Explainable Artificial Intelligence (XAI) - Precepts, Models, and Research Opportunities in Construction

Jul 30, 2026

Upload by Logic

Machine learning and deep learning generate outputs through complex “black box” mechanisms that are difficult to interpret. Explainable artificial intelligence (XAI) provides explanations for how and why model outputs are produced, helping reduce bias and error and improving confidence in decision-making. Limited construction uptake motivates this narrative review, which synthesizes XAI precepts and models and proposes a taxonomy. Suggested research opportunities are intended to address skepticism and hesitancy toward AI adoption in construction.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Background: AI, ML, and DL as black boxes

Need for explainability and risks of non-transparent models

Explainable Artificial Intelligence (XAI)

Definitions and lack of consensus

XAI in construction

Review scope and taxonomy of precepts and models

Advanced Engineering Informatics 57 (2023) 102024 Contents lists available at ScienceDirect Advanced Engineering Informatics journal [homepage: www.elsevier.com/locate/aei](homepage: www.elsevier.com/locate/aei) | Explainable artificial intelligence (XAI): Precepts, models, and opportunities for research in construction
Peter E.D. Love a, Weili Fangb, *, Jane Matthewsc, Stuart Portera, Hanbin Luod, Lieyun Ding d
a School of Civil and Mechanical Engineering, Curtin University, GPO Box U1987, Perth, Western Australia 6845, Australia b Department of Civil and Building Systems, Technische Universit¨at Berlin, Gustav-Meyer-Allee 25, 13156 Berlin, Germany c School of Architecture and Built Environment, Deakin University Geelong Waterfront Campus, Geelong, VIC 3220, Australia d School of Civil Engineering and Mechanics, Huazhong University of Science and Technology, Wuhan 430074, China | | | | | --- | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords: Construction Deep learning Explainability Interpretability Machine learning XAI | | Machine learning (ML) and deep learning (DL) are both branches of AI. As a form of AI, ML automatically adapts to changing datasets with minimal human interference. Deep learning is a subset of ML that uses artificial neural networks to imitate the learning process of the human brain. The ‘black box’ nature of ML and DL makes their inner workings difficult to understand and interpret. Deploying explainable artificial intelligence (XAI) can help explain why and how the output of ML and DL models are generated. As a result, understanding a model’s functioning, behavior, and outputs can be garnered, reducing bias and error and improving confidence indecision-making. Despite providing an improved understanding of model outputs, XAI has received limited attention in construction. This paper presents a narrative review of XAI and a taxonomy of precepts and models to raise awareness about its potential opportunities for use in construction. It is envisaged that the opportunities suggested can stimulate new lines of inquiry to help alleviate the prevailing skepticism and hesitancy toward AI adoption and integration in construction. | | 1. Introduction Artificial Intelligence (AI) is the mimicking of human intelligence processes by machines, especially computer systems. Two major offshoots of AI are Machine Learning (ML) and Deep Learning (DL), which can provide businesses with a wide range of benefits if applied appropriately [47]. The outputs generated by ML and DL models are produced by black boxes that are impossible to interpret [43]. In computer science, the term black box refers to a data-driven algorithm that produces useful information without revealing any information about its internal workings [100]. Consequently, AI models must be continuously monitored and managed to explain their use and the outputs of algorithms [10]. The difference between ML and DL needs to be highlighted at this juncture. In short, ML is a subfield of AI and can learn and adapt without following explicit instructions by using algorithms and statistical models to analyze and draw inferences from patterns of data. In the case ofDL, it is a subfield of ML whereby artificial neural networks (ANN) form the basis of its developed algorithms. The ANN comprises multiple layers that process data and extract progressively higher-level features. Data-driven ML and DL models are good at unearthing associationsand patterns in data, but causality cannot be guaranteed. Prior knowledge needs to be considered to infer causal relationships. The discovered associations emerging from AI-based models may be completely unexpected, not interpretable nor explainable [62]. Thus, with the rapid advancements in AI, people have begun to question the risks associated with its use and have sought to comprehend how an algorithm arrives ata solution [33]. As Anand et al. [7] cogently pointed out, ML

FAQs

Why is explainable artificial intelligence (XAI) needed for machine learning and deep learning models?

ML and DL often operate as “black boxes,” making their internal workings hard to interpret. XAI provides explanations for how and why outputs are generated, improving trust and decision-making while reducing bias and error.

How does the paper describe the relationship between ML and DL?

Machine learning is presented as a subfield of AI that learns from data patterns without explicit instructions. Deep learning is described as a subset of ML that uses artificial neural networks with multiple layers to extract progressively higher-level features.

What does the paper contribute regarding XAI in construction?

It provides a narrative review and a taxonomy of XAI precepts and models. The authors outline potential opportunities to stimulate further research and reduce skepticism about AI adoption in construction.

Jul 30, 2026

Upload by Logic