast more
accurately (Potrawa & Tetereva, 2022). However, these methods can capture non-linear patterns but require extensive
data preprocessing and parameter tuning to avoid overfitting (Baur et al., 2023).

3.1.3. Artificial neural networks

The Artificial neural network (ANN) is one of the stream methods commonly employed for appraisal purposes
(Kamara et al., 2020). The ANN computational model can accept much more variation than HPM can, given the fact
that it does not rely on linear bonds (Foryś, 2022). After this, the ANNs, using a complex system of artificial neurons,
can predict house prices since they know how each real estate-related input variable relates to its corresponding output
(Horvath et al., 2021). There are several studies to test the superiority of HPM and ANN which have been done in real
estate appraisal. In general, ANN has reportedly delivered MAE of 5–10% for property valuation methods, however,
the average HPM is higher and typically ranges from 10 to 15% (Abidoye & Chan, 2017, 2018). However, there are
no specific numbers on how many samples to use when valuing real estate with ANNs can be effective and if this is
a universal principle (Hoxha, 2023). The primary advantage of using ANNs is that they can care for uncertainty input
into the prediction model, provide gain in terms of accuracy, and are time-saving (Foryś, 2022). ANNs operate as
black box apparatuses meaning, the process of how the output values were generated is not explicated (Yasnitsky et
al., 2021). Because of this lack of transparency, ANNs are difficult to utilize in certain scenarios, such as expropriation
and property taxation, where the assessments may be contested and require a legal defense (Rampini & Re Cecconi,
2021).

3.1.4. Hybrid models

The goal of the hybrid model is to reduce counterarguments. This paradigm is described as a system that combines
elements of both new and old components built in an agile manner, creating a synergistic whole (Kamara et al., 2020).
In this particular domain, hybrid refers to the interchangeable use of Combining human intervention (e.g., field
inspection and situations where physical inspection is required to obtain initial property features, the analysis of
results, and final decisions) with computer use (e.g., calculation processes); manual (e.g., for coding variables and
quality control programs); automated solutions (e.g., for assessing the significance of features, comparable properties
selection); and classical learning (e.g., regression, classification, association, clustering, generalizations) with deep
learning and neural nets within ML approaches (e.g., recurrent neural networks, convolutional neural networks,
generative adversarial networks) (Nor & Raheem, 2024; Özöğür Akyüz et al., 2023). These hybrid techniques can
improve the accuracy and dependability of valuation decisions by incorporating domain experience and expert
knowledge into ML-driven valuation models, especially in specialized or complicated real estate markets (El jaouhari
et al., 2023; X. Zhang et al., 2024).

3.1.5. Geospatial models

The geographical components of real estate prediction are examined using geospatial analysis models (Aydinoglu &
Sisman, 2024). Spatial statistics and geographic information systems (GIS) are useful tools for modeling real estate
trends based on proximity and location (Mete & Yomralioglu, 2023). Although they might need unique data, these
techniques are crucial for studying location-specific characteristics(Xia et al., 2022). The standard hedonic price
regression model is not able to effectively address these spatial issues given the substantial spatial autocorrelation and
spatial variability of housing price data (Schirripa Spagnolo et al., 2024). Thus, two spatial modeling paradigms have
been developed as a result of researchers' improvement of the model through the consideration of spatiality. First,
several researchers have combined traditional hedonic price regression with spatial econometrics models (Oust et al.,
2020; R. Wang & Rasouli, 2022). For instance, (Y. Wang et al., 2017) found that the spatial lag and spatial error
model produced better accuracy than the ordinary least squares (OLS) regression model when examining Beijing,
China, home prices. Others have mostly focused on how spatial variability affects housing costs. For example, (Cao
et al., 2019) examined the cost of public housing in Singapore using big data from smart cards and discovered that the
geographically weighted regression (GWR) model performed significantly better than classic linear regression .
Second, some studies estimate housing prices using spatial interpolation in addition to spatial econometric models
(Dambon et al., 2021; Zhou et al., 2020). The primary benefit of this approach is its ability to retrieve the price surface
of the entire region using a small number of sampling points (Aydinoglu & Sisman, 2024). Further, geospatial
modeling has significantly increased accuracy when compared to numerous linear regression models (Saldana-Perez
et al., 2024).

3.1.6. Regression analysis

Through the use of regression analysis, a set of data can be examined to estimate a potential functional relationship
between an independent and dependent variable (Foryś, 2022). Besides, "Multiple" regression methods are
characterized by the dependent variable being a function of multiple factors in general (Liu, 2022). Precisely such
systems need to be used for the study of intricate economic systems such as the real estate market (N. Chen, 2022).
Much like in regression analysis, the appraiser defines the independent factors that impact the value in regressio n
analysis and then chooses the conditional attribute in the same manner (Carbonara et al., 2021). In the last stage, the
appraiser examines the connections between conditional and decisional features (Liu, 2022). Furthermore, since they
are comprehensible and capable of handling numerous predictor variables, both linear and non-linear regression
models are often employed in real estate forecasting (Pai & Wang, 2020). However, if not properly adjusted, they may
overfit and fail to adequately represent non-linear connections (Steurer et al., 2021).

3.1.7. Income and cost approach

The income approach states that the capitalized value of the income that a piece of income-producing real estate
generates both now and in the future, matches the piece's market value (Glumac & Des Rosiers, 2020). The appraiser
assesses the amount, quality, direction, and duration of this income before converting it into an expression of present
value, or market value, using an acceptable capitalization rate (Su et al., 2021). Either the Discounted Cash Flow
(DCF) method or the direct capitalization method is used to estimate the value of the property under investigation,
depending on its kind and complexity (Frodsham, 2024). The DCF method makes sense theoretically and is especially
suitable for assets with many income streams that are not often traded (like hotels and shopping malls) (Čirjevskis,
2021). Furthermore, the cost approach is used to determine a property's value by adding the land value and the
depreciated reconstruction cost (DRC) (Arcuri et al., 2020). According to the cost approach, a real estate asset's market
value is equal to the land purchase price plus building costs, minus depreciation (Leskinen et al., 2020). Further, the
cost approach is mostly applied to single-use, non-income-producing real estate properties (such as schools and
churches) and certain industrial buildings that are rarely sold on the market for property insurance purposes (Su et al.,
2021).

3.1.8. Market approach

The market approach consists of determining the property’s value by contrasting the property under assessment with
similar properties for which transaction prices or sales are available (Tajani et al., 2019). Its accuracy and application
are largely dependent on the level of development and local circumstances of the real estate market, with the valuers'
main duty being application and interpretation (Ghosn et al., 2024). However, there is a wealth of literature regarding
challenges associated with valuation, especially when applying the market approach. Information asymmetry (Li et
al., 2019), valuation variances (Arcuri et al., 2020), uncertainty in valuation (Despotovic et al., 2023), and accuracy
of valuation (H. Lee et al., 2024) are only a few of the challenges covered by these studies. As a result, in reality,
practitioners alter this fundamental model based on the kind of property being valued, variations in the amount of
available market data, technological developments, raised client awareness, requests for greater accuracy in the
assessed values, and generally altered real estate market dynamics (Ghosn et al., 2024; Tajani et al., 2019).

3.2. Integrating AVMs in real estate: strategic applications (RQ1)

The number of published articles in this research area led to the identification of eight separate real estate subsectors.
With 24% of the articles, real estate development was identified as the primary real estate sub-sector. The real estate
investment subsector contained 19% of the documents, followed by land administration (16%). Leasing and renting,
Property transactions, and real estate business only contributed 9%, 7%, and 5% of the papers, respectively. Lastly,
real estate administration and real estate maintenance received less attention.
Two more "segments" were found in light of the existence of several publications about the global real estate
market and the investigation of mixed intended uses in the same study. It should be noted that the studies that
concurrently examined the effects of AVMs on two or more real estate subsectors were included in the "mixed uses"
group. Instead of focusing on a particular intended purpose, the "global real estate market" segment is meant to gather
research that examines the dynamics and trends of global real estate indices to identify the broader impact of AVMs.

3.2.1. Real estate development

The goal of real estate development is to use investments to raise the value of a single property or an area within a
property (Baur et al., 2023). A plot of land within a real estate, a building or its components within a plot, or a land or
water region that forms raw land for real estate can all be the object of real estate development (Jafary et al., 2024b).
Development of real estate can either concentrate on the portion of land that is still undergoing planning or on already
existing real estate and its components (Reite, 2023). Furthermore, AVMs—particularly those that use HPM—
evaluate property values by considering both locational and intrinsic qualities, offering a thorough breakdown of value
drivers (Abidoye & Chan, 2018). Similarly, ML models, especially neural networks, are excellent at analyzing big
data sets to identify intricate patterns and make remarkably accurate predictions about real estate prices while
dynamically responding to shifting market conditions (Pai & Wang, 2020). Furthermore, regression analysis is still
fundamental since it uses statistical modeling to quantify the correlations between values and property attributes (Liu,
2022).

3.2.2. Real estate investment

Investing capital in real estate or its components to generate profit or other advantages is referred to as real estate
investment (Swietek, 2024). According to (Saldana-Perez et al., 2024), the most recent ML models offer precise cost
estimation through data exchange formats, assess risk, growth, and depreciation rates in real estate investments, and
give the owner clear, graphical, and economically significant information. Since ML models are parametric and smart,
(Fazeli et al., 2020) proposed a semi-automated BIM-based cost estimation approach that enables practitioners to
estimate the cost of projects. Further, real estate depreciation was calculated using HPM by (R. Wang & Rasouli,
2022), providing developers and real estate managers with an effective tool for decision assistance. Further, (Lisi,
2019) integrated a market approach, multiple regression analysis, and the implicit prices of housing to evaluate
environmental impacts during the early stages of design and provide building envelope alternatives for decision-
making throughout the building lifecycle.
3.2.3. Land administration

Information regarding the ownership, value, and usage of land and its related resources are recorded and disseminated
through the process of land administration (C. L. Lee et al., 2024). Several research concentrating on land
administration and taxation highlight the new opportunities for real estate valuation making use of the advantages of
machine learning models, particularly by applying and/or expanding the land administration domain model (LADM),
which is an international standard for land administration (Atazadeh et al., 2021; Indrajit et al., 2020). According to
(Rey-Blanco et al., 2024), HPMs offer a comprehensive examination of property values through the dissection of
intrinsic features and locational qualities, facilitating fair land assessment and efficient property taxation. Similarly,
hybrid models combine sophisticated algorithms and conventional valuation methodologies to provide comprehensive
land valuation frameworks by fusing the predictive power of contemporary technologies with the interpretability of
proven methods (Kamara et al., 2020). Likewise, regression analysis measures the connections between values and
land attributes, offering factual information for determining land values and formulating policies (N. Chen, 2022).

3.2.4. Leasing and renting

The transfer of the right to use real estate or any portion of it in exchange for money is known as leasing (Y. Zhang et
al., 2020). This includes land leasing, apartment leasing, and lease-purchase. In a similar vein, renting involves paying
for the privilege of using real estate (Tanrıvermiş, 2020). The fourth-largest literature category was devoted to studies
on leasing and renting, which mostly contained solutions intended to alleviate existing leasing problems, like
intermediary dependencies, inefficiencies, and trust issues (J. Chen et al., 2022; Rosenthal et al., 2022). Further, the
leasing process is frequently manual, labor-intensive, and paper-based (Y. Zhang et al., 2020), which adds to the
industry's non-transparency, high costs, information asymmetry, and fraud.
Through the dissection of intrinsic qualities and locational elements, AVMs-enabled HPMs could assess the worth
of properties and help landlords and renters negotiate for reasonable rental rates that are in line with market dynamics
and property details (Potrawa & Tetereva, 2022). Furthermore, geospatial models improve market segmentation and
strategic property positioning for rental investments by utilizing spatial data to evaluate rental values based on
neighborhood dynamics and proximity to facilities (Aydinoglu & Sisman, 2024). Further, landlords can set
competitive rents and maximize property management techniques with the help of regression analysis, which
quantifies the links between rental property qualities and market rates (Carbonara et al., 2021).

3.2.5. Property transactions

The purchase and transfer of proprietary rights to real estate or its components is known as a property transaction
(Evangelista et al., 2020). Actions about the purchase and sale of real estate, such as real estate exchanges, purchases,
sales, and valuations, are included in the category of property transactions (Jiao & Xu, 2022). AVMs could primarily
assist in resolving the problem of intermediary-dependent, manual, paper-based, expensive, and time-consuming
property transaction processes (Glumac & Des Rosiers, 2020). Furthermore, by breaking down property values into
intrinsic qualities and locational aspects, HPMs may offer a thorough analysis that supports reasonable pricing and
negotiation tactics (Zaki et al., 2022). Similarly, ANNs use massive datasets to forecast real estate values based on
past performance and present market dynamics, providing critical information for pricing competitiveness and
investment choices (Horvath et al., 2021). Further, the cost approach assesses property value based on replacement or
reproduction costs, directing decisions on property renovations and asset assessment (Glumac & Des Rosiers, 2020),
whereas the income approach assesses property worth based on income-generating possibilities, which is crucial for
evaluating investments in commercial real estate (Su et al., 2021).

3.2.6. Real estate business

The activities with financial goals connected to the ownership, use, and production of real estate or the provision of
services to customers within it are collectively referred to as real estate business (Matysiak, 2023). Establishing
services that create value addition for real estate users, and consequently, owners and investors, throughout the real
estate lifecycle, is the fundamental concept of the real estate business (Krämer et al., 2023). AVMs combining several
approaches - like cost, income, and market approach - provide complete valuations in the real estate business helping
guide strategic business decisions (Tajani et al., 2019). These approaches facilitate pricing strategies, risk
management, competitive positioning, and market analysis, which promotes operational excellence and business
success (Glumac & Des Rosiers, 2020). Further, (Kamara et al., 2020) proposed a novel hybrid neural network model
to generate property values instantaneously.

3.2.7. Real estate administration

Real estate administration includes the management of real estate-related finances, operations, human resources, and
information services, as well as the administration of real estate-related legal matters (Renigier-Biłozor et al., 2022).
More specifically, real estate administration can be defined as the management of the complete real estate company
and building management (Wan & Lindenthal, 2023). As per (Rey-Blanco et al., 2024), HPMs break down property
attributes to determine how each one contributes to market value, taking into account the complex preferences of both
buyers and sellers. Meanwhile, ML models, which are distinguished by their capacity for adaptive learning, enhance
valuation accuracy by combining large datasets and spotting complex patterns those conventional methods might miss
(Foryś, 2022). Further, (Saldana-Perez et al., 2024) argued that geospatial models might use spatial data to
contextualize valuations in space and capture localized market dynamics.

3.2.8. Real estate maintenance

The service of keeping real estate in the intended state and with the desired condition, value, and attributes is known
as real estate maintenance (Batista & Marques, 2021). Energy management, facility services, technical services,
cleaning services, trash management, and management of outdoor areas are all included in real estate maintenance
(Yalpir et al., 2021). (Rey-Blanco et al., 2024) assessed market prices and examined the features of the properties
using HPM, which helped with portfolio management and property tax assessments. Moreover, (Ogunfowora &
Najjaran, 2023) developed an ML system to improve maintenance scheduling by estimating the life of equipment and
allocating resources as efficiently as possible using past data. In addition, (Aungkulanon et al., 2024) employed a
hybrid approach to estimate maintenance requirements and lower operating expenses by fusing real-time sensor data
with predictive analytics. Similarly, (Cheng et al., 2020) developed a framework based on BIM and IoT technologies
for property maintenance management to compare the costs of property maintenance to those of comparable assets,
helping to guide decisions about maintenance and redevelopment expenditures.

Furthermore, we summarize the strategic applications of AVM approaches in different real estate subsectors from
the perspective of three fundamental pillars, such as technology, social, and environment, in Table 2, which includes
knowledge from our extensive descriptive and content analysis of academic publications.
Table 2. Strategic applications of AVM approaches in different real estate subsector

AVM approach Real estate Real estate Land Leasing & Property Real estate Real estate Real estate
development investment administration renting transactions business administration maintenance
Real estate
subsector
Hedonic pricing Property and Comparative Land valuation Rental pricing is Estimating Pricing tactics are Regulation Forecasting
models Land valuation pricing analysis for taxation and based on transaction determined by compliance and maintenance
based on of property public policy property prices via market dynamics real estate tax costs by
amenities, features for determination. features. feature-based and consumer assessments. feature-based
location, and investment comparisons. preferences. comparisons.
market purposes.
conditions.
Machine learning Predictive Market trend Automated Forecasting Precise and Predictive Predictive Predictive
models models for risk analysis and assessment for changes in rental dynamic modeling and analytics for maintenance
assessment and portfolio urban planning prices using past estimation of scenario analysis optimized plans are built
development optimization. and zoning. data. property values for business property on insights
profitability. in rapidly intelligence. management and from machine
evolving increased learning and
marketplaces. operational historical data.
effectiveness.
Artificial neural Assessing Risk reduction Automated Nonlinear Forecasting Demand Automated AI-based
networks complex non- through the border appraisal pricing transaction forecasting and property property
linear market examination of and land parcel forecasting for prices in sophisticated management degradation
factors market emotions identification. intricate leasing different market market using forecasts of and
to identify and investing situations. scenarios. segmentation for several variables maintenance
development trends. real estate for trends. requirement
prospects. companies. predictions.
Hybrid models Combining Combining Thorough Dynamic Strong Integrated Coordinated Hybrid model-
several AVM market and valuation for pricing models transaction valuation models administrative based
approaches for economic data infrastructure combining pricing using a for improved decision-making maintenance
development to enable sound development machine variety of decision-making. and property cost
planning and investment and land learning, ANN, assessment valuation systems. forecasting and
accurate decision- registration. and hedonic techniques. optimization.
valuation. making. methods.
Geospatial models Land use- Investment Geographic Spatial analysis Estimating Location-based Effective property Geographical
optimized tactics focusing information of pricing transaction strategic planning management and forecasting for
valuation based on geography- systems (GIS) dynamics and prices with the and business land use planning environmental
on geographic based asset for planning and rental demand. use of intelligence. using GIS. factor-based
and locational valuation. land geographic data. maintenance
considerations. registration. planning.
Regression Linear Forecasting Determination Regression Transaction Regression Regression Modeling
analysis regression for investment of property models to pricing forecasts analysis for model-based maintenance
expenses and performance value for tax forecast rental derived from financial administrative costs using
revenue and analyzing assessment and prices based on regression forecasting and assessments and historical
estimation in market trends. land-use policy. property analysis of past valuation. reporting. attributes and
development attributes. data. trends.
projects.
Market approach pricing new Benchmarking Market value Rental property Assessment Market-driven Reviewing and Evaluating
developments real estate evaluation for comparison derived from pricing tactics for evaluating maintenance
through investments property parcels within the same analogous real corporate market-based expenses via
comparative against market in public and market. estate activities. valuations through market
market analysis. comparables. private transactions. administrative comparisons.
transactions. processes.
Income & cost Project viability Analyzing Assessment of Valuation of Transaction Strategies for Evaluations of Analyzing
approach analysis based capitalization land use rental income pricing income-based real administrative maintenance
on development rates and considering for rental determination estate operations expenses and costs in light of
expenses and revenue through development properties. based on valuation. strategies for the cost
anticipated real estate expenses and income- generating structure and
revenue. investment prospective generating income. prospective
appraisal. profits in the potential. revenue of the
future. property.
3.3. Opportunities and challenges (RQ2)

Navigating the real estate industry's vast AVM landscape reveals an array of opportunities and challenges. There is
enormous complexity in the number and types of data sources that have to be integrated — ranging from market
reports to real estate listings. Moreover, an advanced infrastructure is required to manage the heavy data load in time
and quality (Glumac & Des Rosiers, 2020). The real estate markets also indicate the difficulty of validating accuracy
and so require agile systems that can handle change faster (C. L. Lee et al., 2024).
Opportunities: The real estate industry can benefit a lot from AVMs, mainly in terms of enhancing decision-
making skills, improving accuracy and efficiency (D. Wang & Li, 2019). AVMs can simplify the procedures of
property valuations and consume fewer labor costs and timeframes than conventional methods to evaluate properties
(Bilge & Yaman, 2021). In addition, AVMs allow dynamic decisions in the field of real estate investment and portfolio
management through on-the-fly valuation adjustments because of market fluctuations (Wei et al., 2022). AVMs have
powerful advanced analytics such as ML and ANNs to improve prediction accuracy — allowing investors to anticipate
market movements, and hedge their risk (Abidoye & Chan, 2018). By utilizing geospatial data to contextualize
valuations within local market dynamics, these models enhance the precision of property assessment and strategic
planning (Ghosh & Mukherjee, 2018). Additionally, AVMs also allow for the optimization of resources and reduction
of transactional risks while taking advantage of market opportunities by automating complex valuation processes
(Tekouabou et al., 2024). This in turn ensures a more firm and effective real estate industry (Droj et al, 2024).
Challenges: While there are benefits to using AVMs, there are also many challenges. One of the primary
challenges of AVMs is that because most of them are heavily data-dependent — drawing information from various
databases and sources like property attributes, transaction histories, or market conditions — achieving high-quality
and accessible data becomes an issue to grapple with (Oliveira et al., 2021). Inconsistent or outdated data may then
cause inaccurate values, which in turn could compromise the effectiveness of the model (Yalpir et al., 2021). Model
Transparency and Complexity: In the last background, model transparency is also necessary to bear in mind, and due
to their complexity, advanced techniques such as ANNs and ML are difficult to interpret, which complicates the
understanding of how a decision was made (Foryś, 2022). Externally, they have to face market volatility, and the
influences that come from outside, such as changes in the economy or regulations, cause the model to fail in predicting
the future creating uncertainty in valuation results (Batista & Marques, 2021). In addition, human judgment still plays
an important role in the transactions of real estate, as valuations take into account subjective factors while property
inspections have to be done in person, causing problems for AVMs when they only rely solely on data (Batista &
Marques, 2021). Consequently, taking into account that progress in the field of models and methods used in AVMs
will occur dynamically due to new data sources, improved model approaches, and changing regulatory framework,
the readiness of these models to be applied should always be qualified (Renigier-Biłozor et al., 2022).
Based on the content analysis of the 97 articles, we have grouped challenges and opportunities into the following
segments: (i) Environmental, (ii) Social, and (iii) technological (see Table 3).

Table 3. Summary of opportunities and challenges from the selected papers

Groups Opportunities Challenges
Technological Improved prediction accuracy using neural Issues with data availability and quality
networks and machine learning
Agile decision-making is facilitated by real-time Advanced models' complexity and lack of
value updates. transparency
Combining several data sources (such as Inability to adjust to quickly evolving data
transactional and geospatial) for a thorough environments and technological advancements
analysis
Time and labor costs are decreased through the Data privacy issues and cybersecurity threats
automation of valuation operations.
Using advanced analytics to spot investment Reliance on substantial datasets that aren't always
opportunities and market trends available
Capacity to effectively manage vast amounts of Models must be updated and maintained
data constantly to guarantee relevancy
Environmental Environmental risk assessment supporting Difficulties in accurately incorporating and
proactive maintenance and management of real updating environmental data
estate
Enhancing evaluation of the effects of Variability in the local environment influencing
environmental changes (such as climate change the model's output consistency
and natural disasters) on property values
 Promoting sustainable growth by making well- Complicated interactions between environmental
informed decisions factors
Enhanced distribution of resources for Potential for bias in sources of environmental data
environmental sustainability
Increased accuracy in property value taking into Predictive accuracy is impacted by market
account the dynamics of the local environment volatility and other variables (such as changes in
the economy or regulations).
Social Enhanced stakeholder trust through greater Human monitoring is required to handle
transparency in property appraisal subjective influences in value.
Democratization of real estate appraisal using The potential of algorithmic bias affecting
easily available AVM instruments valuations' equity and fairness
Process simplification improving stakeholder Industry experts' resistance to the changes in
engagement and customer satisfaction standard processes
Improved ability to predict and reduce market risks Ethical issues in using data and making decisions
using algorithms
Assistance in formulating policies with data- Automating real estate transactions while
driven perspectives on housing markets maintaining a requirement for individualized
services
Encouraging more informed investment decisions
from the public and private sectors

4. Discussion

Global real estate markets are developing quickly both as an asset class for investors and as a means of facilitating
human interaction. As a result, the accuracy of their value is becoming a more important indicator of their performance
and appeal to businesses, governments, and communities (Swietek, 2024). Policymakers are facing mounting pressure
to devise tactics and implement novel regulatory structures that enhance the efficiency of the real estate market,
thereby enabling the sector to rival other asset classes, in response to the changing demands of investors, enterprises,
and consumers (Jafary et al., 2024a; Reite, 2023).
To this end, the discussion considers possible areas for future research and development and considers the
strategic implications of including AVMs in real estate valuation processes. The use of AVMs methods in the
evaluation of real estate carries promise for faster, and more accurate property valuation (Despotovic et., 2023). AVM-
driven valuation techniques may be automated and facilitated by effective ML models and HPMs for accurate, faster
property valuations (Evangelista et al., 2020; Potrawa & Tetereva, 2022). These models can capture complex market
dynamics and provide stakeholders with valuable information through the use of large datasets and state-of-the-art
computational methods. In addition to this, the integration of ANNs with real estate appraisal allows stakeholders to
anticipate market behaviors and trends, identify risks, and be able to make sound investment choices (Abidoye &
Chan, 2017). So AVMs-enabled regression analysis can also analyze historical data and market indicators from the
past to help give a better estimate of property values/value trends and potential ([Link], 2022). To ensure trust
together with regulatory compliance when using AVM-based valuation processes methods should be accounted for
(Doumpos et al, 2021). Similarly, the development of acceptance and use of AVMs in real estate needs to consider
bias, data privacy, and interpretability of AVM methodologies (Jafary et al., 2024b).
In the realm of strategic applications, it is evident that an AVM plays a vital role in revolutionizing the real estate
industry such as decision-making, and enhancing efficiency and accuracy of valuation. Such AVMs provide geospatial
data derived, real-time property valuations, as a function of machine learning and advanced analytics rendering
decision-making to the stakeholders in managing their portfolios (Steurer et al., 2021). These models have simplified
the evaluation process reducing human errors, as well as time and cost associated with conventional approaches (Zaki
et al. 2022). “AVMs also allow you to quickly assess market trends and estimate risk, which is a key driver of proactive
property asset management” (Potrawa & Tetereva, 2022). In addition, AVMs are a comprehensive and agile method
to assess property by combining diverse data resources through advanced algorithms (Baur et al., 2023). This results
in data-driven and strategic practices in the real estate sector. Efforts should focus on the development of user-
interpretable interfaces that are specifically tailored to real estate valuation, to improve interpretability and accuracy
of AVM-driven evaluation models. Alternatively, as demonstrated by Baur et al. (2023), those approaches, that delve
into the construction of hybrid AVM systems which combine human expertise with ML skills, offer a way forward
for improved valuation insights and decision support. New insight and possibilities for both accuracy, efficiency, and
transparency of property assessments are provided by the integration of ML-based AVMs in real estate valuation
procedures (Potrawa & Tetereva, 2022). Considering the high demand to access and interpret vast datasets, real estate
stakeholders can harness the transformational capability of AVMs to navigate through contemporary property
mechanisms and formulate an ethical, regulatory, and technical solution that can redefine industry practice in a
resilient sustainable future.

4.1. Future research directions

There is a potential in the future to be extremely important for the real estate sector due to its direct interest, among
other things, in AVMs and real estate value. AVM approaches continue to evolve rapidly and so there are many
research and development opportunities to make the valuation process better. There is a need for more research on
AVMs-driven valuation approaches developed and implemented. These approaches use more advanced ML models
like deep learning, computer vision, and natural processing language to automatize and enhance the procedure of
valuation (Potrawa & Tetereva, 2022). Using massive datasets and advanced computational algorithms, these models
would be able to generate more accurate and insightful property valuations. In addition to this, another advantage of
using AVMs is that it might expand the predictive strength of real estate valuation. In addition, the AVM methods can
detect patterns and relations which are more accurately done by analyzing past data and market trends to forecast with
accuracy which property value drivers move the market. So, the use of predictive analytics can help stakeholders
mitigate risks and select investments with more depth (Hoxha, 2023). Future research can also explore approaches to
include data from diverse sources in AVMs-centric appraisal models. Some of the data is structured, some are
unstructured, such as emotions in social media and satellite images, and property attributes, transaction histories, etc.
An AVM-centric valuation methodology also has the added benefit of providing richer, more comprehensive
evaluations of real property value by capitalizing on these diverse data sources as well (Foryś, 2022).
The focus of further study works in this field would be centered on issues relating to ethics and regulation in real
estate valuation and AVMs (Nor & Raheem 2024; Özöğür Akyüz et al. 2023). The concerns of data privacy, bias,
equity, and regulatory compliance need to be addressed with the increased adoption of AVMs in the valuation process.
Future research work could develop models and recommendations to ensure that AVM-assisted policies are consistent
with ethical norms as well as legal regulations. We further find that exploring hybrid AVM systems, fusing human
knowledge with ML models, is an interesting way for future research. By refining and combining real estate value
knowledge and expert domain experience into AVMs-driven valuation processes, these hybrid systems can increase
the reliability and accuracy of valuation outcomes, particularly in more complex or niche property markets. As more
real estate valuation index data become accessible, future research could replicate the current findings. Furthermore,
supplementary qualitative case studies at the national and international level, for instance, provide additional specific
data about transferred property, such as owners and their addresses, purchase prices, land valuations, land use laws,
and future modifications to land use laws. More in-depth information about the circumstances and structure of
successful AVM investments, taking into account institutional configurations, political environments, regional
contexts, and capability levels, would aid in the analysis and validation of the paper's findings.
More interdisciplinary research on the potential applications of AVMs in the real estate industry is needed. This
research should bridge the gap between the technological perspective and other relevant perspectives, such as those
related to politics, government, economics, the environment, organizations, law, education, and behavior. Since the
industry is inherently resistant to change, promoting advancement will probably need an interdisciplinary strategy. In
this case, some real estate market participants may benefit from the use of design science research tools and related
techniques. Finally, considering the substantial ecological impact of the real estate industry, new ideas and approaches
to climate change and sustainability need to be developed immediately.
Based on this study’s findings, we have proposed some open research questions for future researchers as follows:
• What are the factors and influences that delineate how these innovative technologies can shape performance
(driving accuracy), and efficiency in (scoring) AVMs for real estate transactions?
• What is the relationship between the accuracy of AVM property valuations and the quality, availability, and
integration of data from various sources?
• How will the broader adoption of AVMs in the real estate industry affect market patterns such as property
value trends, market stability, and stakeholder behavior in the future?
• What does it look like to confirm and have deep-seated biases in AVM approaches, be scrutinized, evaluated,
and develop mitigation strategies, ensuring unbiased and fair property assessments?
• What are the key drivers of consumer and real estate professional acceptance and use of AVMs, and how can
user-friendly AVM models be constructed to overcome the adoption barriers?
• How are different hurdles related to regulating and policymaking for AVMs, specifically about data privacy,
enforcement, and standardization in the real estate industry?
 • What roles do the number and quality of data inputs (transactional, economic, geospatial) play in determining
how well AVMs predict property prices?
• To what extent do AVMs affect the long-term trust of stakeholders, and in this case particularly how do these
models impact the cycle of a real estate market, as well as investors' strategies?
• What are the main bottlenecks in terms of AVM adoption and how can they be dealt with given low user
confidence and acceptance?
• In what ways are AVMs most accurate and the least transparent in the eyes of consumers and real estate
professionals, and how could they be improved to ensure better a more reliable product?
• What are the ethical considerations that stem from employing AVMs, specifically in terms of ensuring
transparency, fairness, and privacy within property valuations?
• What are the competitive implications for real estate markets of AVM integration with emerging
technologies, and what are the strategies by which businesses might respond to remain competitive?
• What implications can the adoption of AVMs have on property investment behaviors and market speculation
(especially for volatile or developing markets) in real estate over the long run?

4.2. Limitations

As with any research project, there are significant limitations to our study. First, the reading and selection of the
publications may have been biased subjectively. It's possible that important articles for analysis were left out of the
exclusion criteria used in the literature review technique as they were determined based on the goals of the papers.
From this perspective, we might have missed some literature because we chose articles from just two databases
(Scopus and WoS), even though they are both well-populated. Moreover, the knowledge gained from the content
analysis depends on how we understand the arguments made in the literature review, which could change in other
research projects of a similar kind. Secondly, only articles and reviews were included in this review, which ultimately
resulted in a smaller total number of publications under assessment. Therefore, researchers may choose to consider
more diverse papers in the future. Third, mapping all the strategic applications of AVMs in the real estate industry
requires more than one qualitative study.

5. Conclusion

The adoption of AVMs in valuation procedures has tremendous promise for influencing the future trajectory of the
real estate industry, which continues to play a crucial role in promoting economic growth and development. Therefore,
this study aimed to ascertain the state-of-the-art AVMs being used in the real estate sector as well as the strategic
applications of AVMs promoting data-driven decision-making in the sector. The results of the systematic literature
review showed that there has been a rise in publications recently. A total of 97 pertinent academic papers that discussed
or used AVM techniques for real estate valuation were identified and analyzed. The literature identified eight primary
sub-sectors related to real estate: land administration, leasing and renting, property transactions, real estate
development, real estate investment, real estate administration, and real estate maintenance. The conceptual
framework developed in this research will be insightful for researchers and practitioners who want to further research
in this area.
Ultimately, the integration of AVMs and real estate valuation has the potential to completely transform
conventional valuation procedures by providing novel approaches, resources, and understandings to successfully
negotiate the complexity of the contemporary real estate market. Stakeholders may leverage the transformative power
of AVMs to unlock value, reduce risks, and promote favorable outcomes for the real estate sector and the whole
economy by continuing to collaborate and innovate. Researchers can identify gaps in the literature on the use of AVMs
in real estate and conduct additional can-do research by using the data above. Similarly, the valuers and other real
estate stakeholders - from buyers and sellers to investors and policymakers - will find great practical value in the
insights gained from this study.

References

Abidoye, R. B., & Chan, A. P. C. (2017). Artificial neural network in property valuation: Application framework and research
trend. Property Management, 35(5), 554–571. [Link]
Abidoye, R. B., & Chan, A. P. C. (2018). Improving property valuation accuracy: A comparison of hedonic pricing model and
artificial neural network. Pacific Rim Property Research Journal, 24(1), 71–83.
[Link]
Arcuri, N., De Ruggiero, M., Salvo, F., & Zinno, R. (2020). Automated Valuation Methods through the Cost Approach in a BIM
and GIS Integration Framework for Smart City Appraisals. Sustainability, 12(18), Article 18.
[Link]
Atazadeh, B., Olfat, H., Rajabifard, A., Kalantari, M., Shojaei, D., & Marjani, A. M. (2021). Linking Land Administration Domain
Model and BIM environment for 3D digital cadastre in multi-storey buildings. Land Use Policy, 104, 105367.
[Link]
Aungkulanon, P., Hirunwat, A., Atthirawong, W., Phimsing, K., Chanhom, S., & Luangpaiboon, P. (2024). Optimizing
maintenance responsibility distribution in real estate management: A complexity-driven approach for sustainable
efficiency. Journal of Open Innovation: Technology, Market, and Complexity, 10(1), 100239.
[Link]
Aydinoglu, A. C., & Sisman, S. (2024). Comparing modelling performance and evaluating differences of feature importance on
defined geographical appraisal zones for mass real estate appraisal. Spatial Economic Analysis, 19(2), 225–249.
[Link]
Batista, P., & Marques, J. L. (2021). Automated Housing Price Valuation and Spatial Data. In O. Gervasi, B. Murgante, S. Misra,
C. Garau, I. Blečić, D. Taniar, B. O. Apduhan, A. M. A. C. Rocha, E. Tarantino, & C. M. Torre (Eds.), Computational
Science and Its Applications – ICCSA 2021 (pp. 366–381). Springer International Publishing.
[Link]
Baur, K., Rosenfelder, M., & Lutz, B. (2023). Automated real estate valuation with machine learning models using property
descriptions. Expert Systems with Applications, 213, 119147. [Link]
Bilge, E. C., & Yaman, H. (2021). Information management roles in real estate development lifecycle: Literature review on BIM
and IPD framework. Construction Innovation, 21(4), 723–742. [Link]
Cao, K., Diao, M., & Wu, B. (2019). A Big Data–Based Geographically Weighted Regression Model for Public Housing Prices:
A Case Study in Singapore. Annals of the American Association of Geographers, 109(1), 173–186.
[Link]
Carbonara, S., Faustoferri, M., & Stefano, D. (2021). Real Estate Values and Urban Quality: A Multiple Linear Regression Model
for Defining an Urban Quality Index. Sustainability, 13(24), Article 24. [Link]
Cardone, B., Di Martino, F., & Senatore, S. (2024). Real estate price estimation through a fuzzy partition-driven genetic algorithm.
Information Sciences, 667, 120442. [Link]
Chen, J., Wu, F., & Lu, T. (2022). The financialization of rental housing in China: A case study of the asset-light financing model
of long-term apartment rental. Land Use Policy, 112, 105442. [Link]
Chen, N. (2022). House Price Prediction Model of Zhaoqing City Based on Correlation Analysis and Multiple Linear Regression
Analysis. Wireless Communications and Mobile Computing, 2022(1), 9590704. [Link]
Cheng, J. C. P., Chen, W., Chen, K., & Wang, Q. (2020). Data-driven predictive maintenance planning framework for MEP
components based on BIM and IoT using machine learning algorithms. Automation in Construction, 112, 103087.
[Link]
Čirjevskis, A. (2021). Value Maximizing Decisions in the Real Estate Market: Real Options Valuation Approach. Journal of Risk
and Financial Management, 14(6), Article 6. [Link]
Dambon, J. A., Sigrist, F., & Furrer, R. (2021). Maximum likelihood estimation of spatially varying coefficient models for large
data with an application to real estate price prediction. Spatial Statistics, 41, 100470.
[Link]
Denyer, D., & Tranfield, D. (2009). Producing a systematic review. In The Sage handbook of organizational research methods (pp.
671–689). Sage Publications Ltd.
Despotovic, M., Koch, D., Stumpe, E., Brunauer, W. A., & Zeppelzauer, M. (2023). Leveraging supplementary modalities in
automated real estate valuation using comparative judgments and deep learning. Journal of European Real Estate
Research, 16(2), 200–219. [Link]
Doumpos, M., Papastamos, D., Andritsos, D., & Zopounidis, C. (2021). Developing automated valuation models for estimating
property values: A comparison of global and locally weighted approaches. Annals of Operations Research, 306(1), 415–
433. [Link]
Droj, G., Kwartnik-Pruc, A., & Droj, L. (2024). A Comprehensive Overview Regarding the Impact of GIS on Property Valuation.
ISPRS International Journal of Geo-Information, 13(6), Article 6. [Link]
El jaouhari, A., Arif, J., Samadhiya, A., Kumar, A., & Trinkūnas, V. (2023). Are we there or do we have more to do? Metaverse in
facility management and future prospects. International Journal of Strategic Property Management, 27(3), Article 3.
Evangelista, R., Ramalho, E. A., & Andrade e Silva, J. (2020). On the use of hedonic regression models to measure the effect of
energy efficiency on residential property transaction prices: Evidence for Portugal and selected data issues. Energy
Economics, 86, 104699. [Link]
Fazeli, A., Dashti, M. S., Jalaei, F., & Khanzadi, M. (2020). An integrated BIM-based approach for cost estimation in construction
projects. Engineering, Construction and Architectural Management, 28(9), 2828–2854. [Link] -
01-2020-0027
Foryś, I. (2022). Machine learning in house price analysis: Regression models versus neural networks. Procedia Computer Science,
207, 435–445. [Link]
Frodsham, M. (2024). Practice Briefing: The implications of a move towards explicit discounted cash flow (DCF) models for
property investment valuations. Journal of Property Investment & Finance, ahead-of-print(ahead-of-print).
[Link]
Gaur, A., & Kumar, M. (2018). A systematic approach to conducting review studies: An assessment of content analysis in 25 years
of IB research. Journal of World Business, 53(2), 280–289. [Link]
Ghosn, C., Warren-Myers, G., & Candido, C. (2024). Mapping the International Valuation Standards ESG criteria and sustainability
rating tools adopted at scale by the Australian commercial real estate market. Journal of Property Investment & Finance,
ahead-of-print(ahead-of-print). [Link]
Glumac, B., & Des Rosiers, F. (2020). Towards a taxonomy for real estate and land automated valuation systems. Journal of
Property Investment & Finance, 39(5), 450–463. [Link]
Gröbel, S., & Thomschke, L. (2018). Hedonic pricing and the spatial structure of housing data – an application to Berlin. Journal
of Property Research, 35(3), 185–208. [Link]
Horvath, S., Soot, M., Zaddach, S., Neuner, H., & Weitkamp, A. (2021). Deriving adequate sample sizes for ANN-based modelling
of real estate valuation tasks by complexity analysis. Land Use Policy, 107, 105475.
[Link]
Hoxha, V. (2023). Exploring the predictive power of ANN and traditional regression models in real estate pricing: Evidence from
Prishtina. Journal of Property Investment & Finance, 42(2), 134–150. [Link]
Indrajit, A., van Loenen, B., Ploeger, H., & van Oosterom, P. (2020). Developing a spatial planning information package in ISO
19152 land administration domain model. Land Use Policy, 98, 104111.
[Link]
Jafary, P., Shojaei, D., Rajabifard, A., & Ngo, T. (2024a). Automated land valuation models: A comparative study of four machine
learning and deep learning methods based on a comprehensive range of influential factors. Cities, 151, 105115.
[Link]
Jafary, P., Shojaei, D., Rajabifard, A., & Ngo, T. (2024b). Automating property valuation at the macro scale of suburban level: A
multi-step method based on spatial imputation techniques, machine learning and deep learning. Habitat International,
148, 103075. [Link]
Jiao, M., & Xu, H. (2022). How do Collective Operating Construction Land (COCL) Transactions affect rural residents’ property
income? Evidence from rural Deqing County, China. Land Use Policy, 113, 105897.
[Link]
Kamara, A. F., Chen, E., Liu, Q., & Pan, Z. (2020). A hybrid neural network for predicting Days on Market a measure of liquidity
in real estate industry. Knowledge-Based Systems, 208, 106417. [Link]
Kipper, L. M., Furstenau, L. B., Hoppe, D., Frozza, R., & Iepsen, S. (2020). Scopus scientific mapping production in industry 4.0
(2011–2018): A bibliometric analysis. International Journal of Production Research, 58(6), 1605–1627.
[Link]
Krämer, B., Stang, M., Doskoč, V., Schäfers, W., & Friedrich, T. (2023). Automated valuation models: Improving model
performance by choosing the optimal spatial training level. Journal of Property Research, 40(4), 365–390.
[Link]
Lee, C. L., Yam, S., Susilawati, C., & Blake, A. (2024). The Future Property Workforce: Challenges and Opportunities for Property
Professionals in the Changing Landscape. Buildings, 14(1), Article 1. [Link]
Lee, H., Han, H., Pettit, C., Gao, Q., & Shi, V. (2024). Machine learning approach to residential valuation: A convolutional neural
network model for geographic variation. The Annals of Regional Science, 72(2), 579–599.
[Link]
Leskinen, N., Vimpari, J., & Junnila, S. (2020). Using real estate market fundamentals to determine the correct discount rate for
decentralised energy investments. Sustainable Cities and Society, 53, 101953. [Link]
Li, X., Chen, J., & Ai, X. (2019). Contract design in a cross-sales supply chain with demand information asymmetry. European
Journal of Operational Research, 275(3), 939–956. [Link]
Liberati, A., Altman, D. G., Tetzlaff, J., Mulrow, C., Gøtzsche, P. C., Ioannidis, J. P. A., Clarke, M., Devereaux, P. J., Kleijnen, J.,
& Moher, D. (2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate
health care interventions: Explanation and elaboration. Journal of Clinical Epidemiology, 62(10), e1–e34.
[Link]
Lisi, G. (2019). Sales comparison approach, multiple regression analysis and the implicit prices of housing. Journal of Property
Research, 36(3), 272–290. [Link]
Liu, G. (2022). Research on Prediction and Analysis of Real Estate Market Based on the Multiple Linear Regression Model.
Scientific Programming, 2022(1), 5750354. [Link]
Malek, J., & Desai, T. N. (2020). A systematic literature review to map literature focus of sustainable manufacturing. Journal of
Cleaner Production, 256, 120345. [Link]
Matysiak, G. A. (2023). Assessing the accuracy of individual property values estimated by automated valuation models. Journal
of Property Investment & Finance, 41(3), 279–289. [Link]
Mete, M. O., & Yomralioglu, T. (2023). A Hybrid Approach for Mass Valuation of Residential Properties through Geographic
Information Systems and Machine Learning Integration. Geographical Analysis, 55(4), 535–559.
[Link]
Nor, M. I., & Raheem, M. M. (2024). Assessing the speculative dynamics and determinants of residential apartment rentals in
Mogadishu, Somalia: A hybrid modeling approach. Habitat International, 144, 102995.
[Link]
Ogunfowora, O., & Najjaran, H. (2023). Reinforcement and deep reinforcement learning-based solutions for machine maintenance
planning, scheduling policies, and optimization. Journal of Manufacturing Systems, 70, 244–263.
[Link]
Oliveira, T. C. de, Medeiros, L. de, & Detzel, D. H. M. (2021). Applying data mining algorithms to real estate appraisals: A
comparative study. International Journal of Housing Markets and Analysis, 14(5), 969–986.
[Link]
Oust, A., Hansen, S. N., & Pettrem, T. R. (2020). Combining Property Price Predictions from Repeat Sales and Spatially Enhanced
Hedonic Regressions. The Journal of Real Estate Finance and Economics, 61(2), 183–207.
[Link]
Özöğür Akyüz, S., Eygi Erdogan, B., Yıldız, Ö., & Karadayı Ataş, P. (2023). A Novel Hybrid House Price Prediction Model.
Computational Economics, 62(3), 1215–1232. [Link]
Pai, P.-F., & Wang, W.-C. (2020). Using Machine Learning Models and Actual Transaction Data for Predicting Real Estate Prices.
Applied Sciences, 10(17), Article 17. [Link]
Potrawa, T., & Tetereva, A. (2022). How much is the view from the window worth? Machine learning-driven hedonic pricing
model of the real estate market. Journal of Business Research, 144, 50–65. [Link]
Rampini, L., & Re Cecconi, F. (2021). Artificial intelligence algorithms to predict Italian real estate market prices. Journal of
Property Investment & Finance, 40(6), 588–611. [Link]
Reite, E. J. (2023). Mortgage lending valuation bias under housing price changes and loan-to-value regulations. Finance Research
Letters, 58, 104677. [Link]
Renigier-Biłozor, M., Janowski, A., & d’Amato, M. (2019). Automated Valuation Model based on fuzzy and rough set theory for
real estate market with insufficient source data. Land Use Policy, 87, 104021.
[Link]
Renigier-Biłozor, M., Źróbek, S., Walacik, M., Borst, R., Grover, R., & d’Amato, M. (2022). International acceptance of automated
modern tools use must-have for sustainable real estate market development. Land Use Policy, 113, 105876.
[Link]
Rey-Blanco, D., Zofío, J. L., & González-Arias, J. (2024). Improving hedonic housing price models by integrating optimal
accessibility indices into regression and random forest analyses. Expert Systems with Applications, 235, 121059.
[Link]
Rosenthal, S. S., Strange, W. C., & Urrego, J. A. (2022). JUE insight: Are city centers losing their appeal? Commercial real estate,
urban spatial structure, and COVID-19. Journal of Urban Economics, 127, 103381.
[Link]
Saldana-Perez, M., Guzmán, G., Palma-Preciado, C., Argüelles-Cruz, A., & Moreno-Ibarra, M. (2024). Geospatial modeling of
climate change indices at Mexico City using machine learning regression. Transforming Government: People, Process
and Policy, ahead-of-print(ahead-of-print). [Link]
Schirripa Spagnolo, F., Borgoni, R., Carcagnì, A., Michelangeli, A., & Salvati, N. (2024). A spatial semiparametric M-quantile
regression for hedonic price modelling. AStA Advances in Statistical Analysis, 108(1), 159–183.
[Link]
Sing, T. F., Yang, J. J., & Yu, S. M. (2022). Boosted Tree Ensembles for Artificial Intelligence Based Automated Valuation Models
(AI-AVM). The Journal of Real Estate Finance and Economics, 65(4), 649–674. [Link]
09861-1
Steurer, M., Hill, R. J., & Pfeifer, N. (2021). Metrics for evaluating the performance of machine learning based automated valuation
models. Journal of Property Research, 38(2), 99–129. [Link]
Su, T., Li, H., & An, Y. (2021). A BIM and machine learning integration framework for automated property valuation. Journal of
Building Engineering, 44, 102636. [Link]
Swietek, A. R. (2024). Using automated design appraisal to model building-specific devaluation risk due to land-use change.
Sustainable Cities and Society, 109, 105529. [Link]
Tajani, F., Morano, P., Salvo, F., & De Ruggiero, M. (2019). Property valuation: The market approach optimised by a weighted
appraisal model. Journal of Property Investment & Finance, 38(5), 399–418. [Link]
Tanrıvermiş, H. (2020). Possible impacts of COVID-19 outbreak on real estate sector and possible changes to adopt: A situation
analysis and general assessment on Turkish perspective. Journal of Urban Management, 9(3), 263–269.
[Link]
Tekouabou, S. C. K., Gherghina, Ş. C., Kameni, E. D., Filali, Y., & Idrissi Gartoumi, K. (2024). AI-Based on Machine Learning
Methods for Urban Real Estate Prediction: A Systematic Survey. Archives of Computational Methods in Engineering,
31(2), 1079–1095. [Link]
Thomé, A. M. T., Scavarda, L. F., & Scavarda, A. J. (2016). Conducting systematic literature review in operations management.
Production Planning & Control, 27(5), 408–420. [Link]
Trojanek, R., Gluszak, M., & Trojanek, M. (2024). Public land leases, reforms and (in)stability of municipal revenues in Poland –
The case of Poznan city. Cities, 148, 104877. [Link]
Valdez Gómez de la Torre, F. M., & Chen, X. (2024). Housing price determinants in Ecuador: A spatial hedonic analysis.
International Journal of Housing Markets and Analysis, ahead-of-print(ahead-of-print). [Link]
09-2023-0121
Vieira, E., & Gomes, J. (2009). A comparison of Scopus and Web of Science for a typical university. Scientometrics, 81(2), 587–
600. [Link]
Wan, W. X., & Lindenthal, T. (2023). Testing machine learning systems in real estate. Real Estate Economics, 51(3), 754–778.
[Link]
Wang, D., & Li, V. J. (2019). Mass Appraisal Models of Real Estate in the 21st Century: A Systematic Literature Review.
Sustainability, 11(24), Article 24. [Link]
Wang, R., & Rasouli, S. (2022). Contribution of streetscape features to the hedonic pricing model using Geographically Weighted
Regression: Evidence from Amsterdam. Tourism Management, 91, 104523.
[Link]
Wang, Y., Wang, S., Li, G., Zhang, H., Jin, L., Su, Y., & Wu, K. (2017). Identifying the determinants of housing prices in China
using spatial regression and the geographical detector technique. Applied Geography, 79, 26–36.
[Link]
Wei, C., Fu, M., Wang, L., Yang, H., Tang, F., & Xiong, Y. (2022). The Research Development of Hedonic Price Model-Based
Real Estate Appraisal in the Era of Big Data. Land, 11(3), Article 3. [Link]
Xia, H., Liu, Z., Efremochkina, M., Liu, X., & Lin, C. (2022). Study on city digital twin technologies for sustainable smart city
design: A review and bibliometric analysis of geographic information system and building information modeling
integration. Sustainable Cities and Society, 84, 104009. [Link]
Yalpir, S., Sisman, S., Akar, A. U., & Unel, F. B. (2021). Feature selection applications and model validation for mass real estate
valuation systems. Land Use Policy, 108, 105539. [Link]
Yasnitsky, L. N., Yasnitsky, V. L., & Alekseev, A. O. (2021). The Complex Neural Network Model for Mass Appraisal and
Scenario Forecasting of the Urban Real Estate Market Value That Adapts Itself to Space and Time. Complexity, 2021(1),
5392170. [Link]
Zaki, J., Nayyar, A., Dalal, S., & Ali, Z. H. (2022). House price prediction using hedonic pricing model and machine learning
techniques. Concurrency and Computation: Practice and Experience, 34(27), e7342. [Link]
Zhang, X., Ma, Y., & Wang, M. (2024). An attention-based Logistic-CNN-BiLSTM hybrid neural network for credit risk prediction
of listed real estate enterprises. Expert Systems, 41(2), e13299. [Link]
Zhang, Y., Xian, J., & Huang, M. (2020). Online leasing strategy for depreciable equipment considering opportunity cost.
Information Processing Letters, 162, 105981. [Link]
Zhou, Q., Shao, Q., Zhang, X., & Chen, J. (2020). Do housing prices promote total factor productivity? Evidence from spatial panel
data models in explaining the mediating role of population density. Land Use Policy, 91, 104410.
[Link]