Intelligent Cash-Flow Stress Detection for Small and Medium-Sized Enterprises Using Financial Ratios and Payment-Delay Signals
Keywords:
Cash-Flow Stress, Small and Medium-Sized Enterprises, Financial Ratios, Payment Delays, Logistic Regression, Decision Tree, Interpretable Machine LearningAbstract
Small and medium-sized enterprises (SMEs) can remain profitable while becoming unable to meet near-term obligations because receivables arrive late, short-term liabilities accumulate, or operating cash flow weakens. This study develops a lightweight early-warning approach that combines conventional financial ratios with operational payment-delay signals. A reproducible Python process generated 3,000 clearly synthetic SME-period observations and probabilistically sampled a cash-flow-stress label from an explicit logistic data-generating mechanism. Logistic regression and a constrained decision tree were evaluated under stratified 70/15/15 training, validation, and untouched-test splits. Four required configurations compared ratio-only features with combined financial, payment, and control variables. On the test set, combined logistic regression achieved 0.780 accuracy, 0.790 precision, 0.663 recall, 0.868 specificity, 0.721 F1, 0.856 ROC-AUC, and 0.547 Matthews correlation coefficient. Its F1 and ROC-AUC exceeded the ratio-only logistic model, although bootstrap intervals overlapped. The combined decision tree did not improve over its ratio-only counterpart, demonstrating that additional variables do not automatically generalize. Operating cash-flow ratio, leverage, overdue-invoice percentage, customer and supplier delays, and interest coverage were prominent interpretable signals. The findings support human-reviewed screening rather than automated lending or rejection. Because the data and target are synthetic, real SME validation, temporal testing, calibration, fairness analysis, and cost-sensitive threshold governance remain necessary.
References
A. Malakauskas and A. Lakštutienė, “Financial Distress Prediction for Small and Medium Enterprises Using Machine Learning Techniques,” Engineering Economics, vol. 32, no. 1, pp. 4–14, 2021, doi: 10.5755/j01.ee.32.1.27382.
Y. Tong and Z. Serrasqueiro, “Predictions of Failure and Financial Distress: A Study on Portuguese High and Medium-High Technology Small and Mid-Sized Enterprises,” Journal of International Studies, vol. 14, no. 2, pp. 9–25, 2021, doi: 10.14254/2071-8330.2021/14-2/1.
Y. Zizi, M. Oudgou, and A. El Moudden, “Determinants and Predictors of SMEs’ Financial Failure: A Logistic Regression Approach,” Risks, vol. 8, no. 4, art. 107, 2020, doi: 10.3390/risks8040107.
V. Lefebvre, “Trade credit, payment duration, and SMEs’ growth in the European Union,” International Entrepreneurship and Management Journal, vol. 19, no. 3, pp. 1313–1340, 2023, doi: 10.1007/s11365-023-00871-4.
O. Kaya, “The impact of late payments on SMEs’ access to finance: Evidence from credit rationing and loan terms,” Economic Modelling, vol. 141, art. 106896, 2024, doi: 10.1016/j.econmod.2024.106896.
P. Murro and V. Peruzzi, “Relationship lending and the use of trade credit: the role of relational capital and private information,” Small Business Economics, vol. 59, no. 1, pp. 327–360, 2022, doi: 10.1007/s11187-021-00537-x.
Y. Zizi, A. Jamali-Alaoui, B. El Goumi, M. Oudgou, and A. El Moudden, “An Optimal Model of Financial Distress Prediction: A Comparative Study between Neural Networks and Logistic Regression,” Risks, vol. 9, no. 11, art. 200, 2021, doi: 10.3390/risks9110200.
Y. Aker and A. Karavardar, “Using Machine Learning Methods in Financial Distress Prediction: Sample of Small and Medium Sized Enterprises Operating in Turkey,” Ege Academic Review, vol. 23, no. 2, pp. 145–162, 2023, doi: 10.21121/eab.1027084.
A. Gepp and K. Kumar, “Predicting Financial Distress: A Comparison of Survival Analysis and Decision Tree Techniques,” Procedia Computer Science, vol. 54, pp. 396–404, 2015, doi: 10.1016/j.procs.2015.06.046.
M. Ďurica, J. Frnda, and L. Svabova, “Artificial neural network and decision tree-based modelling of non-prosperity of companies,” Equilibrium. Quarterly Journal of Economics and Economic Policy, vol. 18, no. 4, pp. 1105–1131, 2023, doi: 10.24136/eq.2023.035.
I. Podhorská, J. Vrbka, G. Lăzăroiu, and M. Kováčová, “Innovations in Financial Management: Recursive Prediction Model Based on Decision Trees,” Marketing and Management of Innovations, no. 3, pp. 276–292, 2020, doi: 10.21272/mmi.2020.3-20.
B. Prusak, “Review of Research into Enterprise Bankruptcy Prediction in Selected Central and Eastern European Countries,” International Journal of Financial Studies, vol. 6, no. 3, art. 60, 2018, doi: 10.3390/ijfs6030060.
D. V. Carvalho, E. M. Pereira, and J. S. Cardoso, “Machine Learning Interpretability: A Survey on Methods and Metrics,” Electronics, vol. 8, no. 8, art. 832, 2019, doi: 10.3390/electronics8080832.
A. Barredo Arrieta et al., “Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI,” Information Fusion, vol. 58, pp. 82–115, 2020, doi: 10.1016/j.inffus.2019.12.012.
M. T. Ribeiro, S. Singh, and C. Guestrin, “Why Should I Trust You?: Explaining the Predictions of Any Classifier,” Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, pp. 1135–1144, 2016, doi: 10.1145/2939672.2939778.
F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011.
D. Chicco and G. Jurman, “The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation,” BMC Genomics, vol. 21, art. 6, 2020, doi: 10.1186/s12864-019-6413-7.
T. Saito and M. Rehmsmeier, “The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets,” PLOS ONE, vol. 10, no. 3, art. e0118432, 2015, doi: 10.1371/journal.pone.0118432.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 The Author(s)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
