2026-03-09
Valery Mfondoum, Mylène Noubi Tchatchoua, Homère Ngandam, Ibrahim Mfombie
A productive approach to integrating strategic Foresight and machine learning is the Generalized Strategic Foresight Model embedding Machine Learning Operations (MLOps) (GSF(M)²), a unified governance architecture that combines the interpretive depth of long-term scenario-based Foresight with the adaptivity of real-time machine learning pipelines. The model addresses structural deficiencies in existing decision-making systems, where Foresight methods generate anticipatory insights but lack operationalization mechanisms, while machine learning algorithms automate processes but ignore strategic and participatory context as well as socio-organizational specificity. A systematic literature review following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) methodology (16 publications in each block—Foresight and machine learning lifecycle) identified methodological gaps in both fields when compared against reference architectures. GSF(M)² synthesizes the strengths of both approaches by embedding Foresight logic into adaptive machine learning processes and integrating automated feedback loops into scenario planning. The result is a continuously learning ecosystem that recalibrates scenarios, model parameters, and strategic options in real time. The synthesis of anticipatory analytics, continuous horizon scanning, and data-driven prioritization enhances policymaking effectiveness and institutional agility under conditions of international and technological uncertainty. GSF(M)² represents the first dual-core framework for the co-evolution of strategic Foresight and adaptive algorithms within a unified reflexive governance architecture.
2025-12-17
Alyona Nefedova, Elizaveta Marinina, Ekaterina Antonova, Andoor Alexander Chamech
In the context of intensifying global technological competition and the imperative of achieving scientific and technological sovereignty for the Russian Federation, the exploration of novel formats for training highly qualified personnel assumes critical significance. One promising trajectory in this domain is the Industrial PhD — a framework designed to conduct research in the service of industrial needs, with active collaboration between universities and businesses, underpinned by state support. This model offers a means to bridge the institutional divide between academic research and the applied challenges of the real economy, thereby fostering knowledge transfer and enhancing the innovation capacities of enterprises. This article provides a comprehensive overview of international experiences with Industrial PhD programs, drawing on an analysis of over sixty programs across nineteen countries. We synthesize a range of organizational and financial models, and identify institutional preconditions essential for the sustainability of such programs: tripartite agreements among the university, the industrial partner, and the doctoral candidate; mechanisms of co-funding; systems of dual academic supervision; and flexible arrangements governing intellectual property. Particular attention is devoted to mapping the barriers and challenges encountered by program participants — such as divergent goal-setting between academic and corporate sectors, conflicting expectations, administrative burdens, and risks to academic autonomy. Furthermore, this article outlines potential pathways for adapting the Industrial PhD model in the Russian context. Key conditions for successful implementation are examined: the launch of pilot initiatives at leading technical universities; the development of a robust legal framework to support tripartite interaction; and the institutionalization of state support mechanisms.