[Paper Review] Future of Supply Chain: Challenges, Trends, and Prospects

Cristiana L. Lara, John M. Wassick|arXiv (Cornell University)|Jan 30, 2023

Business Process Modeling and AnalysisBusiness, Management and Accounting3 citations

TL;DR

This paper identifies shared challenges in global supply chains across e-commerce and process industries—particularly in scalability, multi-horizon decision integration, and human-computer coordination—proposing multi-agent systems with rigorous human decision modeling as a promising framework to unify strategic, tactical, and operational decision-making, thereby enhancing supply chain resilience and efficiency.

ABSTRACT

This paper discusses the broad challenges shared by e-commerce and the process industries operating global supply chains. Specifically, we discuss how process industries and e-commerce differ in many aspects but have similar challenges ahead of them in order to remain competitive, keep up with the always increasing requirements of the customers and stakeholders, and gain efficiency. While both industries have been early adopters of decision support tools based on machine intelligence, both share unresolved challenges related to scalability, integration of decision-making over different time horizons (e.g. strategic, tactical and execution-level decisions) and across internal business units, and orchestration of human and computer-based decision-makers. We discuss future trends and research opportunities in the area of supply chain, and suggest that the methods of multi-agent systems supported by rigorous treatment of human decision-making in combination with machine intelligence is a great contender to address these critical challenges.

Motivation & Objective

• To identify common challenges faced by e-commerce and process industries in managing complex, global supply chains.
• To analyze unresolved issues in scaling decision support systems and integrating decisions across strategic, tactical, and execution levels.
• To examine the coordination challenges between human and machine decision-makers in supply chain operations.
• To explore future research directions in supply chain management driven by advances in machine intelligence and human behavior modeling.
• To advocate for multi-agent systems with formal treatment of human decision-making as a key solution path for next-generation supply chains.

Proposed method

• Leverages insights from multi-agent systems (MAS) to model interactions among autonomous decision-making entities in supply chains.
• Integrates machine intelligence with formal models of human decision-making to improve coordination between human and algorithmic agents.
• Proposes a framework that supports decision-making across multiple time horizons—strategic, tactical, and operational—within a unified architecture.
• Emphasizes the need for scalable and interoperable systems that unify internal business units and external partners in decision processes.
• Uses a conceptual model to demonstrate how human cognitive constraints and machine capabilities can be jointly optimized in supply chain systems.
• Draws on existing decision support tools in e-commerce and process industries to inform the design of future hybrid human-machine systems.

Experimental results

Research questions

• RQ1 What are the core challenges shared by e-commerce and process industries in managing global supply chains?
• RQ2 How can decision-making be effectively integrated across strategic, tactical, and execution levels in supply chains?
• RQ3 What role does human decision-making play in hybrid human-machine supply chain systems, and how can it be formally modeled?
• RQ4 In what ways can multi-agent systems enhance scalability and coordination in complex supply chains?
• RQ5 What future research directions are most promising for advancing resilient and efficient supply chain systems?

Key findings

• E-commerce and process industries face parallel challenges in scaling decision support systems and integrating decisions across different time horizons.
• Current systems struggle with the orchestration of human and machine decision-makers, especially across organizational units and timeframes.
• Multi-agent systems that incorporate rigorous models of human behavior show strong potential for unifying multi-level decision-making in supply chains.
• The integration of machine intelligence with human decision-making is critical for achieving both efficiency and adaptability in future supply chains.
• Future research should focus on developing scalable, interoperable frameworks that treat human cognition as a first-class component in supply chain design.
• The proposed approach offers a viable path to addressing unresolved challenges in supply chain resilience, responsiveness, and sustainability.