Machine Learning–Driven Forecasting Pipelines for Financial Volatility Detection in Integrated Enterprise ERP Environments

Abstract

This study examined how machine learning driven forecasting pipelines improve financial volatility detection in integrated enterprise ERP environments, addressing the problem that many organizations possess large volumes of financial and operational data but still struggle to identify early signs of instability in a timely and reliable manner. The purpose of the research was to assess whether ERP data integration, machine learning forecasting pipeline capability, real time analytics capability, forecasting pipeline automation, and machine learning model performance significantly strengthen financial volatility detection and related managerial outcomes in enterprise settings. A quantitative, cross sectional, case-based design was adopted using survey data from 214 valid respondents drawn from cloud enabled and enterprise ERP cases across manufacturing, banking and financial services, retail and trade, logistics and supply chain, and technology and service organizations. Data were collected through a five-point Likert scale questionnaire and analyzed using descriptive statistics, Cronbach’s alpha, Pearson correlation, and multiple regression in SPSS. The findings showed high levels of ERP data integration (M = 4.18, SD = 0.61), machine learning forecasting pipeline capability (M = 4.11, SD = 0.66), real time analytics capability (M = 4.07, SD = 0.70), and financial volatility detection (M = 4.15, SD = 0.63). Correlation results indicated that financial volatility detection was positively associated with ERP data integration (r = 0.62, p < .001), machine learning forecasting pipeline capability (r = 0.71, p < .001), real time analytics capability (r = 0.68, p < .001), forecasting pipeline automation (r = 0.57, p < .001), and machine learning model performance (r = 0.73, p < .001). Regression analysis confirmed that all predictors significantly influenced financial volatility detection, with machine learning model performance emerging as the strongest predictor (β = 0.29, p < .001), and the overall model explaining 52.9% of the variance (R² = 0.529, F = 46.82, p < .001). The study implies that enterprises can strengthen financial monitoring, risk awareness, and decision support by embedding high performing machine learning forecasting systems within integrated ERP based information environments.