Company failure prediction in the construction industry

• Published in: Expert systems with applications Vol. 40; no. 16; pp. 6253 - 6257
• Main Authors: Horta, I.M., Camanho, A.S.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ltd 15.11.2013; Elsevier
• Subjects: Applied sciences; Company failure; Computer science; control theory; systems; Computer systems performance. Reliability; Construction industry; Data processing. List processing. Character string processing; Exact sciences and technology; Expert systems; Failure; Firm modelling; Logistic regression; Logistics; Mathematical models; Mathematics; Memory organisation. Data processing; Operational research and scientific management; Operational research. Management science; Oversampling; Probability and statistics; Regression; Sampling theory, sample surveys; Sciences and techniques of general use; Software; Statistics; Support vector machine; Support vector machines; Logistic regression; Support vector machine; Company failure; Construction industry; Data analysis; Probabilistic approach; Random sampling; Replacement; Oversampling; Data mining; Failures; Modeling; Subsampling; Business model; Specificity; Portuguese; Vector support machine; Sampling methods; Random variable; Firm
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2013.05.045

•The paper proposes an innovative model to predict construction company failure.•The model is based on the SVM technique and its performance is compared with Logistic regression.•The model includes financial and strategic variables as predictors.•Two different sampling techniques to handle imbalanced datasets are tested.•The performance of the SVM model, evaluated using the AUC measure, is very good (over 90%). This paper proposes a new model to predict company failure in the construction industry. The model includes three major innovative aspects. The use of strategic variables reflecting the key specificities of construction companies, which are critical to explain company failure. The use of data mining techniques, i.e. support vector machine to predict company failure. The use of two different sampling methods (random undersampling and random oversampling with replacement) to balance class distributions. The model proposed was empirically tested using all Portuguese contractors that operated in 2009. It is concluded that support vector machine, with random oversampling and including strategic variables, is a very robust tool to predict company failure in the context of the construction industry. In particular, this model outperforms the results obtained with logistic regression.