Analytics-Driven Supply Chain Management
A Future Research Agenda Based on the XAIOR Framework: 2025

Facebook Whatsapp X-twitter Linkedin

Supply chain management (SCM) involves coordinating various functions across a multi-echelon system to manage the flow of goods, services, and processes that transform raw materials into finished products. This article explores the evolution of SCM toward analytics-driven approaches, framed through the three dimensions of the XAIOR framework: Performance Analytics (PA), Attributable Analytics (AA), and Responsible Analytics (RA). This framework offers a pathway for researchers to advance SCM methodologies, addressing challenges and opportunities in the field.

Performance Analytics (PA): Enhancing Efficiency and Effectiveness

Research in SCM has long focused on improving operational efficiency and decision accuracy. Bayesian decision theory, in particular, has emerged as a robust tool for integrating information into supply chain decisions.

Key Studies:

  • Iyer & Bergen (1997) applied Bayesian conjugate pair theory to model responsive supply chain operations, quantifying the impact of information updates on supply chain performance and achieving Pareto improvement. [1]
  • Aronis et al. (2004) leveraged Bayesian information updating for inventory management, optimizing inventory policies by updating parameters based on failure rates. 
  • Choi, Li, & Yan (2006) extended Bayesian models to scenarios with varying variances, emphasizing the importance of sophisticated models and well-chosen observation targets. [3]

Future Research Directions:

Researchers could explore adaptive Bayesian models tailored to dynamic supply chain environments, incorporating real-time data to further enhance predictive accuracy. Extending Bayesian applications to perishable goods or volatile markets could also provide valuable insights.

Did You Know?

Organizations can better understand how decisions are made by integrating explainable AI into supply chains. As businesses transition to this new approach, they would address ethical issues more effectively and attain greater operational results.

Supply chains will be better equipped to manage market risks and resource limits while maintaining ethics once AI is more understood. By using analytics responsibly companies can assess future social and environmental effects to create better and safer supply chains. The ability of AI technologies to make decisions and provide justification for them may determine the future.

Attributable Analytics (AA): Improving Interpretability and Actionability

As computational power and data availability grow, there is a rising need for interpretable and actionable models in supply chain risk analysis. Bayesian networks (BNs) have proven instrumental in this regard, offering probabilistic graphical models that provide clarity in risk relationships.

Key Studies:

  • Garvey, Carnovale, & Yeniyurt (2015) used Bayesian network analysis (BNA) to model risk propagation and dependencies in supply chains. [4]
  • Liu et al. (2021) adopted dynamic Bayesian networks (DBNAs) to address disruptions under worst-case probability scenarios, creating explainable optimization models. [5]
  • Sakib et al. (2021) applied BNA to forecast and manage challenges in oil and gas supply chains, identifying critical factors influencing performance. [6]

Challenges and Opportunities:

Many AI-driven tools in SCM are still opaque, functioning as “black boxes” that limit their usability for operations managers. The integration of explainable AI (XAI) techniques, such as SHAP (Shapley Additive Explanations), has shown promise in improving model transparency.

Future Research Directions:

Researchers could focus on integrating XAI methodologies with AI-driven SCM tools to bridge the gap between automation and interpretability. This could involve developing user-friendly dashboards that translate complex analytics into actionable insights for supply chain managers.

Enhance Your Study
with Professional Assistance

In need of help for your research and to meet your academic dreams? Don’t hesitate to contact us for help at any point of your research process: our services are individualized and comprehensive. Contact us today to learn how we can help you get the help you need to succeed in your academic pursuits!

Contact Us

Conclusion and Call to Action

The XAIOR framework offers a comprehensive lens through which researchers can address pressing challenges in SCM. By advancing methodologies in performance, attributable, and responsible analytics, future studies can contribute to the creation of scalable, interpretable, and ethical supply chains.

Examples of Future Applications:

  1. Industry-Specific Studies: Applying AA to healthcare or renewable energy supply chains to manage risks and optimize resource allocation.
  2. Ethical AI in Procurement: Extending RA to model fraud detection across diverse industries, ensuring fairness and accountability.
  3. Dynamic Decision Systems: Combining PA with real-time analytics for responsive supply chains in e-commerce and perishable goods sectors.

By exploring these dimensions, researchers can shape the future of SCM to be more efficient, transparent, and socially responsible, ensuring its relevance in an increasingly complex global landscape.