1. Introduction

1.1 Background

Like most industries the manufacturing sector is changing rapidly as a result of new technology. Research is currently talking about industry 4.0 and iteration of the industrial revolution where cyber-physical systems, the Internet of Things (IoT), and artificial intelligence (AI) are changing the traditional ways of producing goods and services and the process of working (Alaref, 2025). HRM is drifting towards a state where it can be categorized as people analytics (Cayrat and Boxall, 2022). According to Cayrat and Boxall (2022), people analytics leverages algorithmic hiring to ensure that the company is able to manage and process the high volume of recruitment inputs needed to manage an industrial environment. As a form of definition, algorithmic hiring can be described as the use of statistical models and machine learning to automate the sourcing, screening, and selection of candidates (Hosain et al., 2025). If algorithmic hiring is placed in the context of manufacturing, it becomes apparent that efficiency is an important driver for its adoption. This is because typically the manufacturing industry has been noted to suffer critical skills shortage and turnover of entry-level staff (Sulaiman-Oloko, Agu and Ann, 2026). Theoretically, the automation of the first stages of the process of filtering candidates helps organizations to reduce two important recruitment metrics: time-to-hire and the cost per acquisition (Raji et al., 2024). HR can then focus on high-level strategic decision-making of the business. Machines may be immune to fatigue and some conscious biases that may beset human interviewers, yet there is a phenomenon called "garbage in, garbage out" (Theodorakopoulos, Theodoropoulou and Halkiopoulos, 2025). For example, an algorithm may be trained on data in which a specific demographic is unusually more successful than others. To the algorithm, this may mean that this particular demographic is more likely to be successful in the recruitment process.

The ethical implications also extend to stakeholder trust. According to Theodorakopoulos, Theodoropoulou and Halkiopoulos (2025), the perceived fairness of hiring impacts organizational reputation and the "psychological contract" between employer and employee. This is important because community relations are vital to manufacturing companies based on the impact they can have on the environment (Bristol-Alagbariya, Ayanponle and Ogedengbe, 2024). The goal of this study is to bridge the gap between what is known about technical capability and social responsibility by synthesizing the literature.

1.2 Problem Statement

Most organisations who have implemented hiring via AI have done so with the aim of achieving speed and cost-reduction and overlooking automated discrimination risk (Hosain et al., 2025). As more manufacturing firms start to integrate AI into hiring for workflow optimization, research has not been able to catch up to a point where there is an empirical synthesis of how algorithms affect diversity and workforce stability.

One central issue in this matter is the "transparency paradox", which means that the more complex and "accurate" an algorithm becomes, it might become less interpretable by humans as to its decision-making process (Omeiza et al., 2025). A lack of explainability can make accountability difficult. There is therefore a need to investigate if algorithmic hiring actually delivers on its promise of objectivity or if research can show if it is only perpetuating previously existing inequality.

1.3 Research Aim

To systematically review and synthesise academic literature on algorithmic hiring in the manufacturing industry and pay attention to recruitment efficiency, fairness, and ethical implications.

1.4 Research Questions

To achieve the aim of this study, the following research questions were addressed:

RQ1: How is algorithmic hiring conceptualized and applied in the manufacturing industry?

RQ2: What is the impact of algorithmic hiring on recruitment efficiency and decision-making in the manufacturing industry?

While RQ1 and RQ2 form the core focus, cross-cutting themes related to bias, fairness, and ethics emerged during the analysis and are discussed in the context of these two questions.

1.5 Significance of the Study

This study contributes to academic theory and industrial practice as insights from it adds to what is already known about Industry 4.0. As with many other sectors, manufacturing practitioners and researchers continue to use technology as a way of addressing industry specific labour challenges. This study systematically synthesises knowledge about algorithmic hiring and provides a roadmap for leaders in the manufacturing sector to find a balance between efficiency and fairness. Ethically and socially, the study is significant for its focus on Equality, Diversity, and Inclusion (EDI). Manufacturing firms face pressure from labour unions, regulatory bodies, and local communities on transparency and equity in recruitment. This research provides a framework for HR practitioners to audit their digital tools and ensure that the organisation does not marginalize protected groups or violate international labour standards.

As more jurisdictions start to identify the risks of AI for recruitment, it becomes important for manufacturing firms to ensure that their technical infrastructure complies with legal frameworks. This research then becomes a foundation on which policymakers and corporate governance officers can develop ethical guidelines to protect the rights of candidates.

1.6 Scope of the Study

In order to achieve a focused and rigorous systematic review, the scope of this research has been carefully defined. The study focuses on global academic literature and includes all industrial economies where algorithmic hiring has been implemented or conceptualized. Speaking of conceptualization, the study is built on the pillars of operational efficiency (e.g., time-to-hire, cost-reduction), algorithmic fairness (e.g., bias mitigation, demographic parity), and ethical governance (e.g., transparency, accountability). The study limits its analysis to the manufacturing industry. It pays particular attention to those subsectors where recruitment is high-volume at entry-level and positions are technical and specialized. The review will focus only on peer-reviewed journals and industry reports published between 2015 and 2026. This timeframe is instructive as it is a period of advancements in machine learning and natural language processing (NLP). This study will not collect primary data.

1.7 Structure of the Report

This report has six sections: Introduction, Literature Review, Methodology, Data Analysis and Findings, Discussion and Recommendations for HR practice and Further Research and Conclusion.

2. Literature Review

2.1 Introduction

Algorithmic hiring is one of the ways that many industries, including manufacturing, are using to manage large-scale recruitment. The promise of AH is better efficiency, objectivity, and workforce planning (Zhang and Yencha, 2022). There are risks however to AH adoption with regards to bias, transparency, and ethics. This review provides an overview of what is known about AH in manufacturing with a focus on efficiency, fairness, and ethics. The goal is to provide a critical evaluation of assumptions and identification of contradictions, as well as a picture of where the gaps in research are.

2.2 Manufacturing Labour Context and Adoption of Algorithmic Hiring

With the advent of Industry 4.0, manufacturing has also changed alongside other sectors. Production and labour systems are changing because of AI, automation, and data analytics (Erigbe, 2025). Manufacturing companies have features that separate them from other industries especially as regards the labour market, including high-volume hiring for entry-level roles and demand for technical and hybrid skills (Sullivan, 2022). The sector is currently facing labour shortages and high turnover (Express Employment, 2025), pressures that can incentivise employers to adopt automated hiring. Literature however does not fully support the notion that technology solves labour problems automatically. According to Kassa and Worku (2025), AI simply changes job roles and worker experiences rather than just improving efficiency. Manufacturing has also historically struggled with diversity; Farrukh et al. (2025) state that it is one of the industries where there is a lack of gender representation on the factory floor, raising the concern that new technologies may reproduce old inequalities rather than solve them.

When manufacturing firms adopt AH, they do so to automate tasks such as resume screening, candidate ranking, and skill matching (Umachandran, 2021). The drivers of AH adoption are therefore efficiency and cost reduction. According to Ebrahim and Rajab (2025), AI has been shown to reduce time-to-hire by up to 60%. However, the issue with these claims is that they are mainly from industry reports, and peer-reviewed studies have done little investigation specific to the manufacturing industry. Limon (2025) linked the efficacy of AH in manufacturing to operational needs such as workforce planning and productivity, while Pushpakumara and Ahsan (2025) argue that AH is more relevant to the service industry. In spite of these differences, there is a consensus by scholars that AH can be used as a general HR tool (Li et al., 2021). The limited research that focuses specifically on the manufacturing sector is a gap in the literature.

2.3 Efficiency Outcomes in Algorithmic Recruitment

Hosain et al. (2025) states that efficiency is the main justification for algorithmic hiring in any industry. In the manufacturing context, production needs often require recruitment to be fast and scalable. AI systems can reduce administrative workload of the human recruiters and speed up screening (Amaugo, 2024). Studies have shown that AI-driven recruitment can improve time-to-hire and cost-per-hire (Hukkeri and Pol, 2025), and predictive analytics can identify candidates likely to perform well or stay longer (Onyekachi, Okoro and Iqbal, 2025). The conflict however arises when these claims are measured in the context of bias. According to Marieke and Sach (2024), AI can remove human bias but it can also perpetuate the Gap Trap where algorithms penalize non-linear career paths, potentially masking the systemic exclusion of some demographics.

2.4 Algorithmic Bias and Workforce Inequality

One of the most studied areas in AI decision making is algorithmic bias (Marieke and Sach, 2024). This is the kind of bias that occurs when algorithms produce unfair outcomes based on flawed data on which it has been trained. For a sector that already contains marked inequalities as historical data has shown gender and occupational segregation (Farrukh et al., 2025), the algorithms trained on such data may reproduce outcomes no different from human recruiters. To compound this, research has shown that making hiring decisions by AI can perpetuate discrimination based on gender and race (Sony et al., 2025). Although Murikah, Nthenge and Musyoka (2024) argue that AI can reduce human bias by standardizing decision-making; Walther (2026) takes the opposite view as they argue that AI amplifies bias because it can scale faster and bigger. Research is also talking about new types of bias (Xu, Li and Jiang, 2025). For example, AI systems may favour candidates who use tools similar to those on which it is trained. This is called self-preference bias.

2.5 Transparency and the Black Box Problem

An important dimension to the AH discourse is a part of the broader AI decision making discussion and it has to do with the difficulty in interpreting AI decision making. These kinds of AI are referred to as black box systems (Asatiani et al., 2020). According to Asatiani et al. (2020), a lack of transparency in algorithmic decision making erodes accountability. Omeiza et al. (2025) highlight the transparency paradox where algorithms become more accurate but become less interpretable. This creates a conundrum for organisations because explaining hiring decisions becomes more difficult. This warranted the promotion of Explainable AI (XAI) as a solution. XAI methods can be accurate and reduce bias while improving transparency (Haque, Islam and Mikalef, 2023). The limitation is that much of XAI is still theoretical.

Another argument is that human oversight can make AI decision making more transparent (Holzinger, Zatloukal and Müller, 2024). However, Vicente and Matute (2023) disagree stating that humans tend to follow AI recommendations even when bias is present. The problem here is that if algorithms and humans reinforce the same biases, it is difficult to determine who is responsible for unfair decisions.

2.6 Ethical and Governance Challenges

The ethical questions concerning AH include fairness, accountability, and respect for individuals (Marieke and Sach, 2024). Firstly, applicants must believe that hiring processes are fair. According to Krishnan, Ahmad and Haron (2018), perceived fairness affects the reputation of an organisation and employee commitment. When such an organisation operates in local communities, as many manufacturers do, community relations can be affected if there is perceived unfairness in hiring.

The legal angle of AH can be seen in lawsuits against companies using AI for hiring. Litigants have sued for discrimination based on race, age, and disability (Cowley, 2026). Now, regulators are responding. The EU and some states in the US now require audits of automated hiring systems (Koulianos, 2024). Anti-discrimination laws are now being extended to AI, even though regulatory frameworks are still developing.

2.7 Synthesis of Debates and Research Gaps

There are three main debates in literature about the issue. The first one is the conflict between efficiency and fairness. Research is unconvincing about the trade-offs between speed or scalability and bias can be fully resolved. There is also a lack of consensus about the role of human oversight. Is the operative enough as a safeguard against bias or does the operative reinforce algorithmic decisions? The third issue is the gap between theory and practice. Many solutions like explainable AI have limited evidence of implementation.

This study has identified a dearth of research on AH hiring in manufacturing. The limited evidence on manufacturing makes it difficult to ascertain how AH affects workforce composition, productivity, and diversity in manufacturing over time. The other gap is the lack of integrated frameworks. Research mostly examines efficiency, fairness, and ethics separately. Few studies analyze how they interact in organisations.

2.8 Conclusion

This review shows the opportunities and risks in algorithmic hiring in manufacturing. Its prospects for efficiency are noted but there are concerns about bias, transparency, and ethics. The literature is fragmented and sector-specific analysis is lacking. There is a need for systematic research that integrates all dimensions of the discourse into the manufacturing context. This study addresses this gap by synthesising the evidence in literature.

3. Methodology

3.1 Introduction

This is the section where an explanation how the study was conducted is presented. The research systematically reviews and synthesizes literature on algorithmic hiring in the manufacturing industry. The focus is on three key areas: efficiency, fairness, and ethical implications. The reason for choosing systematic literature review (SLR) as the method was because the topic is still developing. SLR ensures that findings from previous studies can be identified, evaluated and combined transparently (Mengist, Soromessa and Legese, 2020).

3.2 Research Design: Systematic Literature Review Approach

SLR ensures that existing research can be collected and analyzed in a transparent and reproducible way (Mengist, Soromessa and Legese, 2020). A departure from traditional literature review, SLR follows a clear process and this helps to reduce bias and improve reliability. SLR is appropriate for this study because research on AH is still taking shape and little is known about how it specifically affects the manufacturing sector. A systematic approach helps to bring coherence to the limited but scattered insights out there.

This means that the study follows the principles of systematic reviews (transparency, consistency, and replicability). The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Page et al., 2021) and the Critical Appraisal Skills Program (CASP) checklists (Long, French and Brooks, 2020) are two frameworks that guided this study in identifying and selecting studies.

3.3 Search Strategy

The search started from academic databases so that the results can yield a broad spectrum of literature on the subject. The databases included Scopus, Web of Science, ScienceDirect, and Google Scholar. The search terms used for the search were based on the core concepts of the study and they included "algorithmic hiring," "AI recruitment," "machine learning hiring," "automated recruitment," and "algorithmic bias." The terms were combined with these keywords: "manufacturing" and "industrial sector." Boolean operators ("algorithmic hiring" AND "manufacturing", "AI recruitment" AND "bias", "automated hiring" OR "machine learning recruitment"). Filters were applied to limit studies only to those published between 2015 and 2026.

3.4 Inclusion and Exclusion Criteria

The following were the inclusion and exclusion criteria used to ensure that only relevant and high-quality studies were selected.

Inclusion criteria:

• Peer-reviewed journal articles

• Academic conference papers and industry reports

• Studies on algorithmic hiring or AI in recruitment

• Studies on manufacturing or HR in manufacturing

• Publications in English

• Studies published between 2015 and 2026

Exclusion criteria:

1. Non-academic sources

2. Articles outside the selected timeframe

3. Duplicate records.

3.5 Study Selection Process

The process of selecting the studies for analysis was based on the PRISMA framework (Page et al., 2021). This ensured that there was structure and transparency, but more importantly, it ensured that the selected studies were relevant and of a high-quality.

The process of selection was undertaken in four stages.

Stage 1: Identification

After the keywords and search strings had been inputted into the databases, studies meeting the criteria were identified and all the results were recorded for screening.

Stage 2: Screening

This was the stage where the duplicate articles were removed. After this, the remaining studies were screened from the content of their titles and abstracts. This revealed the studies that were not relevant to algorithmic hiring or recruitment. These were removed.

Stage 3: Eligibility

At this point, each of the selected articles were read so that they could be assessed against the inclusion and exclusion criteria.

Stage 4: Inclusion

This left only the studies that met all criteria and these made up the final sample for analysis.

3.6 Data Extraction Process

The data extraction process followed a structured approach. This structure ensured that relevant information could be collected from the selected studies. The structure also helped increase consistency and reduce bias.

These were the information recorded against each study selected:

• Author(s) and year of publication

• Study context (e.g., country, industry)

• Research method used

• Key findings especially as related to efficiency, fairness, and ethics

The data was placed in an extraction table and this ensured that the process of comparing findings across studies was intuitive. It also added a layer of consistency because all the studies were analyzed in the same way. The table also made synthesis easier and improved clarity.

3.7 Quality Assessment of Studies

The CASP checklist was used to assess the quality of the selected studies. The CASP approach is a recognized way to evaluate the credibility, relevance, and rigour of research (CASP, 2024). Based on CASP principles, the studies were assessed on clarity of research aims, appropriateness of methodology, validity of findings, and relevance to this review. The use of CASP logic strengthens the review because outcomes will indicate that the review is based on reliable evidence (CASP, 2024).

3.8 Data Analysis and Synthesis Method

The data was analyzed using thematic analysis, a method regarded as appropriate for identifying patterns across a range of studies (Ahmed et al., 2025). The analysis followed the six-step Braun and Clarke process of thematic analysis, moving from data familiarisation and initial coding through to theme development, review, definition, and final write-up (Ahmed et al., 2025).

3.9 Ethical Considerations

As the study does not involve primary data collection and there are also no human participants, it stands to reason that informed consent was not required. This is research that is based purely on published literature. However, there are still ethical responsibilities that need to be upheld. All sources have been properly cited to avoid plagiarism. The findings of the studies used have also been presented with accuracy, without misrepresentation and without selective reporting. The study also ensures that all interpretations were given on the basis of credible evidence. This helped to maintain academic integrity and to ensure that the conclusions drawn are fair and reliable.

3.10 Limitations of the Methodology

The limitations of the methodology deployed are as follows:

A study which depends entirely on existing literature may find itself limited by the quality and scope of the previous research it used (Shaheen et al., 2023). One important limitation specific to this study is the fact there is limited research that focuses specifically on algorithmic hiring in the manufacturing sector.

The second limitation is slightly related to the first in that the review may be affected by publication bias. Nair (2019) argues that studies with significant findings are more likely to be published than those with less than significant results. Another limitation is in the exclusion criteria which permitted only studies published in English. This may exclude relevant research published in other languages.

3.11 Section Summary

This section has explained the methodology used in the study. A systematic literature review approach was adopted to ensure transparency and structure. The section described the research design, search strategy, inclusion criteria, and study selection process. It also explained how data was extracted and analyzed. The next section presents the findings and analyses them.

4. Data Analysis and Findings

4.1 Introduction

In accordance with the outlined thematic analysis framework, qualitative data was extracted and synthesised from the selected six pieces of academic literature. Analysis was done based on a systematic review of patterns and coding that would help address the research questions posed.

4.2 Theme Identification

Bottom-up method of thematic analysis was used to define the main themes emerging from the selected literature. Firstly, findings of individual studies were categorized as subthemes, which were later incorporated into major categories or themes such as conceptualization, efficiency, and fairness. All themes defined from the literature are common for all six selected studies.

4.3 Features of Selected Studies

The reviewed articles address the topic of AI and algorithmic recruitment practices within different geographical areas. For example, three of the selected studies focus on AI adoption in China (Zheng et al., 2024; Chang & Cheng, 2025; Jiang et al., 2024). Two other articles cover Nigeria (Amaugo, 2024) and India (Umachandran, 2021), respectively. The last study (Limon, 2025) had a global outlook with data from Germany, Japan, and the USA.

The groups of participants and data sources were different for each study. For example, the sample size in one study comprised 423 manufacturing respondents in China (Zheng et al., 2024) and while another one was made up of 301 Nigerian employees (Amaugo, 2024). In another study, there were 1,072 applicants for employment positions (Chang & Cheng, 2025). One more study utilized micro-data of manufacturing firms listed in 2012-2019 (Jiang et al., 2024).

The studies also involved different types of technology for recruitment. They included MIS, ERP, and machine learning algorithms. Some models studied include data-matching algorithms and interview scheduling. Others are robotic process automation and automated machines. The information related to the studies is presented in the data extraction table (Appendix 1).

4.4 RQ1: Conceptualization and Application of Algorithmic Hiring

This theme table examines how the six studies have conceptualized and have applied algorithmic hiring.

Table 1: Conceptualization and Application of Algorithmic Hiring

Theme	Zheng et al., 2024	Limon, 2025	Amaugo, 2024	Umachandran, 2021	Chang & Cheng, 2025	Jiang et al., 2024
Conceptualization of AI/Algorithms	Objective Decision-Maker	Evidence-Based Analytics	Business Process Automation (BPA)	Sustainable Economy Driver	Pattern Recognition Model	Structural Labor Shifter.
Application in Candidate Screening	Blind Screening	MIS Integration.	Machine Learning	Pre-screening & selection	Algorithmic Filtering	High-Skilled Creation
Application in Candidate Matching	Data-Driven Interviews.	Skill-Based Allocation	Predictive Optimization	Success Factor Linkage	Data-Matching Analysis	Technical Role Deployment
Operational Efficacy Goal	Bias Mitigation	Productivity Gains	Market Competitiveness	Cost-Effective Sourcing	Administrative Speed.	Labor Efficiency

Manufacturing research conceptualizes the use of algorithms in recruiting as the transition from intuition to information. Zheng et al. (2024) perceive algorithms as objective evaluators using informational databases in lieu of subjective evaluation. According to Limon (2025), algorithmic hiring is a way of bringing human resource management into evidence-based decision making. The author, Amaugo (2024), conceptualizes algorithms as business process automation aimed at minimizing errors made by humans. For Umachandran (2021), algorithmic hiring is the driving force behind achieving sustainability through electronic recruitment models. In addition, Chang & Cheng (2025) describe algorithmic hiring as a pattern recognition model that detects signals of star performers. For Jiang et al. (2024), the concept of algorithmic hiring is that of structural labor shifters which change what kinds of jobs are available.

In terms of application, the concepts revolve around candidate screening and matching. For instance, according to Zheng et al. (2024), candidate screening is performed anonymously to conceal demographic information. Meanwhile, Limon (2025) focuses on how algorithms are combined with the management system for real-time scheduling. On the other hand, Amaugo (2024) explores how machine learning is implemented to uncover production deficiencies. Umachandran (2021) investigates the use of keywords to match resumes and job descriptions. The focus of Chang and Cheng (2025) is how organisations conduct automated filtering to evaluate initial criteria before starting testing. Jiang et al. (2024) focus on how companies use algorithms to increase the probability of attracting high-skilled technical specialists.

Meanwhile, matching applications make it possible to streamline the hiring process further. Thus, Zheng et al. (2024) argue for the use of chatbots and automated interviews to standardize questioning. According to Limon (2025), algorithms are employed to match employee qualifications with tasks. Amaugo (2024) highlights the application of predictive analytics for forecasting potential supply chain challenges. Umachandran (2021) correlates data about applicants with success factors using pattern recognition. Finally, Chang and Cheng (2025) use data-matching algorithms for selecting candidates based on their personality type and occupational preferences. Jiang et al. (2024) apply these tools to move employees to more efficient roles.

These six different conceptualizations are significant because if there is no agreed-upon definition, making comparisons becomes difficult and without an acceptable comparison, practitioners may struggle to identify the right framework for their context. This is a key finding in addressing RQ1.

4.5 RQ2: Impact on Recruitment Efficiency and Decision-Making

This theme table evaluates the impact of algorithmic hiring on recruitment efficiency and decision-making in manufacturing based on evidence in the studies.

Table 2: Impact on Recruitment Efficiency and Decision-Making

Theme	Zheng et al., 2024	Limon, 2025	Amaugo, 2024	Umachandran, 2021	Chang & Cheng, 2025	Jiang et al., 2024
Mitigation of Subjective Biases	Reshaping Objectivity	Standardized Indicators	Diminishing Human Mistakes	Enforcing Legitimate Selection	Minimizing Administrative Errors	Skill-Based Structural Shift
Speed and Temporal Efficiency	Accelerated Screening.	Improved Planning Accuracy	Reduced Changeover Times	Swift Talent Acquisition.	Minute-Level Selection.	Productivity Gains
Decision-Making Accuracy	Comprehensive Picture	Predictive Precision: Forecasting Problems	Success Factor Linkage	Pattern Recognition.	Mechanism Optimization.
Resource and Cost Efficiency	Direct Cost Savings	Annual Cost Reduction	Operational Cost Savings	Cost-Effective Mode	Expense Minimization.	Labour Cost Control
Decision-Making Constraints	Soft Skill Limitations	Technical Readiness Gap	Infrastructure Barriers	Consulting Assistance Needs.	Validation Risks.	Substitution Pressures.

AH plays a significant role in manufacturing because it eliminates subjectivity and enhances efficiency. Zheng et al. (2024) reveal that AI decreases seven categories of human biases (conformity, affinity, halo, recency, horn, contrast, and overall). According to Limon (2025), standardized measures eliminate subjective impact on HR decisions while Amaugo (2024) points out that human errors in the production process are reduced. Umachandran (2021) claims that logic-driven tools ensure that legally and socially acceptable selection is made. Chang and Cheng (2025) claim that algorithms eliminate the administrative delays that human selectors might be prone to. Jiang et al. (2024) point out that machines substitute error-prone human tasks with high-skill jobs.

Efficiency can be estimated in terms of speed and precision. Zheng et al. (2024) claim that results become faster due to automated resume evaluation. Limon (2025) reveals a 22% increase in the accuracy of labor planning. In Amaugo (2024), AI has been shown to minimize production changeover time from hours to minutes. Umachandran (2021) suggests that talent sourcing can be accelerated through the process of digital verification. Concurring, Chang and Cheng (2025) argue that AH ensures that the best talents can be determined in seconds. Decisions also become more accurate via predictive analytics. Zheng et al. (2024) states that algorithms can give selectors a better understanding of the candidate's capabilities. Limon (2025)'s study states AI attained an 82% level of accuracy in predicting absenteeism. Umachandran (2021) states that algorithms are able to connect candidates with success factors based on their positions. This quality was affirmed by Chang and Cheng (2025) who state that AI can use pattern recognition to identify indicators of top performance.

4.6 Cross-Cutting Theme: Bias, Fairness, Diversity, and Stakeholder Perceptions

This theme table highlights findings on bias, fairness, diversity, and stakeholder perceptions in algorithmic hiring in the studies. These cross-cutting themes support and qualify the findings under both RQ1 and RQ2.

Table 3: Bias, Fairness, Diversity, and Stakeholder Perceptions

Theme	Zheng et al., 2024	Limon, 2025	Amaugo, 2024	Umachandran, 2021	Chang & Cheng, 2025	Jiang et al., 2024
Mitigation of Cognitive Biases	Reduces seven types of bias (conformity, affinity, halo, recency, horn, contrast, and overall)	Standardizes performance indicators.	Diminishes human mistakes and subjective inefficiencies	Solves unconscious biases and enforces legitimate and legal selection.	Minimizes administrative errors and selection delays	Replaces error-prone manual tasks.
Diversity and Inclusion	Using blind screening to remove identifiers	Ensures human capital is managed without considering subjective traits.	-	Equal opportunity via the internet.	Workforce composition should reflect labour availability	-
Stakeholder Perceptions & Resistance	Resistance from candidates and a fear of job loss.	Success depends on user behavior and digital literacy.	Resistance to Change due to fears of job displacement.	Exceptional applicant experience	Staff may feel threatened or insecure	Fears of machines replacing humans.
Ethics and Data Privacy	Data privacy issues; AI replicating past discrimination	Ethical data use and employee agency	Data Privacy and Security Concerns	Data anonymity	Social desirability measures to filter out bias	Protecting people's livelihood
The Human-AI Synergy	Human judgment is indispensable	Socio-technical balance	AI adoption requires strategic planning and personnel training.	AI augments HR rather than replacing it	Managers reserve the right of appointment	The government must balance technology development and the protection of employment.

The studies prove that algorithms may help increase objectivity and help in eliminating cognitive biases. In particular, Zheng et al. (2024) address seven different types of bias (conformity, affinity, halo, recency, horn, contrast, and overall). For example, conformity bias is when an assessor lowers the initial positive assessment of a candidate due to the negative opinions shared by other members of the recruitment team. An example of affinity bias is when a company owner offers an executive position to his employee because they both graduated from the same university and have a common passion for rock climbing. Limon (2025) highlights that management information systems can provide additional transparency hiring in manufacturing companies. According to Amaugo (2024) and Umachandran (2021) respectively, artificial intelligence can eliminate subjective inefficiencies and address unconscious biases during the selection process. Chang & Cheng (2025) stress that when profiling is automated, delays can be avoided. However, while Jiang et al. (2024) state that machinery replaces manual work, they concentrate more on cognitive biases.

To foster diversity and inclusion, organizations implement numerous approaches in the implementation of AH. For instance, Zheng et al. (2024) identifies blind screening as one strategy for ensuring equality. Limon (2025) says that algorithms guarantee that human capital management is performed according to skills irrespective of subjective features. Finally, Umachandran (2021) offers up an argument that the use of internet technology gives people access to opportunities. Amaugo (2024) does not have concrete indicators of diversity but Chang and Cheng (2025) stressed that the structure of the workforce should be built without discrimination but based on labour availability.

It is in the area of stakeholder perceptions that AH implementation begins to witness barriers. Zheng et al. (2024) point out that manufacturing candidates and workers are often unwilling to engage amid fears of losing their jobs. Limon (2025) argues that it is the presence of digital skills that can determine the success of the implementation process. Amaugo (2024) recorded a 21.59% rate of rejection from employees due to the fear of being replaced. This is corroborated by Jiang et al. (2024) who reiterate long standing concerns of people about machines displacing them. Almost all the authors raised the ethical issues of data privacy and the duplication of previous discriminatory practices. Almost all scholars believe in the integration of human and artificial intelligence, where the former's involvement in decision-making is inevitable.

5. Discussion and Recommendations for HR Practice and Further Research

5.1 Addressing RQ1: Conceptualization and Application of Algorithmic Hiring in Manufacturing

The research looked at algorithmic recruitment in manufacturing and evaluated six peer-reviewed articles. It used efficiency, fairness, and ethical concerns as criteria for the review. The six studies had different views of algorithmic hiring. According to Zheng et al. (2024), algorithms are an objective decision maker. Limon (2025) referred to it as evidence-based analytics, while Amaugo (2024) views it within the perspective of business process automation. Umachandran (2021) associated it with sustainability while it was named as pattern recognition by Chang & Cheng (2025). Jiang et al. (2024) referred to algorithmic hiring as a structural labour shifter.

These differences are important because if the definition is not agreed on, making a comparison will not be an easy task. And without an acceptable comparison, it might be difficult for practitioners to identify the right framework to use in practice. Although the studies are focused on manufacturing and one would have expected that definitions would reflect the uniqueness. Yet the definitions appear to be from generalized HR models.

There are six different definitions of algorithmic hiring as it relates to the manufacturing sector. The six studies agree that AH can improve objectivity in approach to recruitment, automation, and use of data in decision-making processes. The two main ways that AH is used for hiring in manufacturing is for screening and matching. Blind screening through algorithms is used to eliminate demographic factors that may be used to identify applicants. Rather, the automation comes from keyword matching involving the alignment of candidate resume and the job position description. Predictive analytics predicts future performance of the candidates.

The conclusion that can be drawn here is that there is no single agreed-upon definition of algorithmic hiring specific to manufacturing. This poses real challenges for practitioners, because without a unified framework, it becomes difficult to benchmark tools, compare outcomes, or set consistent ethical standards across the sector.

5.2 Addressing RQ2: Impact on Recruitment Efficiency and Decision-Making

The study found that algorithmic hiring can improve efficiency and accuracy in the recruitment process. According to Limon (2025), labour planning precision improved by 22% because of AH. Amaugo (2024) explained how AI can decrease the time required for production line changeovers from hours to minutes. Zheng et al. (2024) indicated AI reduces seven forms of human prejudice. The automation of resume evaluation saves time for managers according to Limon (2025). The accuracy of labour planning is also increased by 22 percent (Limon, 2025) while time wasted for handing over to new shifts is considerably shortened (Amaugo, 2024). The talent acquisition process also achieves a faster turnaround time as predictive analytics helps to predict absenteeism by up to 82 percent (Limon, 2025). Algorithms also quickly link applicants to factors that predict success (Umachandran, 2021).

According to industry reports, AI decreases time-to-hire by 60% (Ebrahim and Rajab, 2025), while manufacturers anticipate that AI can improve efficiency by up to double digits (Sharma, 2026). However, studies on AH in manufacturing are limited as most of the evidence in literature comes from other industries. This is confirmed by Pushpakumara and Ahsan (2025) who emphasized that AH is more pertinent to services than manufacturing. This is a paradox because manufacturers' decision making on whether or not to adopt AH is based on data from other industries.

This review also supports the first debate identified in the literature review — the contention between efficiency and fairness. While Hosain et al. (2025) claim efficiency as the main rationale that organizations have for implementing AH, Dadheech et al. (2025) have pointed out that efficiency is overdependent on quantitative data. Our findings seem to validate both points of view. However, the studies agree that the benefits of AH to manufacturing recruitment is limited by the quality of the data. While efficiency gains are proven, they are heavily dependent on data quality and do not automatically guarantee fairness.

5.3 Cross-Cutting Themes: Bias, Ethics, and the Human-AI Synergy

Yet, efficiency does not always address the inequities in the process. The studies are unanimous that training algorithms on biased data can reinforce pre-existing disparities. This is the "garbage in, garbage out" issue. This is where the historical gender and racial imbalance existing in manufacturing that Farrukh et al. (2025) highlighted comes in. If data is not streamlined to remove bias, the inherent prejudices will be replicated by AI algorithms trained using this data.

Another perspective on the weakness side is that the speed of AI might result in systematic discrimination. Marieke and Sach (2024) coined the gap trap where algorithms start to discourage non-linear career trajectories. The issue with this is that non-linear trajectories are typical for women and underrepresented groups. The aggregate insight is that while efficiency might cover systematic discrimination, there is no proof within the literature that efficiency and equity can coexist and this trade-off must be resolved.

According to Zheng et al. (2024), AI mitigates seven types of biases. These are conformist bias, affinity bias, halo bias, recency bias, horn bias, contrast bias, and overall bias. However, there have been conflicting viewpoints on this issue by the other research reviewed. For instance, Walther (2026) argued that AI can increase bias because it can scale faster. On the other hand, Murikah et al. (2024) pointed out that AI can be used to mitigate bias because it makes decision making standard across situations. It must be said though that proof for this viewpoint cannot be said to be sufficient. This is because most of the studies which make the assertion used short-term data. The lack of longitudinal studies is a weakness for this theory. It may be easier for organizations to believe that AI does not demonstrate bias. However, it is important to remember that algorithms simply replicate what is fed into them. In fact, Sony et al. (2025) proved that human discrimination could be replicated by AI based on the data fed into it. This means that human input is still an important aspect of AH.

Another aspect of the AH paradox is the dichotomies of bias reduction and bias amplification. Zheng et al. (2024) showed that algorithms can reduce bias while on the other hand, Xu, Li, and Jiang (2025) discovered self-preference bias as one of the new biases from AH use. This situation confirms that more research focusing on manufacturing is needed to fully ascertain the direction of AH as regards bias.

The studies claimed that AH prevented bias but this could not be proved empirically. The perception of stakeholders to AH is generally negative as fear of job loss leads to rejection from 21.59 percent of candidates (Amaugo, 2024). Data privacy protection is another issue which was considered by almost all the authors. The six studies also recommended that there should be human-AI collaboration to guarantee fairness.

Other ethical issues include transparency, accountability, and data privacy. The adoption of Black-box models can lead to less accountability (Asatiani et al., 2020) while the paradox of transparency can be seen when making an algorithm more interpretable starts to reduce its accuracy (Omeiza et al., 2025). The second debate identified in the literature review relates to the need for human intervention. Human supervision allows greater transparency of AI actions according to Holzinger et al. (2024). However, Vicente and Matute (2023) argue that human operators tend to follow AI recommendations even if these recommendations are biased. This issue remains unresolved within the scope of our review. The gap between theory and practice is also demonstrated by how XAI technology was advocated by Haque et al. (2023) but few studies have demonstrated its implementation in manufacturing.

There are several limitations to this systematic review. First, there were only six articles devoted to manufacturing. It means the basis for analysis is too weak and publication bias might have affected the results based on the few articles available. The chances of getting an article published increase significantly if its conclusion was positive. Also, because each study focused on different countries (China, Nigeria, India, Germany, Japan, and the USA) and the countries do not have the same labour laws, it is impossible to draw universal conclusions from the insights.

6. Recommendations for HR Practice

This section provides recommendations on the use of AH by HR in manufacturing organizations based on the insights derived from these six studies. The first recommendation is that manufacturers must have a complete and comprehensive audit of all data prior to utilizing any kind of AI solutions. Previous hiring data may contain information that may be discriminatory to people of protected characteristics. According to Farrukh et al. (2025), discrimination is an ongoing problem in manufacturing. It is therefore important for manufacturing concerns to test their training dataset for any unfairness and determine if the algorithm discriminates against any population group. Regulatory institutions like the EU Commission now require this kind of testing (Koulianos, 2024).

It is essential that manufacturers avoid making the final decision on hiring solely based on AI. Employees must participate in the process even if AI is involved in the selection. Vicente and Matute (2023) stated that people tend to follow AI's recommendations without too much questioning. The logical solution therefore is for manufacturing recruiters to question algorithmic recommendation. This way, a hybrid approach which involves AI screening and ranking candidates is augmented by humans who make the final decision.

Manufacturing companies also need to implement Explainable Artificial Intelligence (XAI). Black-box algorithms should not be the standard for operations which have human or social impact. Candidates must be given reasons for their selection or rejection. There must be transparent documentation of how the algorithm makes its decision, and the candidates must be entitled to an explanation. Haque, Islam and Mikalef (2023) argue that XAI will improve trust when AI is used for decision making. Another benefit of transparency is it helps to protect the organization from lawsuits.

Employee resistance is one of the biggest barriers when adopting AI. Amaugo (2024) reports that fear of losing jobs led to rejection of technology. The onus rests on the firm to explain to their employees how the process works and what benefits they get from using AI. Digital literacy programs should also be launched because workers need to understand the AI they use for work and the AI whose work may affect them.

There is also a need for employers to audit their systems on a regular basis. That bias is absent presently is not a guarantee that it might not happen later due to changes in data. It is therefore expedient that firms compare the results achieved before and after implementing the AI system in terms of diversity, inclusion and representation. These audits must be performed by third party agents so that an objective conclusion is reached. If bias is detected, appropriate measures need to be implemented to rectify it.

7. Recommendations for Further Research

These gaps need to be addressed in future research endeavors. The need for more empirical studies on the impact of algorithms on workforce diversity and productivity cannot be understated. Future studies should also examine various sub-sectors in the manufacturing industry to determine their unique AI adoption dynamics and requirements. The current literature focuses on examining the efficiency, fairness, and ethics of algorithms used as separate variables. Future studies should go beyond this and attempt to develop comprehensive frameworks for analyzing how these variables are related. This would enable manufacturing organizations to identify the sweet spot which balances efficiency and fairness and adopt ethical practices when using AI. A framework for ethical AI adoption would be valuable in guiding companies through the journey.

There is also a need for studies to be conducted on AI-human collaboration for recruiting in manufacturing. Vicente and Matute (2023) raised concerns about the willingness of people to rely on unethical AI-based recommendations for hiring, however the insights are drawn from non-manufacturing firms. There is therefore a necessity for dedicated studies on manufacturing firms. In the same vein, research can explore various oversight mechanisms and their impacts on human-AI integration.

In addition to the current developments in recruitment in manufacturing, legislation such as the EU AI Act may have implications on recruitment practices. There is a need for research into the implications that these laws have on practice. Comparing recruitment practices in various countries would be useful as the biggest manufacturing firms are usually global organizations and they would need information on compliance with different legal frameworks.

8. Conclusion

The study conducted a systematic review of algorithmic hiring in manufacturing and specifically examined the impact of algorithmic hiring on the efficiency of hiring process, recruitment fairness, and recruitment ethics. Six peer reviewed research articles were used for the review. The selected papers originated from different countries and the studies they reported were conducted in various manufacturing settings.

According to the findings of the selected papers, algorithmic hiring is an emerging trend that is gaining increased attention of manufacturing firms. The use of AI technologies can help organisations achieve higher levels of efficiency during recruiting. Algorithmic hiring allows companies to accelerate application processing and decrease the burden on administrative staff. In addition, algorithmic hiring allows for faster recruitment of suitable employees and more accurate hiring. It does this through predictive analysis of the performance of potential employees.

However, there were several challenges identified in the course of the review. The first issue noted was algorithmic bias. Some studies suggested that AI can help to avoid human bias, while other scholars found that the opposite to be true: that algorithms could reflect the discriminatory attitudes that had previously existed in the organization. This means that working with a biased dataset would result in maintaining discrimination that had been in place before.

The problems of transparency and ethical recruiting also came to light in the literature review. Many AI applications tend to be black boxes and the issue with black boxes is that organizations and candidates cannot explain the rationale behind its decisions. Applicants also do not feel comfortable being selected by means of a machine.

The review also noted that human involvement cannot be eliminated in the process of hiring with algorithms. Human recruiters can ask questions about the suggestions made by artificial intelligence, evaluate the soft skills of applicants, and prevent any inherent bias during the process. An appropriate combination of human and machine work could help companies hire efficiently while being fair in their decisions.

This report also revealed some limitations in the literature on the topic. For example, although the six studies were focused on manufacturing, the search for articles revealed that there were insufficient studies focused on the industry. Most studies on algorithmic hiring investigated other sectors or were conducted as a way of finding a general model.

To sum it all up, algorithmic hiring can provide significant advantages to manufacturing companies. These advantages include efficiency, speed, and informed decision-making. Still, its ethical and legal implications cannot be overlooked. It is important to embrace the use of AI in a responsible manner and incorporate human involvement in the hiring process.

References

Ahmed, S.K., Mohammed, R.A., Nashwan, A.J., Ibrahim, R.H., Abdalla, A.Q., Ameen, B.M.M. and Khidhir, R.M. (2025). Using Thematic Analysis in Qualitative Research. Journal of Medicine, Surgery, and Public Health, [online] 6(6), p.100198. doi:https://doi.org/10.1016/j.glmedi.2025.100198.

Alaref, E. (2025). A Literature Review on Industry 4.0 and Its Technologies. Industry 4.0 - Transforming the Future Beyond Manufacturing [Working Title]. [online] doi:https://doi.org/10.5772/intechopen.1012837.

Amaugo, O. (2024). Impact of AI Adoption on Business Process Automation and Competitiveness in Manufacturing Industry in Nigeria. International Journal of Research and Innovation in Social Science, VIII(IIIS), pp.5321-5330. doi:https://doi.org/10.47772/ijriss.2024.803398s.

Asatiani, A., Malo, P., Nagbol, P.R., Penttinen, E., Rinta-Kahila, T. and Salovaara, A. (2020). Challenges of Explaining the Behavior of Black-Box AI Systems. MIS Quarterly Executive, 19(4), pp.259-278. doi:https://doi.org/10.17705/2msqe.00037.

Bot, R. and Dyaji, A. (2023). The Role of Manufacturing Sector in Nigeria's Economy. [online] 1(2), pp.194-210. Available at: https://www.researchgate.net/publication/366929337_The_Role_of_Manufacturing_Sector_in_Nigeria [Accessed 19 Mar. 2026].

Bristol-Alagbariya, B., Ayanponle, L.O. and Ogedengbe, D.E. (2024). Advanced Strategies for Managing Industrial and Community Relations in high-impact Environments. International Journal of Science and Technology Research Archive, 7(2), pp.076-083. doi:https://doi.org/10.53771/ijstra.2024.7.2.0069.

Cayrat, C. and Boxall, P. (2022). Exploring the Phenomenon of HR analytics: a Study of challenges, Risks and Impacts in 40 Large Companies. Journal of Organizational Effectiveness: People and Performance, 9(4). doi:https://doi.org/10.1108/joepp-08-2021-0238.

Chang, K. and Cheng, K.-T. (2025). The Emerging Technology in Hiring: Insights from Assembly Line Workers and Managers. Administrative Sciences, 15(12), p.463. doi:https://doi.org/10.3390/admsci15120463.

Cowley, S. (2026). Job Applicants Sue A.I. Recruitment Tool Company. [online] Nytimes.com. Available at: https://www.nytimes.com/2026/01/21/business/ai-hiring-tools-lawsuit-eightfold-fcra.html.

Dadheech, R., Karichalil, R., Sunanda, K. and Sindhu, N. (2025). AI in Recruitment Enhancing Efficiency or Replacing Human Judgement. [online] ResearchGate. Available at: https://www.researchgate.net/publication/392233257_AI_in_Recruitment_Enhancing_Efficiency_or_Replacing_Human_Judgement.

Ebrahim, S.S. and Rajab, H.A. (2025). The Future of HR: the Role of AI-Powered Recruitment in Shaping the Modern Workforce. OALib, 12(01), pp.1-22. doi:https://doi.org/10.4236/oalib.1112770.

Erigbe, S. (2025). Reshaping the Workforce: the Impact of AI, Automation, and Robotics on Labor Markets. American Scientific Research Journal for Engineering, Technology, and Sciences, [online] 120(1), pp.360-370.

European Commission (2025). AI Act. [online] European Commission. Available at: https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai [Accessed 19 Mar. 2026].

Express Employment (2025). Top 3 Challenges for Manufacturing in 2025: Skills gap, Turnover and AI Integration. [online] Manufacturing Dive. Available at: https://www.manufacturingdive.com/spons/top-3-challenges-for-manufacturing-in-2025-skills-gap-turnover-and-ai-int/744566/ [Accessed 5 Apr. 2026].

Farrukh, A., Mehddi, A., Sajjad, A. and Yawar, S. (2025). Barriers to Women's Participation and Contribution in the Process Manufacturing industry: Managerial Insights from Multiple Case Studies in an Emerging Economy. Journal of Management & Organization, pp.1-31. doi:https://doi.org/10.1017/jmo.2025.10045.

Haque, A.B., Islam, A.K.M.N. and Mikalef, P. (2023). Explainable Artificial Intelligence (XAI) from a User perspective: a Synthesis of Prior Literature and Problematizing Avenues for Future Research. Technological Forecasting and Social Change, 186, p.122120. doi:https://doi.org/10.1016/j.techfore.2022.122120.

Hassija, V., Chamola, V., Mahapatra, A., Singal, A., Goel, D., Huang, K., Scardapane, S., Spinelli, I., Mahmud, M. and Hussain, A. (2023). Interpreting Black-Box Models: a Review on Explainable Artificial Intelligence. Cognitive Computation, [online] 16(1), pp.45-74. doi:https://doi.org/10.1007/s12559-023-10179-8.

Holzinger, A., Zatloukal, K. and Müller, H. (2024). Is Human Oversight to AI Systems Still possible? New Biotechnology, [online] 85, pp.59-62. doi:https://doi.org/10.1016/j.nbt.2024.12.003.

Hosain, M.S., Amin, M.B., Debnath, G.C. and Rahaman, M.A. (2025). The Use of Artificial Intelligence (AI) in the Hiring process: Job Applicants' Perceptions of Procedural Justice. Computers in Human Behavior Reports, 19, p.100713. doi:https://doi.org/10.1016/j.chbr.2025.100713.

Hukkeri, P. and Pol, S. (2025). The Impact of Artificial Intelligence on Recruitment Efficiency and Onboarding Effectiveness. [online] doi:https://doi.org/10.9790/7439-0203012633.

Jiang, H., Ge, Y., Yang, C. and Yu, H. (2024). How Automated Machines Influence Employment in Manufacturing enterprises? PLOS ONE, 19(3), pp.e0299194-e0299194. doi:https://doi.org/10.1371/journal.pone.0299194.

Kassa, B.Y. and Worku, E.K. (2025). The Impact of Artificial Intelligence on Organizational Performance: The Mediating Role of Employee Productivity. Journal of Open Innovation: Technology, Market, and Complexity, 11(1), p.100474. doi:https://doi.org/10.1016/j.joitmc.2025.100474.

Koulianos, I. (2024). AI and HR Legislation: A Look at the EU and U.S. Developments in 2024. [online] VectorVMS. Available at: https://vectorvms.com/blog/total-talent-management/ai-hr-legislation-eu-us-2024/ [Accessed 5 Apr. 2026].

Krishnan, R., Ahmad, N.A.F. binti and Haron, H. (2018). The Effect of Employees Perceived Fairness of Performance Appraisal Systems on Employees Organizational Commitment. International Journal of Academic Research in Business and Social Sciences, 8(3). doi:https://doi.org/10.6007/ijarbss/v8-i3/3941.

Li, L., Lassiter, T., Oh, J. and Lee, M.K. (2021). Algorithmic Hiring in Practice. Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society, pp.166-176. doi:https://doi.org/10.1145/3461702.3462531.

Limon, G.Q. (2025). Workforce Analytics in manufacturing: a Review of MIS Tools for Labor planning, Absenteeism monitoring, and Productivity Optimization. 01(01), pp.90-114. doi:https://doi.org/10.63125/79geej86.

Long, H.A., French, D.P. and Brooks, J.M. (2020). Optimising the Value of the Critical Appraisal Skills Programme (CASP) Tool for Quality Appraisal in Qualitative Evidence Synthesis. Research Methods in Medicine & Health Sciences, 1(1), pp.31-42. doi:https://doi.org/10.1177/2632084320947559.

Marieke, A. and Sach, A. (2024). Michael Is Better than Mehmet: Exploring the Perils of Algorithmic Biases and Selective Adherence to Advice from Automated Decision Support Systems in Hiring. Frontiers in Psychology, 15. doi:https://doi.org/10.3389/fpsyg.2024.1416504.

Mengist, W., Soromessa, T. and Legese, G. (2020). Method for Conducting Systematic Literature Review and Meta-Analysis for Environmental Science Research. MethodsX, [online] 7(7), pp.1-11. doi:https://doi.org/10.1016/j.mex.2019.100777.

Murikah, W., Nthenge, J.K. and Musyoka, F.M. (2024). Bias and Ethics of AI Systems Applied in Auditing - a Systematic Review. Scientific African, [online] 25(e02281). doi:https://doi.org/10.1016/j.sciaf.2024.e02281.

Nair, A.S. (2019). Publication Bias - Importance of Studies with Negative results! Indian Journal of Anaesthesia, [online] 63(6), p.505. doi:https://doi.org/10.4103/ija.ija_142_19.

Omeiza, D., Bhattacharyya, R., Jirotka, M., Hawes, N. and Kunze, L. (2025). A Transparency paradox? Investigating the Impact of Explanation Specificity and Autonomous Vehicle Imperfect Detection Capabilities on Passengers. Transportation Research Part F Traffic Psychology and Behaviour, 109, pp.1275-1292. doi:https://doi.org/10.1016/j.trf.2025.01.015.

Onyekachi, A., Okoro, U. and Iqbal, Z. (2025). Predictive Analytics in Talent Management: Enhancing Recruitment and Retention through AI. [online] ResearchGate.

Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D., Shamseer, L., Tetzlaff, J.M., Akl, E.A., Brennan, S.E., Chou, R., Glanville, J., Grimshaw, J.M., Hrobjarttsson, A., Lalu, M.M., Li, T., Loder, E.W., Mayo-Wilson, E., McDonald, S. and McGuinness, L.A. (2021). The PRISMA 2020 statement: an Updated Guideline for Reporting Systematic Reviews. British Medical Journal, [online] 372(71). doi:https://doi.org/10.1136/bmj.n71.

Pushpakumara, T.D.C. and Ahsan, F.J. (2025). Artificial Intelligence Adoption in Service Industries: A Systematic Literature Review of Key Drives, Barriers, Challenges, and Strategies. International Journal of Innovative Science and Research Technology, [online] 10(3), pp.2240-2256. doi:https://doi.org/10.38124/ijisrt/25mar1490.

Raji, N., George, V., Iyer, R.S., Sharma, S. and Mahabub, B.S. (2024). Revolutionizing Recruitment the Role of Artificial Intelligence in Talent Acquisition. ShodhKosh Journal of Visual and Performing Arts, [online] 5(1). doi:https://doi.org/10.29121/shodhkosh.v5.i1.2024.2141.

Shaheen, N., Shaheen, A., Ramadan, A., Hefnawy, M.T., Ramadan, A., Ibrahim, I., Hassanein, M., Ashour, M.E. and Flouty, O. (2023). Appraising Systematic Reviews: a Comprehensive Guide to Ensuring Validity and Reliability. Frontiers in Research Metrics and Analytics, 8(8). doi:https://doi.org/10.3389/frma.2023.1268045.

Sharma, S. (2026). AI in Manufacturing 2026: Applications, ROI & Implementation. [online] Azilen Technologies. Available at: https://www.azilen.com/blog/ai-in-manufacturing/ [Accessed 5 Apr. 2026].

Sony, M.M.A.A.M., Amin, M.B., Ashraf, A., Islam, K.M.A., Debnath, N.C. and Debnath, G.C. (2025). Bias in AI-driven HRM systems: Investigating Discrimination Risks Embedded in AI Recruitment Tools and HR Analytics. Social Sciences & Humanities Open, [online] 12, p.102082. doi:https://doi.org/10.1016/j.ssaho.2025.102082.

Sulaiman-Oloko, A., Agu, C.O. and Ann, B. (2026). Staff Turnover and Performance of Manufacturing Firms in Nigeria. International Journal of Entrepreneurship and Forensic Management Research. [online] doi:https://doi.org/10.70382/nijefmr.v11i7.028.

Sullivan, D.W. (2022). Manufacturing Hiring Trends: Employers Seeking More Diverse Skills. International Journal of Management & Information Systems (IJMIS), 16(3), p.247. doi:https://doi.org/10.19030/ijmis.v16i3.7077.

Theodorakopoulos, L., Theodoropoulou, A. and Halkiopoulos, C. (2025). Cognitive Bias Mitigation in Executive Decision-Making: A Data-Driven Approach Integrating Big Data Analytics, AI, and Explainable Systems. Electronics, 14(19), p.3930. doi:https://doi.org/10.3390/electronics14193930.

Umachandran, K. (2021). Application of Artificial Intelligence for Recruitment in Manufacturing Industries. Journal of Emerging Technologies, 1(1), pp.11-18. doi:https://doi.org/10.57040/jet.v1i1.39.

Vicente, L.G. and Matute, H. (2023). Humans Inherit Artificial Intelligence Biases. Scientific Reports, 13(1). doi:https://doi.org/10.1038/s41598-023-42384-8.

Walther, C.C. (2026). When Human Bias Meets Machine-Scale Amplification. [online] Centre for International Governance Innovation. Available at: https://www.cigionline.org/articles/when-human-bias-meets-machine-scale-amplification/ [Accessed 5 Apr. 2026].

Xu, J., Li, G. and Jiang, J. (2025). AI Self-preferencing in Algorithmic Hiring: Empirical Evidence and Insights. doi:https://doi.org/10.2139/ssrn.5417394.

Zhang, L. and Yencha, C. (2022). Examining Perceptions Towards Hiring Algorithms. Technology in Society, 68, p.101848. doi:https://doi.org/10.1016/j.techsoc.2021.101848.

Zheng, F., Zhao, C., Usman, M. and Poulova, P. (2024). From Bias to Brilliance: the Impact of Artificial Intelligence Usage on Recruitment Biases in China. IEEE Transactions on Engineering Management, 71, pp.1-13. doi:https://doi.org/10.1109/tem.2024.3442618.