This article analyzes the potential of predictive analysis as a tool to support the management of municipal services, using the Municipal Services of Abrantes as a case study. This study uses the application of three forecasting models (ARIMA, ETS and the Power BI forecast functionality) applied to five economic-financial indicators (general liquidity, reduced liquidity, financial autonomy, solvency and return on assets). The analysis is based on a time series of data collected between 2011 and 2023, focusing the predictive analysis on the years 2024 and 2025 (data for 2024 is not yet available and the year 2025 is still in progress). The performance of the models was evaluated using three error metrics: RMSE; MAE and MAPE, allowing to compare the accuracy of the models used. The results obtained demonstrate that predictive models offer a high degree of reliability and usefulness for financial planning, particularly in the context of municipal services. The predictions obtained point to a trend of stability in the analyzed indicators, with controlled levels of uncertainty, thus validating the application of these models in supporting decision-making. It is concluded that predictive analysis constitutes a significant added value for the modernization of the management of the entities under analysis, promoting more informed and rigorous decision-making.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Predictive Analysis and Decision Making: A New Perspective in the Management of Municipal Services

  • Bernardo Fernando,
  • Augusta Ferreira,
  • Carlos Santos

摘要

This article analyzes the potential of predictive analysis as a tool to support the management of municipal services, using the Municipal Services of Abrantes as a case study. This study uses the application of three forecasting models (ARIMA, ETS and the Power BI forecast functionality) applied to five economic-financial indicators (general liquidity, reduced liquidity, financial autonomy, solvency and return on assets). The analysis is based on a time series of data collected between 2011 and 2023, focusing the predictive analysis on the years 2024 and 2025 (data for 2024 is not yet available and the year 2025 is still in progress). The performance of the models was evaluated using three error metrics: RMSE; MAE and MAPE, allowing to compare the accuracy of the models used. The results obtained demonstrate that predictive models offer a high degree of reliability and usefulness for financial planning, particularly in the context of municipal services. The predictions obtained point to a trend of stability in the analyzed indicators, with controlled levels of uncertainty, thus validating the application of these models in supporting decision-making. It is concluded that predictive analysis constitutes a significant added value for the modernization of the management of the entities under analysis, promoting more informed and rigorous decision-making.