<p>This study presents a game-theoretic analysis of a bi-level supply chain comprising a single vendor and multiple buyers, based on an integrated stochastic inventory model. The vendor offers trade credit to buyers who order a quantity above a certain threshold, encouraging larger orders and boosting overall supply chain performance. The production process is imperfect, leading to a proportion of defective items that are identified and returned to the vendor after a screening process at the buyer’s end. Both the defective rate and system lead time are treated as stochastic variables. The market demand is modeled as price-sensitive, influencing buyers’ ordering decisions. A key feature of the model is the allowance of planned shortages with full backlogging, assuming customers are willing to wait for the next delivery. The research contrasts decentralized and centralized decision-making strategies within the supply chain using a game-theoretic framework, capturing strategic interactions between the vendor and buyers. To solve the models, a stochastic genetic algorithm (GA) is employed, supported by numerical examples. The results reveal that centralized coordination significantly enhances the expected total profit of the supply chain compared to decentralized strategies. Sensitivity analyses further illustrate the impact of critical parameters such as demand rate, customer price sensitivity, defective product percentage, backorder cost, vendor’s interest loss, and buyer’s interest charges. The findings underscore the importance of supply chain integration and game-theoretic coordination, showing that planned backorders, despite associated costs, can lead to increased profitability when managed effectively. Graphical illustrations support the theoretical insights, and practical implications along with avenues for future research are discussed to validate and extend the model’s applicability.</p>