Development of a predictive model of financial performance indicators for organizations with integrated management systems, supported in artificial neural networks
Development of a predictive model of financial performance indicators for organizations with integrated management systems, supported in artificial neural networks
Purpose – Artificial neural networks (ANN) are efficient in dealing with complex problems, therefore, its use for the development of prediction methods of key performance indicators (KPI) can prove as a source of competitive advantage for companies. The objective of this work was to evaluate the use of ANN to perform financial KPI forecasts in companies that use Integrated Management Systems.
Design/methodology/approach – Using the Orbis Europe database, indicators were collected from companies that have certified Integrated Management Systems. Data of 50 Portuguese certified organizations was used.
The Palisade NeuralTools software was used to develop the Artificial Neural Networks (ANN). 3 Models were developed for the prediction of 10 indicators: Model A predicts the indicators of 10 companies (for 2018 and 2019) using their own results from 2010 to 2019; Models B predicts the indicators of two companies for 2019, using an average of the indicators from 2010 to 2019 data from various companies; and Model C predicts the indicators of two companies for 2019 using an average of the indicators from 2010 to 2018 data from various companies. Findings – The interpretation of ANN was done through graphics and relative error analysis. The results were based on data provided by the software and analyzed through graphs and tables for each of the 3 predictive models.
A comparison was made between forecasts and actual values, which allows us to conclude that the results using the ANN are satisfactory, although some indicators have performed better in one model than in another.
It was also possible to conclude that the use of the studied predictive models can bring competitive advantages through the improvement of processes, since the ANN also calculates which variables had the most impact on the prediction of each indicator, thus providing better precision in decision-making.
Originality/value – This study demonstrates positive results in the use of Artificial Neural Networks for the development of a forecast model for financial performance indicators, aimed at companies that use integrated management systems.
Quality Management, Integrated Management Systems, Artificial Intelligence
