Production Planning & Control

Artificial Intelligence–Enabled Production Systems: Empirical Evidence, Implementation Pipelines, and Industrial Applications

Artificial intelligence (AI) is rapidly moving from a standalone “analytics tool” to an operational capability that reshapes how production systems are planned, controlled, and improved in real time. In production contexts, AI can be understood as digital technologies performing tasks and decisions traditionally associated with human intelligence (Pomerol, 1997). Recent operations and production research highlights that AI’s operational value increasingly depends on data availability, interoperability, and sharing across heterogeneous systems, enabling AI to sense, learn, and act within production environments rather than merely supporting off-line analysis (Dubey et al., 2020; Papadopoulos et al., 2022; Spanaki et al., 2018).

As AI becomes embedded in industrial practice, it increasingly augments other production technologies by enabling learning-driven planning and control, data-driven quality and maintenance decisions, and closed-loop optimisation across cyber–physical production environments. Empirical studies already demonstrate promising task-level pathways—for example, deep reinforcement learning for shop-floor rescheduling based on visual planning artefacts (Palombarini and Martinez, 2021), deep learning approaches that extract value from unstructured maintenance logs to predict breakdowns (Usuga-Cadavid et al., 2021), and data-driven methods for multi-robot task allocation (Murugappan et al., 2021). However, what remains underdeveloped is system-level, empirically validated knowledge on how these point solutions are integrated with production technologies and operational data environments to form reliable end-to-end decision-and-execution workflows that generalise beyond individual tasks or single settings (Smyth et al., 2024).

In practice, AI will not augment production technologies as expected unless implementation conditions are addressed across the relationships between business needs, data readiness, and AI-enabled decision workflows. Before AI can be operationalised, organisations must clarify the business decision and process context, specify what data are needed for those decisions, and prepare usable production data through definition, collection, storage, integration, and cleaning—forming a foundational implementation pipeline that starts before AI model building (Mao et al., 2025). Evidence also indicates that beyond the “how-to-combine” question, we still lack robust empirical insight into the enabling and constraining factors—data readiness, organisational readiness, and human–AI interaction—that determine whether AI-enabled solutions can be adopted, implemented, and routinised as operational workflows in real production settings (Papadopoulos et al., 2022; Merhi and Harfouche. 2023; Agostinho et al., 2023). Therefore, this Special Issue focuses on empirical studies that examine the contexts of using AI to augment production technologies, while also addressing the upstream implementation pipeline and socio-technical conditions that determine whether such augmentation can be realised in practice.

Topic Areas

This Special Issue seeks to advance understanding of how AI is embedded in, and enhances, digital production technologies, with a clear emphasis on empirical evidence from real operational contexts (e.g., industrial datasets, system deployments, case studies, field experiments, digital trace data, and production/maintenance logs). Submissions that are purely conceptual, literature-review based, or modelling-driven without empirical validation are outside the scope of this Special Issue.

To ensure clarity for prospective authors, this Special Issue defines empirical research as studies that use real-world operational evidence, such as production and maintenance logs, IoT/sensor streams, machine/robot telemetry, MES/ERP traces, digital twin event logs (when linked to real operations), image/video inspection data, industrial case studies, field experiments, or multi-site archival datasets. Empirical studies may employ machine learning, optimisation, simulation, or digital twin experimentation, but only when the analysis is grounded in real operational data and/or validated in industrial settings with measurable outcomes (e.g., improvements in scheduling performance, quality, throughput, downtime, or decision accuracy).

Studies based solely on synthetic data, stylised simulations, benchmark datasets without production-operational grounding, or purely conceptual/literature-review contributions are outside the scope of this Special Issue. In addition to application-focused empirical papers, we explicitly encourage submissions that examine the full AI implementation pipeline in production systems, including business–AI alignment, where production decision problems are translated into AI use cases and data requirements; data readiness, where  governance/standards, monitoring, preparation/cleaning, and query/access enable reliable AI use; and deployment and workflow embedding, where AI outputs are integrated into human–AI decision workflows and operational routines.

While many emerging deployments take workflow-oriented or agent-like forms, an agentic framing is not required. Instead, we welcome any empirically grounded study that demonstrates how AI is operationalised across the production decision pipeline and linked to measurable production outcomes. To improve coherence, we organise the Special Issue topics along two dimensions, as can be seen from Table 1: (i) stages of AI implementation in production systems—from business–AI alignment (design), to data readiness (pre-implementation), to deployment and workflow embedding (implementation); and (ii) the production technologies and operational contexts in which AI is used to augment decision-making and execution (e.g., cyber–physical systems/digital twins, additive manufacturing, blockchain-enabled data infrastructures). This structure helps position contributions not only by ‘what technology’, but also by ‘where they sit in the implementation pipeline’.

Table 1. Special Issue topics by AI implementation stage and production technology

Table 1 provides a high-level map of the SI scope. Topic areas below offer illustrative (non-exhaustive) examples of suitable empirical submissions. Across these topics, we particularly welcome papers that examine how business needs are translated into AI design choices and production decision workflows, how data readiness conditions enable or constrain implementation, and how AI outputs are embedded into human–AI decision workflows in real production settings. This staged implementation logic is treated as a cross-cutting theme across the SI rather than as a separate topic area.

• AI-Augmented Production Planning and Control: studies that focus on predictive and prescriptive analytics empowered by AI for production planning, sequencing, scheduling, and dispatching; studies that explore reinforcement learning and autonomous decision-making for factory operations; studies that examine AI-supported capacity management, workflow optimisation, and constraint handling in real operational environments.
• Business–AI Alignment and AI Readiness Pipeline for Production Systems: empirical studies that examine how production decisions and business processes are translated into AI use cases and data requirements (i.e., what data are needed for what decisions); studies that investigate data governance/standards, data quality monitoring, data preparation, cleaning, integration, and query/access in production settings; studies that develop and empirically validate measures of AI readiness (e.g., data readiness, process readiness, organisational readiness) for AI-enabled production technologies.
• AI-enabled Quality and Process Control: studies that investigate machine learning and deep learning for real-time defect detection, predictive quality control, and anomaly identification; studies based on sensor data, image analytics, or high-frequency production data to evaluate automated inspection; studies that examine AI-driven adaptive process regulation and feedback mechanisms within active production systems, including studies that document the upstream data preparation and readiness conditions required to deploy such planning and control solutions.
• Agentic and Workflow-Oriented Deployment in Production: empirical studies that investigate how AI-enabled workflows connect data preparation, querying, and analysis to production decisions and execution (e.g., natural-language interfaces to databases/BI, AI-assisted reporting and decision support, and early forms of production “agents” that operationalise decision guidance); studies that evaluate operational performance impacts and human–AI interaction within such workflows.
• AI-Enabled Cyber–Physical Production Systems and Digital Twins: studies that develop and empirically assess intelligent digital twins for real-time optimisation, forecasting, and scenario evaluation; studies that investigate hybrid physics–AI approaches embedded within cyber–physical production environments; studies that provide evidence on continuous learning, data flow, and synchronisation between physical systems and their digital representations; studies that analyse AI-enabled control using real sensor/machine/robot data, including edge AI for decentralised and real-time decision-making; and studies that examine human–robot collaboration enhanced by AI-based perception, adaptive learning, and decision support tools.
• AI and Additive Manufacturing: studies that apply AI to generative design, topology optimisation, and process parameter tuning in AM contexts; studies that empirically evaluate AI-enabled defect prediction, geometry correction, and build-path optimisation; studies that explore AI-enabled coordination and management across distributed AM networks; studies that investigate how end users leverage AI in design to enhance additive manufacturing production and enable product customisation.
• AI and Blockchain for Production Systems: studies that provide empirical evidence on AI-enhanced traceability, verification, and secure data sharing in production networks; studies that explore the integration of AI with blockchain to support auditability, transparency, and decentralised coordination; studies that examine the operational use of smart contracts in automating production activities.
• Human–AI Interaction and Organisational Transformation: studies that investigate human-in-the-loop AI and operator–AI collaboration in production environments; studies that examine how operators, engineers, and managers guide AI systems, interpret AI outputs, evaluate AI-generated recommendations, and decide whether to accept, adapt, or override AI-supported decisions; studies that document organisational readiness, capability development, skill transformation, and adoption challenges associated with AI-enabled production systems; studies that examine explainability, trust, transparency, and governance mechanisms influencing AI deployment on the factory floor.
• Empirical Innovations in Data, Measurement, and Evaluation: studies that utilise field experiments, industry datasets, IoT streams, digital twin logs, video analytics, or multimodal data to evaluate the performance and impact of AI-enabled production technologies; studies that demonstrate measurable improvements in productivity, quality, flexibility, or decision accuracy resulting from AI integration.
• Extended Production Contexts (Production-Adjacent Physical Operations): studies that examine AI embedded in production-adjacent execution systems where planning/control decisions are central, such as warehousing and fulfilment operations; studies that empirically evaluate AI-enabled AMRs for picking, transport, and sorting; studies that assess real-time slotting/routing, vision systems for object recognition and damage detection, and workforce–robot coordination, with explicit implications for operational control and performance.

References:

Agostinho, C., Dikopoulou, Z., Lavasa, E. et al. (2023) ‘Explainability as the key ingredient for AI adoption in Industry 5.0 settings’, Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2023.1264372

Dubey, R., Gunasekaran, A., Childe, S.J., Bryde, D.J., Giannakis, M., Foropon, C., Roubaud, D. and Hazen, B.T. (2020) ‘Big data analytics and artificial intelligence pathway to operational performance under the effects of entrepreneurial orientation and environmental dynamism: A study of manufacturing organisations’, International Journal of Production Economics, 226, 107599. https://doi.org/10.1016/j.ijpe.2019.107599

Mao, Y., Holmström, J., & Cheng, Y. (2025). Applying machine learning to predict production capacity for engineer-to-order products: Learning from wind turbine industry. Technological Forecasting and Social Change, 219, 124295. https://doi.org/10.1016/j.techfore.2025.124295

Merhi, M.I. and Harfouche, A. (2023) ‘Enablers of artificial intelligence adoption and implementation in production systems’, International Journal of Production Research, pp. 1–15. https://doi.org/10.1080/00207543.2023.2167014

Murugappan, E., Subramanian, N., Rahman, S., Goh, M. and Chan, H.K. (2021) ‘Performance analysis of clustering methods for balanced multi-robot task allocations’, International Journal of Production Research. https://doi.org/10.1080/00207543.2021.1955994

Palombarini, J.A. and Martinez, C.E. (2021) ‘End-to-end online rescheduling from Gantt chart images using deep reinforcement learning’, International Journal of Production Research. https://doi.org/10.1080/00207543.2021.2002963

Papadopoulos, T., Sivarajah, U., Spanaki, K., Despoudi, S. and Gunasekaran, A. (2022) ‘Editorial: Artificial Intelligence (AI) and data sharing in manufacturing, production and operations management research’, International Journal of Production Research, 60(14), pp. 4361–4364. https://doi.org/10.1080/00207543.2021.2010979

Pomerol, J.C. (1997) ‘Artificial intelligence and human decision making’, European Journal of Operational Research, 99(1), pp. 3–25. https://doi.org/10.1016/S0377-2217(96)00378-5

Smyth, C., Dennehy, D., Fosso Wamba, S., Scott, M. and Harfouche, A. (2024) ‘Artificial intelligence and prescriptive analytics for supply chain resilience: a systematic literature review and research agenda’, International Journal of Production Research, 62(23), pp. 8537–8561. https://doi.org/10.1080/00207543.2024.2341415

Spanaki, K., Gurguc, Z., Adams, R. and Mulligan, C. (2018) ‘Data supply chain (DSC): Research synthesis and future directions’, International Journal of Production Research, 56(13), pp. 4447–4466. https://doi.org/10.1080/00207543.2017.1399222

Usuga-Cadavid, J.P., Lamouri, S., Grabot, B. and Fortin, A. (2021) ‘Using deep learning to value free-form text data for predictive maintenance’, International Journal of Production Research. https://doi.org/10.1080/00207543.2021.1951868