The discourse surrounding managerial effectiveness, organisational agility and AI within the mining sector is both multifaceted and expansive, encompassing a broad spectrum of theoretical perspectives, practical applications and critical case studies.

Understanding the foundations of this discussion requires exploring several key theoretical lenses:

Managerial effectiveness is critical for the growth and survival of any firm, particularly in the high-stakes, capital-intensive mining industry (Rastogi 2018). Bartol and Martin (1991), as cited in Rastogi (2018), define managerial effectiveness as the ability of an organisation to set and achieve its goals. This definition emphasises the importance of both strategic planning and execution.

Managerial effectiveness is not only about achieving organisational goals but also about effectively utilising resources and skills. According to Anandan et al. (2017), managerial effectiveness involves using managerial skills and strategies through the workforce to meet organisational requirements. This perspective highlights the importance of leadership in driving productivity and fostering a collaborative work environment.

Organisational agility is increasingly vital for maintaining a competitive advantage in today’s dynamic market environment. Žitkienė and Deksnys (2018) note that globalisation, rapid technological advancements, competition, disruptive business models and evolving consumer preferences constantly challenge organisations. These challenges necessitate agile strategies to adapt and thrive.

Talbolt et al. (2015) argue that agility is no longer optional but a necessity for business organisations in the 21st century. The ability to respond swiftly to external environmental changes distinguishes successful organisations from those that struggle. This responsiveness is crucial for mining companies facing volatile commodity markets and regulatory landscapes.

Organisational agility is not a static state but an ongoing process of adaptation and innovation. Alzoubi (2011) suggests that while a firm can become increasingly agile, it can never be concretely agile because the business environment is continually evolving. This continuous evolution underscores the need for mining companies to embrace a culture of innovation and continuous improvement.

Organisational agility and innovation are closely linked, with agile organisations often characterised by a culture of innovation. Harraf (2013), as cited in Talbolt et al. (2015), defines a culture of innovation as one where the organisation is constantly evaluating its systems, structures, procedures and teams to find new and improved methods for performing functions and providing services.

Artificial intelligence is transforming how organisations are managed by enabling machines to perform tasks previously done by humans (Chewrnova & Chernov 2018). This technological shift has profound implications for managerial effectiveness and organisational agility. Scarcello (2019) notes that AI is becoming essential for the success of many organisations and is one of the most impactful technologies of the 21st century.

Machine learning (ML), which is a subset of AI, allows software applications to become more accurate at predicting outcomes. Burns (2023) defines ML as a type of AI that enables software applications to improve their accuracy in predicting outcomes. Similarly, Nilson (1998) views ML as the changes in systems that perform tasks associated with AI. This predictive capability can significantly enhance decision-making processes in the mining industry.

Artificial intelligence drives organisational agility by enabling companies to adapt quickly to changing conditions and make informed decisions based on data analysis. By adopting these technologies, mining companies can improve efficiency, reduce costs and enhance their overall performance. In addition to that, Scarcello (2019:42) asserted that ‘Artificial Intelligence (AI) is proving to be essential for the success of many organisations and one of the most impactful technologies of the 21st century’. Other authors claim that AI has ushered in fourth industrial revolution. This means that for an organisation which is not agile, it will be difficult to compete because AI will bring with it a host of advantages, which include low cost of production, quicker production turnaround, quality output and many more benefits brought by AI whose name was postulated by McCarthy in 1956.

Burns (2023) explains that ‘Machine learning (ML) is a type of AI that allows software applications to become more accurate at predicting outcomes’. This resonates with Nilson’s (1998) view that ‘Machine learning usually refers to the changes in systems that perform tasks associated with AI’.

Because of the nature of mining operations, which are costly and unsafe as well as unpredictable in terms of the market variable, there has been a growing trend in the studies with regard to the evolution of the way mining operations should be managed through the adoption of AI. The studies have started from those who have to do with the definition of AI, forms of AI, application of AI and the effects of AI as well as ethical considerations with regard to AI use in the mining industry.

We looked for studies in three major online libraries: Web of Science, Scopus and IEEE Xplore. These libraries cover a wide range of scientific and technical topics relevant to AI and mining. We chose these libraries to find as many useful studies as possible from different areas such as computer science, engineering, geology and business.

Web of Science: We used Web of Science because it has a good reputation for indexing important journals in many fields. This helped us to find key research and new trends in AI or ML for mining.

Scopus: Scopus covers even more journals and conference articles than Web of Science, so we used it to find a broader range of studies, including those from smaller journals and conferences.

IEEE Xplore: This library focuses on electrical engineering and computer science, so we used it to find specific technical details about AI algorithms and how they are used in mining.

Our search was designed to find different types of studies, including research with data, theoretical ideas, case studies and reviews. We started with a test search to make sure our search terms worked well. We adjusted the search terms as needed based on the test results and as we continued to monitor the available literature.

We used the following search terms:

We chose these terms to cover the different ways AI is used in the mining industry, including both technical and management aspects. Using the word ‘AND’ ensured that we only found studies that discussed both AI and mining. Terms such as ‘Predicting Breakdowns’ and ‘Improving Efficiency’ helped us find studies about specific AI applications in mining operations, while ‘Location Analysis’, ‘Better Management’ and ‘Changing Quickly’ helped us find studies about spatial analysis, management effectiveness and organisational flexibility.

We only looked at studies published between 2019 and 2025 to focus on the most recent developments. This timeframe ensured that we captured the most current research and applications of AI in the mining sector. We focused on this timeframe because technology is rapidly changing, and we wanted our findings to be relevant to current industry practices and challenges.

We also used backward and forward citation analysis to find articles that our main search might have missed. This involved checking the references of the studies we found and tracking who cited those studies in later publications. This helped us to make sure our review was as complete as possible.

To make sure we only included the most relevant and high-quality studies, we used specific rules for including and excluding studies. These rules were based on our research goals and the scope of our review, ensuring that we selected studies that provided valuable information about the role and impact of AI on management and flexibility in the mining industry.

After searching and screening the studies, we used a standard form to collect important information from each study. The form was designed to capture both numbers and descriptions, giving us a complete picture of the study’s goals, methods, findings and limitations.

The form included details such as:

We then summarised the data using descriptions and a simple t-test to compare mines that use AI versus those that do not. Descriptive synthesis gave us a general overview of the main ideas and findings across the studies. We identified common themes and insights related to AI applications, benefits, challenges and spatial considerations.

To address the study objective assessing how AI-enabled, spatially informed decision-making enhances managerial effectiveness and organisational agility in mining, we begin with a foundational quantitative comparison of average outcomes. This quantitative comparison of average outcomes that is usually generated through the use of various AI models such as descriptive AI, predictive AI and prescriptive AI is a hallmark of enhancing mining companies to be effectively managed and responsive to the various variable in its environment. However, this comparison is intentionally embedded within the SAPF so that observed average differences are interpreted as manifestations of the mechanisms described by SAPF (spatial decision integration, rapid re-planning and cross-zone AI-enabled coordination), rather than as isolated statistical artefacts. The progression moves from a simple average comparison towards analyses that reveal spatial and causal links governing performance across the mine network. In essence, the quantitative comparison informs why AI should be adopted because they are a guide to the doubting mining companies who are laggards in adopting this new technology:

The t-test was chosen as a simple and widely used statistical test for comparing the means of two independent groups. The t-test provides a p-value, which indicates the probability of finding the observed difference in means (or a more extreme difference) if there is no true difference between the groups. A p-value less than a pre-determined significance level (e.g. 0.05) is typically considered statistically significant, indicating that there is strong evidence to reject the null hypothesis of no difference between the groups:

The interpretation of the t-test results was conducted in the context of the specific metrics being analysed and the limitations of the data. We acknowledged that correlation does not equal causation and that the observed differences between the groups may be because of other factors not accounted for in the analysis. The interpretation was also guided by the existing literature and the input of industry experts, ensuring that the conclusions were meaningful and relevant to the mining industry.

The following is a more technical breakdown of the components (Equation 1): Y(l)=X(l)β+ε(l)[Eqn 1] where:

Now, the important part: because things close together are often more similar, the error ε(l) is related to the error at nearby locations. A simple way to model this is to say (Equation 2): ε(l)=ρ (average ε of nearby locations)+noise[Eqn 2]

Here, ρ (rho) tells you how strong this relationship is. If ρ is close to 1, it means that if the error is high in one place, it is likely to be high in nearby places too. Noise is just random stuff that is not related to anything else.

To ensure our findings are trustworthy, we took several steps:

Our systematic review, conducted according to our methodology, identified 87 relevant studies published between 2019 and 2025. These studies provide a comprehensive overview of how AI is currently being used in the mining industry. Some of the selected studies are categorised by their description in Table 2.

To gain an in-depth understanding of the application of AI in the mining industry and how it may enhance managerial effectiveness and organisational agility, various studies regarding the origins of AI, how it is applied, the types of AI that mining firms can use, the impacts of the new technologies in the form of AI, ethical considerations with regard to the adoption of AI as well as the role played by AI to foster the adoption of a CBM were analysed. This was done to ensure that the novel SAPF for the use of AI will be designed based on the literature available on AI application in the mining industry. The idea was to eliminate all caveats and limitations that may arise because of lack of sufficient data.

Artificial intelligence-circular business model is important in mining companies, especially when it can help the company to create, capture and deliver value to improve resource efficiency by extending the lifespan of the mineral products and parts to improve environmental, social and economic benefits. Three roles must be played by AI in mining CBM, which are to slow, narrow and close the loops (Ritala, Bocken & Konietzko 2023). Mining companies must use AI to slow resource loops, that is, to increase the life of the mineral products and its components. A mining company can use AI-CBM to reduce resource flow that ultimately reduces costs, which is referred to as narrowing the business model, and this can be done through increasing efficiency in the value chain. Artificial intelligence–enabled closing business models focus on closing the loop of resources post consumptions of the resources and this can be done through reuse, remanufacture and recycle. The identified studies showed that AI is being applied in many different areas of mining. We will explore these areas in the next section:

The reviewed studies consistently showed several benefits of using AI in mining:

Despite the benefits, there are also challenges to adopting AI in mining:

The spatially aware predictive framework offers a robust and data-driven approach to enhancing managerial effectiveness and organisational agility in the mining industry. To simplify the SAPF, it is important to start by defining the concept of being spatially aware. A plethora of definitions exist in literature, which ranges from authors viewing spatial awareness as the understanding of the spatial dimensions and configurations of both the environment and the objects within it (Mihilewizic 1999). Ishikawa (2023) postulates that spatial awareness is the cognitive mapping, which involves acquiring, coding, recalling and decoding information about a relative location and attributes of phenomena in everyday spatial. While this definition by Ishikawa (2023) has managed to relate spatiality to place and location, a more comprehensive explication of spatial awareness would be to define it as the understanding of oneself and relationships within a given space (Gross et al. 2025). This means that from a circular mining business model, the concept of spatiality involves using new technologies such as AI in line with the attributes existing in that mining location. This eliminates a haphazard approach to implementation of AI. It addresses the limitations of current AI implementations by integrating spatial analysis techniques, providing a more comprehensive and nuanced understanding of the factors influencing mining performance. This allows managers to make more informed decisions, optimise resource allocation and improve overall operational efficiency. This framework moves beyond the theoretical by providing a practical roadmap for implementation. The integration of spatial awareness is not merely an add-on but a fundamental shift in how AI is applied, recognising that mining operations are inherently spatial processes.

This article is skewed towards diagnostic AI, which falls under business analytics but also leans towards harnessing the benefits of prescriptive analytics, which entails the prescription of the best option that mining companies can take to enjoy managerial effectiveness and organisational agility. Consequently, for effectiveness, the SAPF must have the following components:

The use of these components of data integration, spatial data analysis, predictive modelling, optimisation algorithms, visualisation and decision support gives room for an AI-CBM, which incorporates the benefits of predictive AI, descriptive AI and prescriptive AI. In this case, the AI-CBM mining organisation will be able to be managed effectively and as such be responsive to a dynamic business environment.

Spatial regression models relationships between variables while accounting for spatial dependence. If you want to determine if AI adoption really improves mine output, you use the model to determine:

Ethical clearance to conduct this study was obtained from the University of Zambia Biomedical Research Ethics Committee on 10 December 2024 (No. 6030-2024).

The mining company concludes that AI-driven predictive maintenance is effective in reducing maintenance costs. Additionally, they learn that supplier relationships and equipment management strategies are critical.

The SAPF offers a robust and data-driven approach to enhancing managerial effectiveness and organisational agility in the mining industry. It addresses the limitations of current AI implementations by integrating spatial analysis techniques, providing a more comprehensive and nuanced understanding of the factors influencing mining performance. This allows managers to make more informed decisions, optimise resource allocation and improve overall operational efficiency. This framework moves beyond the theoretical by providing a practical roadmap for implementation. The integration of spatial awareness is not merely an add-on but a fundamental shift in how AI is applied, recognising that mining operations are inherently spatial processes.

The success of the framework hinges on several key factors:

The implementation of the framework should be phased, starting with pilot projects in specific areas of the mining operation. This allows for iterative refinement and adaptation of the framework to the specific needs of the organisation.

Managerial effectiveness and organisational agility are paramount for success in the increasingly complex and dynamic mining industry. The integration of advanced technologies such as AI offers a powerful means to enhance operational efficiency, improve decision-making and achieve sustainable growth. The SAPF provides a practical roadmap for implementing these technologies, realising their full potential and moving beyond fragmented and spatially unaware applications.

The framework’s emphasis on spatial data analysis and integration addresses a critical gap in the current literature, recognising the inherent spatial nature of mining operations. By explicitly incorporating spatial considerations into the data integration, modelling and decision-making processes, the framework enables more informed and effective management of mining resources and operations.

While challenges remain, the potential benefits of the framework are substantial, including increased operational efficiency, improved decision-making, enhanced sustainability and improved safety. By embracing a data-driven and spatially-aware approach, mining companies can enhance their competitiveness, improve their environmental performance and create a safer and more sustainable working environment.

This study, while providing a comprehensive overview and a practical framework, is subject to certain limitations that warrant acknowledgement. These limitations also point towards promising avenues for future research:

Future research should be directed towards:

The findings of this study carry significant implications for various stakeholders in the mining ecosystem, including mining companies, government policymakers and technology providers.