Abstract :

Predicting B2B Invoice Delays Is Critical For Corporate Liquidity, Yet Traditional Models Often Fail To Capture The Volatility Of Customer Behavior. We Implemented A Strong Predictive Framework Has Been Developed Using Ensemble Learning Techniques And Survival Analysis To Forecast Invoice Payment Behavior. In Addition To Feature Engineering Techniques, This Study Has Introduced A Unique Customer Payment Risk Index And Temporal Decay Weighting To Emphasize Payment History Over Time To Account For Dynamic Changes In Customer Behavior. The Experimental Design Was Time-aware In Nature, And Regressor Techniques Like XGBoost, Random Forest, And LightGBM Were Used In Conjunction With Cox Proportional Hazards Analysis. The Experimental Results Reveal That The Proposed Regressor Technique Using LightGBM With The Introduced Risk Index Has Outperformed Other Techniques In Terms Of Root Mean Squared Error (RMSE) And Mean Absolute Error (MAE). The Study Has Also Used Paired T Tests And Wilcoxon Signed-rank Tests To Determine The Level Of Statistical Significance In Experimental Results. In Addition To That, This Study Has Also Used SHAP (SHapley Additive ExPlanations) Analysis To Determine The Level Of Interpretability In Financial Forecasting In Business-to-business Transactions. The Study Has Found That The Proposed Techniques Have Significantly Enhanced The Level Of Financial Forecasting In Business-to-business Transactions.

Published:

15-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

2328 - 2333

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite