Project – The Influence of Artificial Intelligence-Based Risk Forecasting on Project Delivery Performance in Emerging Economies.

INTRODUCTION

1.1 Background to the Study

The construction industry plays a fundamental role in economic development through the provision of residential buildings, commercial facilities, transportation infrastructure, educational institutions, healthcare facilities, and other essential public assets. The successful delivery of construction projects contributes to employment generation, capital formation, urban development, and improved living standards. However, construction projects are often exposed to uncertainties arising from fluctuating material prices, labour shortages, design changes, funding constraints, adverse weather conditions, regulatory requirements, and coordination difficulties among project participants. These uncertainties can affect project completion time, expenditure, quality, and the achievement of intended objectives. Mansfield, Ugwu, and Doran (1994) established that construction projects in Nigeria experience substantial cost variations and time overruns, while Aibinu and Jagboro (2002) demonstrated that construction delays adversely affect completion time and cost. These findings illustrate the importance of effective risk identification, assessment, and forecasting in achieving successful project delivery. In contemporary construction management, risk forecasting is increasingly important because it allows project teams to anticipate possible problems and formulate preventive measures before uncertainties develop into significant project failures.

Artificial intelligence (AI) has emerged as an important technological development with applications in data analysis, pattern recognition, prediction, decision support, and automated problem-solving. In construction management, AI-based risk forecasting refers to the application of intelligent computational techniques to historical project records, current site information, financial data, scheduling information, procurement records, and other relevant variables to estimate the likelihood and potential consequences of future project risks. Such techniques may include machine learning, artificial neural networks, natural language processing, optimisation algorithms, and other predictive analytical methods. Unlike approaches that rely exclusively on periodic reports or professional judgement, AI-based forecasting can identify relationships among multiple variables and update predictions as new information becomes available. Tian et al. (2025), in their systematic literature review of artificial intelligence in construction risk management, identified applications of machine learning in predictive modelling, natural language processing in document and compliance risk management, and optimisation algorithms in resource and scheduling decisions. These applications suggest that AI-based risk forecasting can support more informed project planning and control. Nevertheless, its effectiveness depends on the quality of available data, the suitability of the selected models, and the capacity of project personnel to interpret and apply the forecasts.

Construction project risk management involves the systematic identification, analysis, evaluation, treatment, and monitoring of uncertainties that may influence project objectives. Traditionally, construction professionals have used risk registers, expert assessments, probability-impact matrices, contingency allowances, progress reports, and periodic performance reviews to manage project risks. Although these methods remain valuable, they may be insufficient when projects involve large volumes of data, complex interactions among risk factors, or rapidly changing operating conditions. AI-based forecasting provides an opportunity to strengthen these conventional methods by detecting emerging patterns and estimating the likelihood of events such as schedule slippage, cost escalation, material shortages, equipment breakdown, and productivity decline. Tian et al. (2025) found that AI applications in construction risk management extend across cost, safety, schedule, quality, and supply-chain concerns, demonstrating the breadth of possible forecasting applications. Similarly, Wuni (2026) examined critical risks associated with AI implementation in construction projects and identified technical, financial, managerial, knowledge-related, legal, and data-related dimensions. This evidence indicates that AI can support construction risk management, but its adoption also requires deliberate attention to technological reliability, organisational readiness, and responsible decision-making.

Project delivery performance refers to the extent to which a construction project achieves its planned objectives within the specified time, approved budget, required quality standards, and agreed scope. Although the traditional measures of project success emphasise time, cost, and quality, project delivery performance may also incorporate safety, resource utilisation, client satisfaction, stakeholder expectations, and the functional suitability of the completed facility. A project that is completed but exceeds its budget substantially or fails to meet essential quality requirements cannot be regarded as fully successful. In the Nigerian construction industry, Aibinu and Jagboro (2002) found that delays had significant effects on the completion cost and duration of the building projects they studied, illustrating the close relationship between risk-related disruptions and delivery outcomes. Consequently, the value of AI-based risk forecasting should be assessed not merely by its ability to generate predictions but also by whether its outputs help project teams make timely interventions and improve actual delivery performance. The present study therefore considers whether AI-based forecasting of project risks is associated with improved cost performance, schedule adherence, and achievement of specified construction quality requirements in selected projects in the Federal Capital Territory (FCT), Abuja.

Nigeria’s construction industry operates within an environment in which projects may be affected by funding difficulties, changes in material prices, procurement delays, design deficiencies, contractual disputes, labour-related challenges, and weaknesses in project coordination. These factors can interact, making it difficult for project managers to predict the eventual effects of individual risks using isolated assessments. Mansfield et al. (1994) investigated causes of delays and cost overruns in Nigerian construction projects and documented substantial deviations from initial contract expectations. Aibinu and Jagboro (2002) also found that delays were associated with cost and time overruns, disputes, litigation, and, in some cases, project abandonment. Although these studies provide important evidence about longstanding delivery problems, they do not establish the current level of AI adoption or the effectiveness of AI-based risk forecasting in contemporary Nigerian construction projects. This distinction is important because the availability of digital technologies does not automatically translate into better project outcomes. There is therefore a need to investigate how construction professionals use predictive technologies, whether such technologies inform project decisions, and the extent to which forecasting contributes to improved delivery performance under Nigerian operating conditions.

Emerging economies face particular opportunities and constraints in adopting advanced digital technologies for infrastructure and construction management. AI-based forecasting may help organisations make better use of historical project data, identify recurring risk patterns, allocate resources more efficiently, and respond to emerging problems before they generate substantial losses. However, adoption may be constrained by inadequate digital infrastructure, fragmented project records, limited access to specialised expertise, implementation costs, inconsistent data standards, and concerns about the reliability and transparency of algorithmic predictions. The World Bank (2026), in its World Development Report 2026: The Promise of Artificial Intelligence, emphasised that AI could support productivity in developing economies while highlighting the importance of addressing constraints involving connectivity, electricity, skills, and institutional capacity. In construction, these broader concerns are particularly relevant because predictive systems require sufficiently accurate and consistent information about project costs, schedules, procurement, labour, and progress. Consequently, the influence of AI-based risk forecasting in emerging economies should not be assumed to be identical to its influence in highly digitised construction markets. Empirical investigation is needed to establish whether the technology improves project delivery and what organisational or technical conditions strengthen or weaken its contribution.

The Federal Capital Territory, Abuja, provides a relevant setting for examining the relationship between AI-based risk forecasting and construction project delivery performance because it contains public and private developments associated with residential accommodation, commercial activities, administrative functions, education, healthcare, and urban infrastructure. Construction projects within the territory may involve multiple stakeholders, different procurement arrangements, complex design requirements, and varying levels of digital capability. These characteristics make effective risk forecasting potentially valuable for anticipating project delays, controlling cost variations, monitoring construction progress, and supporting timely managerial intervention. However, evidence that AI-based risk forecasting is technically promising at the international level does not establish how extensively it is used or how effective it is among selected construction projects in Abuja. Tian et al. (2025) identified opportunities for AI applications in construction risk management while highlighting continuing challenges associated with data quality, model interpretability, and workforce skills. In view of these concerns, this study focuses on selected construction projects in the FCT to examine the extent of AI-based risk forecasting adoption, its contribution to identifying and predicting project risks, and its relationship with project delivery performance. The study is intended to provide context-specific evidence that may guide construction professionals, project owners, contractors, consultants, and relevant public institutions in determining how predictive technologies can be applied effectively within the Nigerian construction environment.

1.2 Statement of the Problem

Despite the importance of the construction industry to economic development, the successful delivery of construction projects remains a major concern in Nigeria. Construction projects may experience delays, cost overruns, disputes, variations in scope, and difficulties in meeting specified quality standards. Mansfield et al. (1994) identified substantial cost variations and time overruns in Nigerian construction projects, while Aibinu and Jagboro (2002) demonstrated the adverse effects of delays on project completion time and cost. These problems suggest that weaknesses in anticipating and managing project risks can undermine the achievement of construction objectives. Although conventional risk management practices remain important, the extent to which they enable project teams to anticipate interconnected risks and respond before significant disruptions occur requires further investigation, particularly in projects operating under changing economic and organisational conditions.

A second problem concerns the limitations of relying predominantly on traditional approaches to risk assessment and project monitoring. Construction risks can develop through interactions among material prices, procurement schedules, labour productivity, funding availability, design changes, and site conditions. When these factors are assessed separately or reviewed only after a problem becomes evident, project managers may have limited opportunity to implement preventive measures. AI-based risk forecasting offers a potential means of identifying patterns in project data and predicting emerging risks before they significantly affect delivery outcomes. Tian et al. (2025) identified machine learning, natural language processing, and optimisation techniques as relevant tools for construction risk management. Nevertheless, the availability of these technologies does not establish that they are being effectively adopted by construction organisations in Abuja or that their predictions consistently result in improved project performance. The practical relationship between predictive capability and actual managerial action therefore remains an important issue for investigation.

A third problem relates to the challenges of implementing AI-based forecasting in emerging economies. The usefulness of predictive systems may depend on the availability of reliable project records, appropriate digital infrastructure, skilled personnel, organisational support, and the capacity to interpret algorithmic outputs. Wuni (2026) identified technical, financial, managerial, knowledge-related, legal, and data-related risks associated with AI implementation in construction projects. Similarly, the World Bank (2026) highlighted the importance of infrastructure, skills, connectivity, and institutional capacity in enabling developing economies to benefit from artificial intelligence. These concerns raise questions about whether construction organisations in the FCT possess the necessary capabilities to adopt AI-based forecasting and whether barriers to implementation reduce its potential contribution to project delivery performance. Without examining these issues, recommendations for adopting AI in construction management may overlook the practical conditions that determine whether the technology can deliver meaningful benefits.

The fourth problem is the need for context-specific empirical evidence on the influence of AI-based risk forecasting on the performance of selected construction projects in the Federal Capital Territory, Abuja. Existing studies establish that construction risks contribute to delivery difficulties and that AI offers potential applications in risk prediction and management; however, these findings do not, by themselves, demonstrate the strength or direction of the relationship between AI-based forecasting and project delivery outcomes among selected construction projects in Abuja. In particular, the extent to which forecasting supports cost control, schedule adherence, and quality achievement, as well as the constraints affecting its practical application, requires focused investigation. This study therefore seeks to examine the influence of AI-based risk forecasting on project delivery performance in selected construction projects in the FCT, Abuja, with a view to generating evidence that can inform construction management practices and technology adoption decisions in Nigeria.

1.3 Aim and Objectives of the Study

1.3.1 Aim of the Study

The aim of this study is to examine the influence of artificial intelligence-based risk forecasting on project delivery performance in selected construction projects in the Federal Capital Territory, Abuja.

1.3.2 Specific Objectives

The specific objectives are to:

1. Examine the extent of adoption of artificial intelligence-based risk forecasting among selected construction projects in the Federal Capital Territory, Abuja.

2. Assess the influence of AI-based risk forecasting on cost performance in selected construction projects in the FCT, Abuja.

3. Determine the influence of AI-based risk forecasting on schedule adherence and timely completion of selected construction projects in the FCT, Abuja.

4. Identify the major challenges affecting the implementation and effective utilisation of AI-based risk forecasting in selected construction projects in the FCT, Abuja.

1.4 Research Questions

The following research questions will guide the study:

1. To what extent is artificial intelligence-based risk forecasting adopted in selected construction projects in the Federal Capital Territory, Abuja?

2. How does AI-based risk forecasting influence cost performance in selected construction projects in the FCT, Abuja?

3. To what extent does AI-based risk forecasting influence schedule adherence and timely project completion in selected construction projects in the FCT, Abuja?

4. What are the major challenges affecting the implementation and effective utilisation of AI-based risk forecasting in selected construction projects in the FCT, Abuja?

1.5 Research Hypothesis

The following null hypothesis is formulated in line with the central purpose of the study:

H₀: Artificial intelligence-based risk forecasting has no statistically significant influence on project delivery performance in selected construction projects in the Federal Capital Territory, Abuja.

The hypothesis may be tested using an appropriate statistical technique based on the study design, measurement of the variables, and data collected from respondents and, where available, project records.

1.6 Significance of the Study

The findings of this study are expected to be useful to the following groups.

1.6.1 Construction Project Managers

The study may help project managers understand how AI-based forecasting can support early identification of risks, improve project monitoring, and strengthen decisions relating to cost, schedules, procurement, and resource allocation. The findings may also help managers identify the conditions under which forecasting outputs can be translated into practical preventive measures.

1.6.2 Contractors and Construction Firms

The study may provide contractors with information on the potential value of adopting AI-based risk forecasting in their project management processes. By examining the relationship between forecasting and delivery performance, the findings may assist firms in evaluating whether investments in predictive tools, digital record systems, and staff training are justified by their operational needs.

1.6.3 Clients and Project Owners

Public and private project owners may benefit from a clearer understanding of how predictive risk management can support project oversight, financial planning, progress monitoring, and accountability. The study may also inform decisions concerning reporting requirements, risk registers, digital project records, and the use of forecasting tools by consultants and contractors.

1.6.4 Government and Policy-Making Institutions

The study may provide context-specific evidence for institutions responsible for construction procurement, infrastructure development, public project monitoring, and digital transformation. Its findings may inform policies and guidelines that encourage responsible technology adoption while addressing challenges involving data quality, technical capacity, transparency, and implementation costs.

1.6.5 Academic Researchers and Students

The study may contribute to academic discussions on artificial intelligence, construction risk management, project performance, and technology adoption in emerging economies. It may also provide a foundation for further studies examining AI applications in other Nigerian states, infrastructure categories, procurement systems, and project delivery environments.

1.6.6 Technology Developers and Consultants

Developers of construction management software and digital advisory services may use the findings to better understand the operational requirements of construction firms in Abuja. The study may help identify user concerns and implementation barriers that should be considered when developing predictive tools suited to local construction practices.

1.7 Scope of the Study

The study is limited to the influence of artificial intelligence-based risk forecasting on project delivery performance in selected construction projects in the Federal Capital Territory, Abuja, Nigeria.

Content scope: The study covers the adoption of AI-based risk forecasting, its relationship with cost performance, its influence on schedule adherence and timely completion, and the challenges affecting its implementation and utilisation.

Geographical scope: The geographical area is the Federal Capital Territory, Abuja. The study will focus on selected construction projects within the territory rather than all construction projects in Nigeria.

Respondent scope: Depending on access and the availability of relevant information, respondents may include project managers, quantity surveyors, civil engineers, architects, construction managers, site supervisors, contractors, consultants, and other professionals involved in project planning, risk management, and delivery.

Methodological scope: The study may adopt a survey-based design, supported where feasible by relevant project records. Data may be collected through structured questionnaires and, where access is granted, documentary reviews of cost estimates, approved schedules, progress reports, risk registers, and completion records. The final method of analysis will depend on the research design, sample size, measurement scales, and availability of reliable data.

The study does not assume that every selected project already uses AI-based forecasting. The extent of adoption will be established empirically, and the investigation will distinguish between actual AI-based predictive applications and conventional digital tools that do not employ AI forecasting.

1.8 Limitations of the Study

The study may encounter limitations relating to access to confidential project information, incomplete historical records, differences in the level of AI adoption, and the willingness of professionals to provide information. Some respondents may have limited experience with AI-based forecasting, which could affect their ability to assess its practical contribution. In addition, projects may differ in size, complexity, procurement method, funding arrangements, and stage of completion, potentially affecting comparisons of delivery performance.

To minimise these limitations, the researcher should clearly define AI-based risk forecasting, select respondents with relevant professional experience, protect the confidentiality of participants, and use available documentary evidence to support questionnaire responses. Where project records are unavailable, the study should acknowledge its reliance on respondents’ assessments rather than presenting perceptions as independently verified project outcomes.

The interpretation of findings should also recognise that an observed relationship between forecasting and delivery performance does not necessarily establish causation. Other factors, including project financing, managerial competence, contractor experience, and changes in economic conditions, may influence project outcomes.

1.9 Operational Definition of Terms

The following terms are defined according to how they are used in this study.

Artificial Intelligence (AI): The application of computational technologies capable of performing tasks such as pattern recognition, data analysis, prediction, and decision support that would otherwise require substantial human analytical effort.

AI-Based Risk Forecasting: The use of AI techniques, particularly machine learning and predictive analytics, to analyse project information and estimate the likelihood or potential consequences of risks that may affect construction project objectives.

Construction Project: A planned undertaking involving the construction, alteration, renovation, or development of a building or infrastructure facility within the Federal Capital Territory, Abuja.

Risk Management: The systematic process of identifying, analysing, evaluating, treating, and monitoring uncertainties that may affect project objectives.

Project Delivery Performance: The extent to which a construction project achieves its intended objectives in terms of cost, completion time, quality, and agreed scope.

Cost Performance: The extent to which actual or forecast project expenditure conforms to the approved project budget and cost baseline.

Schedule Adherence: The extent to which construction activities and overall project completion conform to the approved project schedule.

Emerging Economies: Economies undergoing structural and industrial transformation, with expanding economic activity and development needs but potentially uneven access to technology, infrastructure, capital, and specialised skills.

Technology Adoption: The process through which an organisation evaluates, acquires, integrates, and uses a technology within its operations.

Selected Construction Projects: The building or infrastructure projects deliberately chosen for investigation within the Federal Capital Territory, Abuja, according to the study’s selection criteria.

1.10 Organisation of the Study

The research is organised into five chapters.

Chapter One presents the introduction, background to the study, statement of the problem, aim and objectives, research questions, research hypothesis, significance, scope, limitations, and operational definitions of terms.

Chapter Two reviews relevant conceptual, theoretical, and empirical literature on artificial intelligence-based risk forecasting, construction risk management, project delivery performance, and technology adoption in emerging economies. It also identifies the research gap and presents the conceptual framework.

Chapter Three explains the research methodology, including the research design, study population, sample size and sampling technique, sources of data, research instruments, validity and reliability procedures, methods of data collection, and techniques for data analysis.

Chapter Four presents the data collected, analyses the results in line with the research objectives, tests the hypothesis, and discusses the findings in relation to relevant literature.

Chapter Five summarises the findings, presents the conclusion, provides recommendations, identifies the study’s contribution, and suggests areas for further research.

Project – The Influence of Artificial Intelligence-Based Risk Forecasting on Project Delivery Performance in Emerging Economies.