Judging Category
Basic or Experimental Research
Student Rank
Graduate
College
Business
Faculty Sponsor
Dr. John Mello jmello@astate.edu
Description
This study evaluates the effectiveness of artificial intelligence (AI) in reducing the bullwhip effect within a simulated supply chain using the Beer Game model. Three decision-making approaches were compared: traditional human-only ordering with limited information, AI-assisted ordering using ChatGPT, and collaborative decision-making with full information sharing across supply chain tiers. Supply chain performance was evaluated over four simulation rounds using key operational metrics, including order variability, inventory levels, backorders, and customer service performance. Results indicate that traditional human decision-making led to higher order variability and stronger fluctuations due to reactive decisions and limited information visibility. AI-assisted ordering improved decision consistency and partially stabilized ordering patterns. Collaborative decision-making produced the most stable system performance, with lower inventory variability, reduced backorders, and higher service levels. These findings demonstrate that full information sharing combined with coordinated decision-making outperforms both human-only and AI-assisted approaches in improving supply chain stability, reducing the bullwhip effect, and enhancing operational efficiency.
Disciplines
Operations and Supply Chain Management
License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Recommended Citation
Vo, Ngan; Cao, Khang; and Lama, Nikita, "Evaluating the Effectiveness of AI in Minimizing the Bullwhip Effect in Supply Chains" (2026). Create@State. 10.
https://arch.astate.edu/evn-createstate/2026/posters/10
Included in
Evaluating the Effectiveness of AI in Minimizing the Bullwhip Effect in Supply Chains
This study evaluates the effectiveness of artificial intelligence (AI) in reducing the bullwhip effect within a simulated supply chain using the Beer Game model. Three decision-making approaches were compared: traditional human-only ordering with limited information, AI-assisted ordering using ChatGPT, and collaborative decision-making with full information sharing across supply chain tiers. Supply chain performance was evaluated over four simulation rounds using key operational metrics, including order variability, inventory levels, backorders, and customer service performance. Results indicate that traditional human decision-making led to higher order variability and stronger fluctuations due to reactive decisions and limited information visibility. AI-assisted ordering improved decision consistency and partially stabilized ordering patterns. Collaborative decision-making produced the most stable system performance, with lower inventory variability, reduced backorders, and higher service levels. These findings demonstrate that full information sharing combined with coordinated decision-making outperforms both human-only and AI-assisted approaches in improving supply chain stability, reducing the bullwhip effect, and enhancing operational efficiency.
