Generic Multi-Agent AI Framework for Weighted Dynamic Corridor Price Optimisation

Abstract

The objective of this analysis is to address the challenges encountered by pricing systems in managing real-time market dynamics. This study presents a fundamental theoretical framework focused on taxonomy and ontology for a domain-specific multi-agent artificial intelligence (AI), serving as an internal price advisor to optimize pricing strategies for products and services. The system is designed to function in conjunction with other corporate AI systems and an Enterprise Resource Planning System (ERP). The ERP serves as a high-quality data foundation, and several other internal and external sources can provide essential data with varying quality. Methods: The proposed AI model builds upon the Weighted Dynamic Corridor Price Optimization framework, which integrates cost-plus and value-based pricing methodologies within a non-linear price corridor bounded by lower and upper thresholds. In the context of supply chain integration, fully-cooperative pricing models can apply Nash equilibrium to enhance supply chain profitability, whilst semi-cooperative models mitigate information asymmetry through the principal-agent theory. The findings from the theoretical analysis of the generic industry- and product-agnostic multi-agent AI system suggest the system’s potential capacity for dynamically computing optimal prices. A generative AI module could facilitate real-time decision-making, enabling sales teams and similar stakeholders to simulate scenarios and refine pricing strategies. In conclusion, the proposed AI system should be capable of delivering adaptive, context-aware, and data-driven recommendations. Depending on its application, the AI system could become very complex, susceptible to errors, and require significant maintenance. Future research should focus on customizing the proposed AI system for specific industries and product categories and validating its applicability through empirical research.