Scholarlink

<p>India’s Unified Payments Interface (UPI) Person-to-Merchant (P2M) ecosystem has experienced exponential growth. However, traditional pricing models fail to account for transaction reliability, consumer sentiment, and fraud risk. This limits the ability of platforms to maximize revenue while ensuring stable and trustworthy transactions. Optimized Dynamic Pricing Model (ODPM), an AI-driven framework integrating Random Forest regression for demand forecasting, Valence Aware Dictionary and sEntiment Reasoner (VADER) sentiment analysis for real-time behavioral insights, and k-means clustering for merchant segmentation. ODPM dynamically computes adaptive prices based on reliability (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({R}_{t}\)</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi>R</mi> <mi>t</mi> </msub> </math></EquationSource> </InlineEquation>), sentiment (<InlineEquation ID="IEq2"> <EquationSource Format="TEX">\({S}_{t}\)</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi>S</mi> <mi>t</mi> </msub> </math></EquationSource> </InlineEquation>), and fraud risk (<InlineEquation ID="IEq3"> <EquationSource Format="TEX">\({F}_{t}\)</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi>F</mi> <mi>t</mi> </msub> </math></EquationSource> </InlineEquation>) to maximize revenue across diverse merchant groups. UPI P2M data (2021–2024) demonstrates ODPM achieves a 12% revenue uplift at an optimal dynamic price factor of 1.1, with total revenue of USD 6,371.32&#xa0;M, root mean square error (RMSE) 4.2%, and R<sup>2</sup> = 0.68 (<i>p</i> &lt; 0.01). Fraud-related revenue volatility reduced by 15%, and the model outperforms static and ARIMA-based pricing approaches. ODPM provides a practical, interpretable, and prescriptive framework for real-time, high-frequency pricing. It advances revenue management theory by integrating behavioral and risk factors and highlights urban–rural heterogeneity in transaction dynamics, offering actionable insights for fintech platforms.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimizing revenue and pricing for UPI transactions using AI-based dynamic pricing models

  • 期刊论文

  • 发表日期：

    2025年12月18日

查看全文
  • V. Prema Kumari,
  • S. Antony Raj

摘要

India’s Unified Payments Interface (UPI) Person-to-Merchant (P2M) ecosystem has experienced exponential growth. However, traditional pricing models fail to account for transaction reliability, consumer sentiment, and fraud risk. This limits the ability of platforms to maximize revenue while ensuring stable and trustworthy transactions. Optimized Dynamic Pricing Model (ODPM), an AI-driven framework integrating Random Forest regression for demand forecasting, Valence Aware Dictionary and sEntiment Reasoner (VADER) sentiment analysis for real-time behavioral insights, and k-means clustering for merchant segmentation. ODPM dynamically computes adaptive prices based on reliability ( \({R}_{t}\) R t ), sentiment ( \({S}_{t}\) S t ), and fraud risk ( \({F}_{t}\) F t ) to maximize revenue across diverse merchant groups. UPI P2M data (2021–2024) demonstrates ODPM achieves a 12% revenue uplift at an optimal dynamic price factor of 1.1, with total revenue of USD 6,371.32 M, root mean square error (RMSE) 4.2%, and R2 = 0.68 (p < 0.01). Fraud-related revenue volatility reduced by 15%, and the model outperforms static and ARIMA-based pricing approaches. ODPM provides a practical, interpretable, and prescriptive framework for real-time, high-frequency pricing. It advances revenue management theory by integrating behavioral and risk factors and highlights urban–rural heterogeneity in transaction dynamics, offering actionable insights for fintech platforms.