Comparative Analysis of Time Series Forecasting Methods in Workforce Planning using Predictive Analytics

This study compares three approaches of labor force forecasting namely, ARIMA, Exponential smoothing, LSTM. In this regard, this is evidence that ARIMA has more accuracy and stability, having RMSE = 0.025 and MAE = 0.015. According to exponential smoothing results, there is an acceptable balance between precision and reliability which have RMSE value of 0.90 and MAE value of 0.18 severally. While this introduces some complexity in the model, the LSTM can readily identify complex patterns with just about 0.035 standard error to the mean and an average mistake size of just about 0.022. The strength of ARIMA and LSTM against economic indicators, as well as Exponential Smoothing’s sensitivity to industry trends. From the point of view of validation, the models were generally acceptable; ARIMA exhibited stability but Exponential Smoothing brought about flexibility. Through such an insight, organizations can use information for making wise decisions by choosing a certain type of strategy basing on specific priorities for the company, dynamics within this industry or field, as well as right balance that may be required while planning for workforce.