International Journal of Scientific Research and Engineering Development

With the fast pace at which the market for computing devices, which includes laptops, desktop computers, smart phones, tablets, and embedded systems, is evolving, there has been a pressing need to provide accurate demand forecasts. In many cases, traditional statistical techniques, such as ARIMA and exponential smoothing, have proven insufficient in capturing the complex dynamics of the demand patterns in technology markets. In this paper, an AI/ML demand forecasting framework will be introduced for computing devices. Specifically, the proposed technique involves the application of machine learning algorithms including random forest, gradient boosting (such as XGBoost or LightGBM), LSTM, and hybrid ensembles to predict demand for computing devices. Heterogeneous data sets such as sales data, macroeconomic indicators, sentiment analysis of social media feeds, and supply chain data can be incorporated in this process. Experimental results show that this hybrid model outperforms conventional methods in terms of its ability to accurately predict demand (low Mean Absolute Percentage Error). The system can be used by suppliers/manufacturers, retailers, and supply chain managers to make decisions on procurement and production.

G.Anbu, R.Ramakrishnan,"Demand Prediction for Computing Devices Using ML-Based Models" International Journal of Scientific Research and Engineering Development, V9(3): Page(1747-1753) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.